Green European 2016

Downloaded by [University of California, San Diego] at 18:58 11 February 2017
Empirical, and methodologically sophisticated, this excellent volume is the first

systematic comparative study of environmental attitudes using the ISSP in over
two decades. It explores with great insight environmental attitudes and behaviour
in Europe and elsewhere.
Jon Pammett, Professor of Political Science
at Carleton University, Canada
Firmly based on data from international social surveys, this volume provides a
varied and fascinating overview of European environmental perceptions, attitudes
and actions. In applying insightful interpretations on a rich collection of avail-
able data, it fills a major gap in our understanding of how Europeans think and
act about environment. Moreover, the contributions to the book represent a truly
European-wide set of analytical angles to the social dimensions of environment.
Highly recommended, both as an indispensable resource for further scholarly
research, and as a source of inspiration in grasping the complex dynamics of citi-
zen environmental concern.
Dr. C.S.A. (Kris) van Koppen, Associate Professor
Environmental Policy at Wageningen University and
former Chair of the Environment & Society Research
Network of the European Sociological Association
This Page is Intentionally Left Blank

Green European
Green European addresses the quest for a better understanding of European

type(s) of environmentalism. This monograph focuses on public attitudes and
behaviours and the culturally rooted, as well as country-specific, differences. The
book addresses the wider issue that many European countries are rendered ‘green’
or as having an advanced environmental awareness, but the question, ‘How green
are Green Europeans really?’ is yet to be answered.
The book covers a variety of unique data-driven comparative studies and is
divided into three parts:
•• The first addresses perceptions of environmental and technological threats

and risks.
•• The second part deals with environmental activism in Europe.
•• The third discusses environmental attitudes and environmental concerns and
their imminent link to personal pro-environmental behaviour.
The empirical comparative nature of the contributions is enabled by data from the
International Social Survey Programme (ISSP).
Audronė Telešienė is Associate Professor of Sociology at the Institute of Public

Policy and Administration, Kaunas University of Technology, Lithuania. She
is a member of the national team implementing the International Social Survey
Programme (ISSP) in Lithuania. Her research interests include environmen-
tal sociology, sociology of science and technology, risk communication, social
change and social analytics.
Matthias Gross is Professor of Environmental Sociology at the Helmholtz

Centre for Environmental Research, UFZ, in Leipzig, and the University of Jena,
Germany. His recent research focuses on public perceptions of renewable energy,
ignorance and security, real-world experiments, and the changing role of civil
society in environmental policy.
Routledge/European Sociological Association Studies in
European Societies
Series editors: Thomas P. Boje, Max Haller, Martin Kohli and Alison Woodward
13 Internationalisation of Social 18 Religion, Identity and

Sciences in Central and Politics
Eastern Europe Germany and Turkey in
The ‘catching up’ - a myth or a Interaction
strategy? Edited by Haldun Gülalp and
Edited by Ilona Pálné Kovács and Günter Seufert
Dagmar Kutsar
19 Post-identity?
14 Baltic Biographies at Culture and European
Historical Crossroads Integration
Edited by Li Bennich-Björkman and Edited by Richard McMahon
Aili Aarelaid-Tart
20 Artistic Practices
15 Mobilizing Regions, Social Interactions and Cultural
Mobilizing Europe Dynamics
Expert knowledge and scientific Edited by Tasos Zembylas
planning in European regional
development 21 Collective Memories in
Sebastian M. Büttner War
Edited by Krzysztof Kosela ,
16 The Road to Social Europe Elena Rozhdestvenskaya,
A contemporary approach to Victoria Semenova and Irina
political cultures and diversity Tartakovskaya
in the EU
Jean-Claude Barbier 22 Green European
Environmental behaviour and
17 The Social Meaning of attitudes in Europe in a historical
Children and Fertility and cross-cultural comparative
Change in Europe perspective
Edited by Anne Lise Ellingsæter, Edited by Audronė Telešienė and
An-Magritt Jensen and Merete Lie Matthias Gross
Green European
Environmental behaviour and

attitudes in Europe in a
historical and cross-cultural
comparative perspective
Edited by Audronė Telešienė

and Matthias Gross
First published 2017
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2017 Audronė Telešienė and Matthias Gross
The right of the editors to be identified as the authors of the

editorial material, and of the authors for their individual chapters,
has been asserted in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
Names: Telesiene, Audrone, editor. | Gross, Matthias, 1969- editor.
Title: Green European : environmental behaviour and attitudes
in Europe in a historical and cross-cultural comparative
perspective / edited by Audrone Telesiene and Matthias Gross.
Description: Abingdon, Oxon ; New York, NY : Routledge, 2016.
Identifiers: LCCN 2016003278| ISBN 9781138123953 (hardback) |
ISBN 9781315648491 (ebook)
Subjects: LCSH: Environmentalism—Social aspects—Europe. | Green
movement--Europe.
Classification: LCC GE199.E85 G74 2016 | DDC 304.2/8094—dc23
LC record available at http://lccn.loc.gov/2016003278
ISBN: 978-1-138-12395-3 (hbk)

ISBN: 978-1-315-64849-1 (ebk)
Typeset in Times New Roman

by Swales & Willis Ltd, Exeter, Devon, UK
Contents
List of figures ix
List of tables xii
Notes on contributors xiv
Foreword: a brief history of sociological research on
environmental concern xix
RILEY E. DUNLAP
List of common abbreviations xxvii

Acknowledgements xxviii
Introduction: how green are Green Europeans? 1

MATTHIAS GROSS AND AUDRONĖ TELEŠIENĖ
PART I
Reconsidering environmental threats and risks 11
1 The perception of environmental threats in a global and

European perspective 13
MARKUS HADLER AND KLAUS KRAEMER
2 Vulnerable and insecure? Environmental and technological

risk perception in Europe 31
AISTĖ BALŽEKIENĖ AND AUDRONĖ TELEŠIENĖ
3 Changing concern about environmental threats and risks:

a longitudinal and multilevel perspective on the
relationship between values and interests 56
ANDRÉ SCHAFFRIN AND ALEXANDER SCHMIDT-CATRAN
viii Contents
PART II
Personal activist profiles in Europe 89
4 Explaining environmental activism by national cultures:

the hypothesis of hysteresis 91
JEAN-PAUL BOZONNET
5 Explaining membership in pro-environmental groups:

a comparative European perspective 111
EGLĖ BUTKEVIČIENĖ AND VAIDAS MORKEVIČIUS
6 Petitioner, contributor, protester: the profile of Europeans

performing different ‘public-sphere’ pro-environmental
behaviours 133
IOSIF BOTETZAGIAS, NIKOLETA JONES AND
CHRYSOVALANTIS MALESIOS
PART III
Exploring European green mindsets and behaviours 155
7 How far for the money? Affluence and democratic

degrowth potential in Europe 157
MLADEN DOMAZET AND BRANKO ANČIĆ
8 Environmental concern in cross-national comparison:

methodological threats and measurement equivalence 182
JOCHEN MAYERL
9 Youth environmental consciousness in Europe:

the influence of psychosocial factors on
pro-environmental behaviour 205
LEENA HAANPÄÄ
10 Dilemmas of sustainability in turbulent times:

the Portuguese case in context 221
JOÃO GUERRA, LUÍSA SCHMIDT AND SUSANA VALENTE
Index 243
Figures
1.1 Dimensions 1 and 2 of environmental threat perception and

position of countries 21
1.2 Dimensions 1 and 3 of environmental threat perception and
position of countries 22
2.1 Perception of risk across European countries’ means of
risk index, where 1 is ‘not at all dangerous’ and 5 is
‘extremely dangerous’ (ISSP 2010) 40
2.2 Perceptions of different types of risk at individual level.
Mean scores (ISSP 2010) 41
2.3 Risk perception across socio-demographic characteristics.
Mean scores (ISSP 2010) 42
2.4 Impact of affluence on risk perception: x-axis: average
aggregate risk-perception index (2010, N = 24,927);
y-axis: GDP per capita (2010, US$) 49
2.5 Impact of human development level on risk perception: x-axis:
average aggregate risk-perception index (2010, N = 24,927);
y-axis: non-income Human Development Index 2010 49
2.6 Impact of democracy on risk perception: x-axis:
y-axis: Democracy Index 2010 50
2.7 Impact of pace of societal transformations on risk perception:
x-axis: average aggregate risk-perception index (2010, N = 24,927);
y-axis: average annual Human Development Index growth (%)
in 2000–2010 50
2.8 Impact of environmental health on risk perception: x-axis:
y-axis: Environmental Health Index 2010 51
3.1 Post-materialism and awareness, efficacy, and willingness-to-pay
by geographical regions 79
5.1 Membership in environmental groups 117
x List of figures
5.2 Thresholds of fuzzification for outcome variable: proportion

of population participating in groups to preserve or
protect environment 118
5.3 Impact of pro-environmental awareness and behaviours of
country populations on levels of their public involvement in
pro-environmental groups: results of fsQCA126
5.4 Impact of existing societal institutions on levels of their
public involvement in pro-environmental groups:

results of the fsQCA128
6.1 Comparison of differences between European and
rest-of-the-world countries concerning performance of three,
exclusive, public-sphere environmental behaviours. Results
of decision-tree analysis for 2000 and 2010 data 150
7.1 III 2011, Q11a (Active Degrowth: Environment: protect by
economic growth) indicators’ scatter of values for
individual European country samples 171
7.2 III 2011, Q10a (Active Degrowth: Worry too much about
environment over jobs) indicators’ scatter of values for
7.3 III 2011, Q19c (Passive Degrowth: Economic progress will
slow down without environmental protection) indicators’
scatter of values for individual European country samples 173
7.4 III 2011, Q11b (Passive Degrowth: Economic growth:
harms the environment) indicators’ scatter of values for
8.1 Measurement model of environmental concern (mean structure
not shown) 194
9.1 Model predicting pro-environmental behaviour
(cf. Oreg & Katz-Gerro 2006) 210
9.2 SEM results of path model. Values are reported as
standardized coefficients 215
10.1 Extent of self-assessed knowledge of causes of and solutions
to environmental problems in Portugal (ISSP 2010, N = 1,022) 224
10.2 Environmental knowledge index (causes and solutions)
within ISSP countries (ISSP 2010, N = 35,051) 225
10.3 Environmental knowledge by GNI per capita, HDI and
inequality within ISSP countries (ISSP 2010, N = 35,051;
UNDP, 2013) 226
10.4 General environmental concern by ISSP country and region
(ISSP 2010, N = 35,051) 228
10.5 Environmental knowledge and general environmental concern
(ISSP 2010, N = 35,051; UNDP, 2013) 229
10.6 To protect the environment, this country is doing too little by
ISSP country and region (ISSP 2010, N = 35,051) 230
List of figures xi
10.7 Importance of environment concern and other priorities

(ISSP 2010, N = 35,051) 232
10.8 Economic growth always harms the environment (ISSP 2010,
N = 35,051) 233
10.9 Economic growth harms the environment. Evolution between
2000 and 2010–12 (ISSP 2010, N = 35,051; ISSP 2000,
N = 28,389) 233
10.10 Country needs economic growth to protect environment

(ISSP 2010, N = 35,051) 234
10.11 To protect the environment: best approach for citizens and
for businesses (ISSP 2010, N = 35,051) 236
10.12 In order to protect the environment, government should . . .
(ISSP 2010, N = 35,051) 237
10.13 More information/education on environmental protection
versus coercive laws and GNI (ISSP 2010, N = 35,051) 238
Tables
1.1 Perception of environmental threats across countries (ISSP 2010) 19

1.2 Correlations among societal affluence, EU membership and threat
perceptions23
1.3 Individual-level determinants of environmental threat perceptions 24
2.1 List of macrolevel measures, indicators and data sources 39
2.2 Multiple linear regression analysis of structural factors as
determinants of risk perception 43
2.3 Multiple linear regression analysis of cognitive factors, trust and
environmental concern as determinants of risk perception 44
2.4 Multiple linear regression analysis for all predictors of risk
perception at individual level 46
2.5 Macrolevel data and risk-perception index values (mean score)
for European countries 47
2.6 Correlation matrix of dependent variable of risk-perception index
and independent context variables 48
3.1 Confirmatory factor analysis of awareness, efficacy and willingness 68
3.2 Number of respondents per country–year used in statistical
analysis (after list-wise deletion) 71
3.3 Multilevel models of awareness, efficacy and willingness with
breakdown of cross-sectional and longitudinal effects 74
4.1 Percentages of environmental activism (ISSP 2010) 95
4.2 ‘Environmental activism index: at least 1 action’ according to
socio-demographic variables (ISSP 2010) 97
4.3 Percentages and binary logistic regression (exp B): ‘at least one
environmental action’ vs. ‘no action’ 103
5.1 Pro-environmental awareness and behaviour of populations and
membership in groups to preserve or protect environment:
raw and fuzzified data 119
5.2 Existing societal institutions in country and membership in groups
to preserve or protect environment: raw and fuzzified data 121
5.3 Results of the analysis of necessary conditions for positive
outcomes (high country levels of membership in groups to
List of tables xiii
preserve or protect environment): impact of pro-environmental

awareness and behaviours of population 123
5.4 Pro-environmental awareness and behaviour of populations and
membership in groups to preserve or protect environment:
complete truth table for positive outcomes (inclusion cut-off = 0.9) 124
5.5 Results of the analysis of necessary conditions for positive
outcomes (high country levels of membership in groups to
preserve or protect environment): impact of societal institutions 127

5.6 Existing societal institutions and membership in groups to
preserve or protect environment: complete truth table for
positive outcomes (inclusion cut-off = 0.91) 127
6.1 European citizens’ exclusive engagement in one out of three
public-sphere pro-environmental behaviours over the past 5 years
(ISSP 2000; ISSP 2010) 140
6.2 Discriminant characteristics of European citizens who have
engaged exclusively in one out of three public-sphere
environmental behaviours over the period 1996–2000 144
6.3 Discriminant characteristics of European citizens who have
engaged exclusively in one out of three public-sphere
environmental behaviours over the period 2006–2010 146
6.4 Comparison of 2000 and 2010 profiles of Europeans who
have ‘donated money to an environmental group’ vs. ‘signed a
petition’ vs. ‘demonstrated/protested for the environment’
over the past 5 years 147
7.1 Instruments: Attitudes of Concern and Activation (ACA) index,
Attitudes of Material Sacrifice (AMS) index, active degrowth
indicators, passive degrowth indicators 164
7.2 Mean values of indices and indicators for European and RoW
groups of countries participating in ISSP 2010 Environment module 169
7.3 Mean values of indices and indicators for European subsamples 170
7.4 Correlations between national individual concern (ACA) and
sacrifice (AMS) indices and active and passive degrowth indicators
(ISSP 2010: Q10a, Q11a, Q11b, Q19c) 174
8.1 Concepts of factorial invariance (FIV) in cross-national studies 188
8.2 ISSP 2010 indicators of environmental concern used in this study 191
8.3 Evaluation of measurement model across Europe 196
8.4 Evaluation of measurement model across the world 197
8.5 Comparison of mean environmental concern across the world:
latent and manifest approaches (reference group is Austria) 198
9.1 Number of respondents, means and standard deviations of variables
used in SEM analysis 214
Contributors
Branko Ančić holds bachelor and doctoral degrees from the University of Zagreb.
He is a post-doctoral fellow at the Institute for Social Research in Zagreb,
Centre for Research in Social Inequalities and Sustainability. He has been a
member of the national team implementing the International Social Survey
Programme (ISSP) in Croatia. His scientific and research interests include
sociology of religion, sociology of health and sociology of sustainable devel-
opment. His publications include an analysis of atheism and non-religiosity
in Europe (Oxford Handbook of Atheism, Oxford University Press, 2014), an
exploration of connection between religion and health (Sociologija i prostor,
2011), religion in south-east Europe (in Hannapi, H. South-East Europe in
Evolution, Routledge, 2014) and interpretation of degrowth attitudes in Europe
(Teorija in praksa, 2015).
Aistė Balžekienė is Associate Professor and Sociology Research Group
Coordinator at the Institute of Public Policy and Administration, Kaunas
University of Technology, Lithuania. Her research interests include compara-
tive research on risk perception and risk discourse, social aspects of technolog-
ical and environmental risks, the social impact of technologies, environmental
sociology and social research methodology. She is involved in various scien-
tific projects researching risk and security governance, environmental compen-
sation. She is a member of the national team implementing International Social
Survey Programme (ISSP) in Lithuania.
Iosif Botetzagias is an Assistant Professor of Environmental Politics and Policy
at the Department of the Environment, University of the Aegean, Greece.
His research interests include pro-environmental behaviours, environmental
non-governmental organizations, Green parties and environmental policy
and management, on which he has published several book chapters and arti-
cles in scientific journals such as Environmental Politics, Environment &
Behaviour, Global Environmental Politics and the Journal of Environmental
Policy & Management.
Jean-Paul Bozonnet is Associate Researcher in Sociology at PACTE-CNRS
(Politiques publiques, ACtion politique, TErritoires; National Center for
Notes on contributors xv
Scientific Research) – Sciences Po, Grenoble University, France. His main

research interests are related to international comparative perspectives on
environmental attitudes and behaviours, and to the topic of an ecological
grand narrative (how it shapes environmental discourses and public policies).
His recent work has been published as Les valeurs des Européens, Évolution
et clivages (2014), Une jeunesse differente? Les valeurs des jeunes Français
depuis 30 ans (2012), Manuel de sociologie de l’environnement (2012) and
Controverses climatiques, sciences et politique (2012), and also published as

articles in the following journals: French Politics (2013), Revue politique et
parlementaire (2012),
Eglė Butkevičienė is Associate Professor in Sociology and Public Policy at the
Institute of Public Policy and Administration, and the Dean of the Faculty of
Social Sciences, Arts and Humanities at Kaunas University of Technology.
Her research interests include civil society, community development, non-
governmental organizations, social movements, social innovations, social
impact of technologies, virtual communities and social networking. She has par-
ticipated in a number of international projects, e.g. 6th Framework Programme
project ‘iCamp’, CINEFOGO (Civil Society and New Forms of Governance
in Europe), and ICT for All. She is a member of national teams implementing
the International Social Survey Programme (ISSP) and the European Social
Survey (ESS) in Lithuania.
Mladen Domazet graduated in Physics and Philosophy from the University
of Oxford, and completed a doctorate in Philosophy of Science at the
University of Zagreb. He is the Research Director at the Institute for
Political Ecology in Zagreb, Croatia. His publications reflect a diversity of
interdisciplinary interests and collaborations, from analysis of Wikipedia
as a complex network to recent books Alice Returns from Wonderland:
Ontological Frameworks for Explanation from Contemporary Quantum
Theories (KruZak, 2012) and Sustainability Perspectives from the European
Semi-periphery (Institut za društvena istraživanja; Heinrich Böll Stiftung
Hrvatska, 2014).
Riley E. Dunlap is Dresser Professor and Regents Professor of Sociology at
Oklahoma State University. A long-term specialist in environmental sociol-
ogy, his current work focuses primarily on political polarization over climate
change and climate-change denial. He chaired the American Sociological
Association’s Task Force on Sociology and Global Climate Change, and
is Senior Editor of the resulting volume, Climate Change and Society:
Sociological Perspectives (Oxford University Press, 2015).
Matthias Gross is Full Professor of Environmental Sociology at the Helmholtz
Centre for Environmental Research, UFZ, Leipzig, and, by joint appointment,
at the University of Jena, Germany. His recent research focuses on the evolu-
tion of alternative energy systems, the centrality of ignorance in engineering,
xvi Notes on contributors
and experimental practices in society. Together with Linsey McGoey he has

recently co-edited the Routledge International Handbook of Ignorance Studies
(2015). He is author of several books including Renewable Energies (together
with Rüdiger Mautz, Routledge, 2015) and Ignorance and Surprise: Science,
Society, and Ecological Design (MIT Press, 2010).
João Guerra is currently a post-doctoral fellow of the Portuguese Foundation
for Science and Technology (FCT), carrying out research on participation

and environmental justice at ICS/ULisboa, the Social Sciences Institute of
University of Lisbon. He holds a Ph.D. (2011) and a Master’s degree (2005)
in Social Sciences from ICS/ULisboa, and a degree in Sociology and Planning
from ISCTE/IUL, University Institute of Lisbon (1998). He is a member of
the Executive Committee of OBSERVA (Observatory of Environment and
Society) and he is also an ECO XXI national commission member; a pro-
ject which assesses annually the characteristics and practices of sustainabil-
ity among Portuguese municipalities. Since 1998 he has carried out several
research projects focused on environmental issues and their interactions with
society, governance, civic participation, sustainable development, environmen-
tal education, environmental justice, etc. He is also the author and co-author
of many articles and books on issues of sustainability, environment justice and
civic participation.
Leena Haanpää is a senior researcher who works at the Child and Youth
Research Institute (CYRI) at the University of Turku. She graduated first in
Italian language and culture, and later in marketing. Her thesis (2007) dealt
with environmental sociology and the sociology of consumption. Currently
her research interests are children and youth: their active citizenship, partici
pation, environmental consciousness and consumerism.
Markus Hadler is a senior lecturer at the Department of Sociology, Macquarie
University, Sydney, Australia. His research focuses on questions of politi-
cal sociology, social inequality and environmental sociology. As one of the
Austrian representatives to the ISSP, he was in charge of the implementation
of the ISSP Environment Survey in Austria, which was funded by the Austrian
Science Fund (Project #P22575-G17).
Nikoleta Jones is a senior lecturer in Sustainability at the Global Sustainability
Institute, Anglia Ruskin University, UK.
Klaus Kraemer is Full Professor of Applied Sociology: Economy, Organization,
Social Problems, at the Department of Sociology, University of Graz, Austria.
His main areas of research are economic sociology and environmental sociology.
Chrysovalantis Malesios is a mathematical statistician, and is currently occu-
pied as a teaching fellow at the Department of Rural Development, Democritus
University of Thrace, Greece. He has worked as a statistician at the Institute of
Notes on contributors xvii
Statistical Research Analysis and Documentation (ISRAD), Athens University

of Economics and Statistics, Greece. He has published papers in internation-
ally refereed journals such as: Renewable and Sustainable Energy Reviews,
Scientometrics, Journal of Socio-Economics, European Societies and South
European Society and Politics.
Jochen Mayerl is Junior Professor for Methods of Empirical Social Research at
the Department of Social Sciences, University of Kaiserslautern, Germany. He

specializes in the study of attitude-behaviour relations, environmental concern,
social context analysis and survey methodology.
Vaidas Morkevičius is Senior Researcher at the Institute of Public Policy
and Administration, Kaunas University of Technology, Lithuania, and is
the national representative at the European Social Survey (ESS). He was
one of the founders of the Lithuanian Data Archive for Social Sciences
and Humanities (www.lidata.eu), and has been member of the national
team implementing the International Social Survey Programme (ISSP) in
Lithuania. His research interests include political communication, elite and
parliamentary studies, e-democracy, survey research, applied statistics and
qualitative comparative analysis.
André Schaffrin is a researcher at the EA European Academy of Technology
and Innovation Assessment, Germany. He is Project Coordinator for ‘Land
Use and Energy Supply at the Municipal Level’ (EnAHRgie) and a member of
the Managing Committee of the European COST (Cooperation in Science and
Technology) Action: Innovations in Climate Governance: Sources, Patterns
and Effects (INOGOV), at the EA European Academy of Technology and
Innovation Assessment.
Luísa Schmidt is a principal researcher at the Institute of Social Sciences,
University of Lisbon (ICS-ULisboa). At ICS-UL she is part of the Scientific
Council and leads the OBSERVA (Observatory of Environment, Territory
and Society). She runs and coordinates several projects in this field of work,
e.g. nationwide surveys on: environmental values and policy issues; local
administration, sustainable development and environmental/energy public
policies; environmental education and energy social and climate-change
issues. This scientific research has resulted in several books and peer-
reviewed papers over the past few years. Luísa is part of the scientific com-
mittee of the interdisciplinary Ph.D. on ‘Climate Change and Sustainable
Development Policies’; a partnership involving two Universities of Lisbon
(ULisboa and UNL). She is a member of the National Council for the
Environment and Sustainable Development, and the Working Group on
Sustainable Development, at the European Environment Advisory Council.
She is also a member of the national team implementing the International
Social Survey Programme (ISSP) in Portugal.
xviii Notes on contributors
Alexander Schmidt-Catran is a research assistant at the Institute of Sociology

and Social Psychology, at the Chair for Empirical Social and Economic
Research, University of Cologne, Germany. His research interests cover wel-
fare attitudes, social inequality, income poverty, migrants and attitudes towards
migrants, as well as statistical methodology.
Audronė Telešienė is Associate Professor in Sociology and Acting Head of
the Institute of Public Policy and Administration, Kaunas University of

Technology, Lithuania. At her university she also serves as a member of the
University Study Programme Committee. Her research interests include envi-
ronmental sociology, sociology of science and technology, risk communica-
tion, social analytics, subjective well-being and social change. She is a member
of the national team implementing the International Social Survey Programme
(ISSP) in Lithuania, and also has been a member of the national team imple-
menting the European Social Survey (ESS). She has published numerous arti-
cles in the field of environmental sociology and has co-authored Manuel de
sociologie de l’environnement (Presses de l’Université Laval, 2012).
Susana Valente is a research fellow at the Foundation of the Science Faculty,
University of Lisbon (FFCUL), taking part in an interdisciplinary team on
Climate Change (ClimAdPTLocal Program). She holds a Ph.D. (2013) in
Social Sciences from the Social Sciences Institute, University of Lisbon (ICS/
ULisboa), a Master’s degree (2001) and a degree in Sociology from ISCTE/
IUL (University Institute of Lisbon; 1994). She has been a founding member
of OBSERVA (Observatory of Environment and Society) since 1996, and has
carried out several research projects focused on environmental and sustainable
development issues from a sociological perspective, studying social percep-
tions, attitudes and practices among Portuguese people. More recently, she has
been working on urban waste, with a special focus on social participation and
public waste policy. She works frequently in community-based research pro-
jects and on stakeholder engagement.
Foreword
A brief history of sociological research
on environmental concern
Riley E. Dunlap
In the USA environmental issues began to receive increasing attention in the

1960s, and then moved prominently onto the public agenda with the massive
celebration of the first ‘Earth Day’ on 22 April, 1970 – the largest demonstration
in the nation’s history. Not surprisingly, one of the first reactions of sociolo-
gists to the emergence of environmental quality as a social problem and major
societal issue was to examine public opinion toward environmental issues.
Surveys of the general public (Murch 1971), as well as of selected populations
such as college students (Dunlap and Gale 1972), were quickly conducted and
published, as were analyses of data collected by major polling firms (McEvoy
1972). These studies typically employed single-item measures of ‘opinions’ on
issues like the seriousness of pollution, or sometimes sets of three or four items
dealing with the perceived causes of pollution, possible solutions, and so forth
(Dunlap and Gale 1972).
One of the first priorities of sociologists was to examine the social and politi-
cal characteristics associated with pro-environmental opinions, in an effort to
clarify the socio-political sources of public concern for environmental quality
(Buttel and Flinn 1974). By the end of the 1970s there was a sufficient number
of such studies to establish empirical generalizations about the ‘social bases’ of
pro-environmental opinions in the USA. Education (positive), age (negative) and
political ideology (liberalism) were found to be the best predictors, followed by
residence (urban) and political party identification (Democratic), and these pre-
dictors remained stable for at least another decade (Van Liere and Dunlap 1980;
Jones and Dunlap 1992). There is some evidence that, more recently, the key
predictors have changed in the USA, with gender (female) and race (non-white)
often being related significantly to pro-environmental views (Liu, Vedlitz and
Shi 2014). Of special note, at least in the USA, is that since the 1990s political
ideology and party identification have become the strongest predictors of envi-
ronmental attitudes in general (McCright, Xiao and Dunlap 2014) and of views
of climate change in particular (McCright and Dunlap 2011). Furthermore, there
is evidence that political polarization over climate change is extending to several
other nations, especially Australia, Canada and the UK (McCright, Dunlap and
Marquart-Pyatt 2016, and references therein).
xx Riley E. Dunlap
Studies of public opinion toward environmental issues have continued to be

a staple of environmental sociology, and of course have now been employed in
numerous nations around the world. Longitudinal studies tracking trends in pub-
lic opinion, typically using data from polling firms or established surveys such
as the General Social Survey in the USA, became popular (Dunlap and Scarce
1991), and more recently have become increasingly sophisticated methodologi-
cally (Melis, Elliot and Shryane 2014; Pampel and Hunter 2012) and extended to
the cross-national level (Nawrotzki and Pampel 2014). As anthropogenic climate

change has emerged as the central environmental problem of our time, efforts
have been made to track trends in public concern about the threat it poses, and
factors influencing changes over time (Brulle, Carmichael and Jenkins 2012).
While the types of studies discussed thus far represent major tendencies in
sociological work, they typically represented analyses of ‘public opinion’ and,
as noted above, often employed single-item indicators or other simple measures
of pro-environmental opinions. But two other early contributions of sociologists
were: (1) clarifying the concept of ‘environmental attitudes’ or ‘environmental
concern’, which Dutch sociologist Peter Ester (1981) noted were commonly
equated (and which others, such as German sociologist Brand [1997], preferred to
label ‘environmental consciousness’), and developing more sophisticated, multi-
item measures of it; and (2) employing and developing social-psychological
models to predict such concern and pro-environmental behaviours (Ester and Van
der Meer 1982). I discuss these two efforts in order.
While its use rose rapidly throughout the 1970s, environmental attitudes/
concern was an ambiguous concept, leading to widely diverse operationalizations
as Heberlein (1981) noted. Over time great effort has gone into clarifying both the
nature of the concept of environmental concern and its operationalization (Dunlap
and Jones 2002; Heberlein, 2012; Schaffrin 2011). Early attempts to develop
more sophisticated operationalizations of environmental concern employing
multi-item measures (Van Liere and Dunlap 1981) have been greatly extended
via the use of confirmatory factor analysis (CFA) and structural equation mod-
elling (Xiao and Dunlap 2007; Marquart-Pyatt 2015). This work suggests that
environmental concern, while clearly multifaceted, is a meaningful concept that
can be measured in a range of nations (Xiao, Dunlap and Hong 2013).
Over time efforts to explain variation in environmental concern became more
sophisticated, incorporating a variety of cognitive variables along with demo-
graphic variables (Dietz, Stern and Guagnano 1998). But much of the interest
in understanding environmental concern stemmed from the assumption that it
would be related to environmental behaviours, and Heberlein (1972) deserves
credit as the first sociologist (at least in the USA) to apply an established social-
psychological theory – the ‘norm-activation model’ (NAM) – in an effort to
explain pro-environmental behaviours. While the NAM stimulated several
studies and has been extended in new directions (Vandenbergh 2005), it also
provided a foundation for two major advances in understanding the sources of
environmental behaviours.
Foreword: A brief history of research xxi
First, a debate over whether or not Heberlein’s results suggested the emer-
gence of a ‘land ethic’, as he claimed (Dunlap and Van Liere 1977; Heberlein
1977), clarified the distinction between human and environmental conse-
quences of environmental behaviours – a distinction credited with stimulating
the search for ‘biospheric’ along with more traditional ‘egoistic’ and ‘altruis-
tic’ values (Stern, Dietz and Kalof 1993; Stern and Dietz 1994). This tripartite
distinction has now become well-established and widely employed in efforts
to explain environmental behaviours (Groot and Steg 2008). Second, a series

of extensions of Heberlein’s basic model were made by Stern, Dietz and col-
leagues (Stern, Dietz and Black 1985–86; Stern, Dietz and Guagnano 1995) that
gradually evolved into the ‘Value-Belief-Norm’ (VBN) theory of environmen-
tal activism (Stern, Dietz, Abel, Guagnano and Kalof 1999). VBN incorporates
measures of the three types of values, beliefs (frequently measured with the
New Ecological Paradigm Scale [Dunlap 2008]) and norms (as originally sug-
gested by Herblein), and has become the most widely used theoretical model
for explaining pro-environmental behaviours – albeit more often by psycholo-
gists than sociologists at this point in time (e.g., Steg, Dreijerink and Abrahamse
2005). For a somewhat similar model, see Best and Mayerl (2013).
Of course, a variety of other social-psychological perspectives, including
rational choice theory and the theory of planned behaviour (Diekmann and Schmidt
1998; Oreg and Katz-Gerro 2006), are also used by sociologists in analysing
individual-level environmental behaviours. Furthermore, in Europe – especially
in the 1980s and 1990s – a great deal of attention was focused on explaining
political behaviours, such as support for Green parties and involvement in anti-
nuclear movements (Rootes 1997), while more recently the emphasis on analyses
of individual-level environmental behaviours has been supplemented by analyses
of environmental practices (Spaargaren 2011).
In the 1990s, when sociologists were making important contributions to under-
standing environmental behaviours, a final critical development – and the one
most directly relevant to this volume – in sociological work on environmental
concern also took place: the emergence of cross-national studies of citizen concern
for the environment. In rapid succession there was Gallup’s 1992 Health of the
Planet Survey (HOP), the 1993 International Social Survey (ISSP) with an envi-
ronmental module, and the 1990–93 wave of the World Values Survey (WVS)
which included several environmental items, together providing a wealth of data
for cross-national comparisons of environmental concern. On the downside, the
indicators of environmental concern in these surveys were not ideal, as they were
designed more as measures of public opinion than of environmental attitudes.
Yet, this new wealth of multinational datasets stimulated a good deal of
sociological interest in studies of environmental concern at the cross-national
(and near international) level. One of the first issues addressed was the impact
of national affluence and post-material (PM) values on citizen concern for the
environment, as earlier work in Europe had suggested that environmentalism
largely arose from post-World War II affluence and the post-material values it
xxii Riley E. Dunlap
generated (Lowe and Rüdig 1986). However, results of the HOP directly chal-
lenged the PM values thesis, as on many dimensions of environmental concern
citizens of poor nations were found to be more concerned than their wealthy
counterparts (Dunlap, Gallup and Gallup 1993; Brechin and Kempton 1994;
Dunlap and Mertig 1995), leading to a major debate (with six contributions)
over the role of PM values in generating international environmental concern in
a March 1977 ‘Forum’ in Social Science Quarterly. Related debates continued
between proponents of various versions of an ‘affluence hypothesis’, that sees

environmental concern as generated by both individual and national prosper-
ity (Franzen 2003), and a ‘globalization hypothesis’ that sees environmental
concern as stimulated by awareness of and exposure to serious ecological deg-
radation (Brechin 1999; Dunlap and York 2008).
In recent years, as part of what might be termed a ‘quantitative revolution’
in environmental sociology, more sophisticated analyses employing multilevel
modelling have allowed researchers to investigate the combined effects of per-
sonal- and national-level characteristics on environmental concern. However,
the results are quite mixed in terms of the role of affluence, with some studies
(Franzen and Meyer 2010) finding both personal and national affluence leading
to higher levels of concern, and other studies finding conflicting results, typi-
cally at the national level but sometimes the personal as well (Fairbrother 2012;
Lo 2014). Individual-level PM values, however, seem to fare better (Hadler and
Haller 2011; Fairbrother 2012; Nawrotzki and Pampel 2014). These and other
studies (Knight and Messer 2012) make clear that the relationships between
affluence (and PM values) and environmental concern may vary, depending on
how concern is measured, with willingness-to-pay items often producing dif-
ferent results from those measuring the perceived seriousness of environmental
problems or the risks they pose. The fact that many of the environmental items
in the WVS and ISSP surveys have been used in subsequent waves is begin-
ning to allow for valuable longitudinal analyses that add additional insight into
the effects of economic conditions on environmental concern (Fairbrother 2012;
Knight and Messer 2012), and future longitudinal studies will hopefully bring
greater clarity to the issue.
Multilevel modelling has enabled researchers to go well beyond examining the
impacts of economic conditions by incorporating a variety of national-level vari-
ables such as urbanization, population density, democratization, inequality and
position in the world polity (Marquart-Pyatt 2012), allowing for tests of a wide
range of macrolevel theoretical perspectives (Hadler and Haller 2011; Knight and
Messer 2012; Givens and Jorgenson 2013; Jorgenson and Givens 2014) other than
post-material values. And, crucially, it has also allowed researchers to examine
the potential effects of indicators of nations’ environmental conditions (pollution
levels, eco-footprints, etc.), thus turning studies of environmental concern into
analyses of societal-environmental interactions (long deemed the core of envi-
ronmental sociology). Unfortunately, the environmental indicators are often far
from ideal, and the national level of aggregation likely attenuates their effects on
Foreword: A brief history of research xxiii
individuals’ levels of environmental concern, given the great variation in envi-

ronmental degradation within larger nations. Perhaps for this reason the impact
of environmental conditions on levels of environmental concern (at both the indi-
vidual and national levels) is, thus far, somewhat ambiguous (Fairbrother 2012;
Knight and Messer 2012; Marquart-Pyatt 2012; Jorgenson and Givens 2014).
Nonetheless, we can expect major progress in achieving clarity with the continu-
ing emergence of new data on nations’ environmental conditions.
In sum, sociological work on environmental concern has come a long way

since 1970. Surveys at the local and regional levels within nations (Hamilton and
Saito 2015), as well as national-level surveys (Best and Mayerl 2013), are still
widely used, but there is obviously a rapidly growing interest in comparative,
cross-national studies. This corresponds to the trend toward increasing use of
cross-national studies of all types within environmental sociology.
This trend is well represented by the current volume, which focuses primar-
ily on comparing various aspects and dynamics of environmental concern across
Europe and sometimes other regions as well (most often with ISSP data). The
chapters in this volume reflect and often extend the themes in cross-national
research just noted, as well as others deriving from earlier emphases in environ-
mental-concern research. They show, for example, a sensitivity to the importance
of analysing different dimensions of environmental concern (including percep-
tions of environmental risks) and creating measures with appropriate techniques
such as CFA, and even examining whether the measures are equivalent across
nations; a special interest in understanding the sources of environmental behav-
iours and environmentalism; the use of multilevel modelling to investigate the
effects of a wide range of individual- and national-level factors (including envi-
ronmental conditions); the use of longitudinal analyses; and the application of
a wide range of appropriate theoretical perspectives from well-established ones,
such as Inglehart’s post-materialism theory and Beck’s risk society theory, to
innovative notions like ‘Hysteresis’. In combination, the chapters not only illus-
trate the value of comparative studies of environmental concern but also extend
our knowledge in important ways – primarily, but not exclusively, to Europe.
Anyone interested in understanding environmental attitudes and behaviours, espe-
cially in Europe, will benefit from reading this volume.
References
Best, Henning and Jochen Mayerl. 2013. ‘Values, Beliefs, Attitudes: An Empirical Study
on the Structure of Environmental Concern and Recycling Participation.’ Social Science
Quarterly 94:691–714.
Brand, Karl-Werner. 1997. ‘Environmental Consciousness and Behaviour: The Greening
of Lifestyles.’ In M. Redclift and G. Woodgate (eds.), The International Handbook of
Environmental Sociology. pp. 204–217. Cheltenham, UK: Edgar Elgar.
Brechin, Steven R. 1999. ‘Objective Problems, Subjective Values, and Global Environmen
talism: Evaluating the Postmaterialist Argument and Challenging a New Explanation.’
Social Science Quarterly 80:793–809.
xxiv Riley E. Dunlap
Brechin, Steven R. and Willet Kempton. 1994. ‘Global Environmentalism: A Challenge to

the Postmaterialism Thesis?’ Social Science Quarterly 75:245–69.
Brulle, Robert J., Jason Carmichael and J. Craig Jenkins. 2012. ‘Shifting Public Opinion
on Climate Change: An Empirical Assessment of Factors Influencing Concern Over
Climate Change in the U.S., 2002–2010.’ Climatic Change 114:169–188.
Buttel, Frederick H. and William L. Flinn. 1974. ‘The Structure of Support for the
Environmental Movement, 1968–1970.’ Rural Sociology 39:56–69.
deGroot, Judith I. M. and Linda Steg. 2008. ‘Value Orientations to Explain Beliefs Related
to Environmental Significant Behavior: How to Measure Egoistic, Altruistic, and
Biospheric Value Orientations.’ Environment and Behavior 40:330–354.
Diekmann, A. and P. Schmidt. 1998. ‘Special Section: Testing Rational Choice Models of
Behavior Involving Environmental Issues.’ Rationality and Society 10:77–78.
Dietz, Thomas, Paul C. Stern and Gregory A. Guagnano. 1998. ‘Social Structural and
Social Psychological Bases of Environmental Concern.’ Environment and Behavior
30:450–471.
Dunlap, Riley E. 2008. ‘The NEP Scale: From Marginality to Worldwide Use.’ Journal of
Environmental Education 40:3–18.
Dunlap, Riley E. and Richard P. Gale. 1972. ‘Politics and Ecology: A Political Profile of
Student Eco-Activists.’ Youth and Society 3:379–397.
Dunlap, Riley E., George H. Gallup, Jr. and Alec M. Gallup. 1993. ‘Of Global Concern:
Results of the Health of the Planet Survey.’ Environment 35 (November):7–15, 33–39.
Dunlap, Riley E. and Robert Emmet Jones. 2002. ‘Environmental Concern: Conceptual
and Measurement Issues.’ In R.E. Dunlap and W. Michelson (eds.), Handbook of
Environmental Sociology. pp. 482–524. Westport, CT: Greenwood Press.
Dunlap, Riley E. and Angela G. Mertig. 1995. ‘Global Concern for the Environment: Is
Affluence a Prerequisite?’ Journal of Social Issues 51:121–137.
Dunlap, Riley E. and Rik Scarce. 1991. ‘The Polls – Poll Trends: Environmental Problems
and Protection.’ Public Opinion Quarterly 55:713–734.
Dunlap, Riley E. and Kent D. Van Liere. 1977. ‘Land Ethic or Golden Rule.’ Journal of
Social Issues 33:200–207.
Dunlap, Riley E. and Richard York. 2008. ‘The Globalization of Environmental Concern
and the Limits of the Post-Materialist Explanation: Evidence from Four Cross-National
Surveys.’ Sociological Quarterly 49:529–563.
Ester, Peter. 1981. ‘Environmental Concern in the Netherlands.’ In T. O’Riordan and
R. K. Turner, eds. Progress in Resource Management and Environmental Planning,
Vol. 3. pp. 81–108. Chichester: John Wiley & Sons.
Ester, Peter and F. van der Meer. 1982. ‘Determinants of Individual Environmental
Behaviour: An Outline of a Behavioural Model and Some Research Findings.’ The
Netherlands’ Journal of Sociology 18:57–94.
Fairbrother, Malcolm. 2012. ‘Rich People, Poor People, and Environmental Concern:
Evidence across Nations and Time.’ European Sociological Review 29:910–922.
Franzen, Axel. 2003. ‘Environmental Attitudes in International Comparison: An Analysis
of the ISSP Surveys 1993 and 2000.’ Social Science Quarterly 84:297–308.
Franzen, Axel and Retro Meyer. 2010. ‘Environmental Attitudes in Cross-National
Perspective: A Multilevel Analysis of the ISSP 1993 and 2000.’ European Sociological
Review 26:219–234.
Givens, Jennifer E. and Andrew K. Jorgenson. 2013. ‘Individual Environmental Concern
in the World Polity: A Multilevel Analysis.’ Social Science Research 42:418–431.
Foreword: A brief history of research xxv
Hadler, Markus and Max Haller. 2011. ‘Global Activism and Nationally Driven Recycling:
The Influence of World Society and National Contexts on Public and Private
Environmental Behavior.’ International Sociology 26:315–345.
Hamilton, Lawrence C. and Kei Saito. 2015. ‘A Four-Party View of US Environmental
Concern.’ Environmental Politics 24:212–227.
Heberlein, Thomas A. 1972. ‘The Land Ethic Realized: Some Social Psychological
Explanations for Changing Environmental Attitudes.’ Journal of Social Issues 28:79–87.
Heberlein, Thomas A. 1977. ‘Norm Activation and Environmental Action: A Rejoinder to

R. E. Dunlap and K. D. Van Liere.’ Journal of Social Issues 33:207–211.
Heberlein, Thomas A. 1981. ‘Environmental Attitudes.’ Zeitschrift fur Umweltpolitik
2:241–270.
Heberlein, Thomas A. 2012. Navigating Environmental Attitudes. Oxford and New York:
Oxford University Press.
Jones, Robert Emmet and Riley E. Dunlap. 1992. ‘The Social Bases of Environmental
Concern: Have They Changed Over Time?’ Rural Sociology 57:28–47.
Jorgenson, Andrew K. and Jennifer E. Givens. 2014. ‘Economic Globalization and
Environmental Concern: A Multilevel Analysis of Individuals Within 37 Nations.’
Environment and Behavior 46:848–871.
Knight, Kyle W. and Benjamin L. Messer. 2012. ‘Environmental Concern in Cross-
National Perspective: The Effects of Affluence, Environmental Degradation, and World
Society.’ Social Science Quarterly 93:521–537.
Liu, Xinsheng, Arnold Vedlitz and Liu Shi. 2014. ‘Examining the Determinants of Public
Environmental Concern: Evidence from National Public Surveys.’ Environmental
Science & Policy 39:77–94.
Lo, Alex Y. 2014. ‘Negative Income Effect on Perception of Long-Term Environmental
Risk.’ Ecological Economics 107:51–58.
Lowe, Philip D. and Wolfgang Rüdig. 1986. ‘Review Article: Political Ecology and the
Social Sciences – The State of the Art.’ British Journal of Political Science 16:513–550.
Marquart-Pyatt, Sandra T. 2012. ‘Contextual Influences on Environmental Concerns
Cross-Nationally: A Multilevel Investigation.’ Social Science Research 41:1085–1099.
Marquart-Pyatt, Sandra T. 2015. ‘Public Opinion about the Environment: Testing
Measurement Equivalence across Countries.’ International Journal of Sociology
45:309–326.
McCright, Aaron M. and Riley E. Dunlap. 2011. ‘The Politicization of Climate Change:
Political Polarization in the American Public’s Views of Global Warming.’ Sociological
McCright, Aaron M., Riley E. Dunlap, and Sandra Marquart-Payatt. 2016. ‘Political
Ideology and Views about Climate Change in the European Union.’ Environmental
Politics 25:338–358.
McCright, Aaron M., Chenyang Xiao and Riley E. Dunlap. 2014. ‘Political Polarization
on Support for Government Spending on Environmental Protection in the USA,
1974–2012.’ Social Science Research 48:251–260.
McEvoy, James III. 1972. ‘The American Concern with the Environment.’ In W. R. Burch,
Jr., N. H. Cheek, Jr. and L. Taylor (eds.), Social Behavior, Natural Resources, and the
Environment. pp. 214–236. New York: Harper & Row.
Melis, Gabriella, Mark Elliot and Nick Shryane. 2014. ‘Environmental Concern Over
Time: Evidence from the Longitudinal Analysis of a British Cohort Study from 1991 to
2008.’ Social Science Quarterly 95:905–919.
xxvi Riley E. Dunlap
Murch, Arvin W. 1971. ‘Public Concern for Environmental Pollution.’ Public Opinion
Nawrotzki, Raphael J. and Fred C. Pampel. 2014. ‘Cohort Change and the Diffusion of
Environmental Concern: A Cross-National Analysis.’ Population & Environment
35:1–25.
Oreg, Shaul and Tally Katz-Gerro. 2006. ‘Predicting Proenvironmental Behavior Cross-
Nationally: Values, the Theory of Planned Behavior, and Value-Belief-Norm Theory.’
Environment and Behavior 38:462–483.

Pampel, Fred C. and Lori M. Hunter. 2012. ‘Cohort Change, Diffusion, and Support
for Environmental Spending in the United States.’ American Journal of Sociology
118:420–448.
Rootes, Christopher A. 1997. ‘Environmental Movements and Green Parties in Western and
Eastern Europe.’ In M. Redclift and G. Woodgate (eds.), The International Handbook
of Environmental Sociology. pp. 319–348. Cheltenham, UK: Edgar Elgar.
Schaffrin, André. 2011. ‘No Measure without Concept: A Critical Review on the
Conceptualization and Measurement of Environmental Concern.’ International Review
of Social Research 1:11–31.
Spaargaren, Gert. 2011. ‘Theories of Practices: Agency, Technology, and Culture. Exploring
the Relevance of Practice Theories for the Governance of Sustainable Consumption
Practices in the New World-Order.’ Global Environmental Change 21:813–822.
Steg, Linda, Lieke Dreijerink and Wokje Abrahamse. 2005. ‘Factors Influencing the
Acceptability of Energy Policies: A Test of VBN Theory.’ Journal of Environmental
Psychology 25:415–425.
Stern, Paul C. and Thomas Dietz. 1994. ‘The Value Basis of Environmental Concern.’
Journal of Social Issues 50:65–84.
Stern, Paul C., Thomas Dietz and J. Stanley Black. 1985–86. ‘Support for Environmental
Protection: The Role of Moral Norms.’ Population and Environment 8:204–222.
Stern, Paul C., Thomas Dietz and Gregory A. Guagnano. 1995. ‘The New Ecological
Paradigm in Social-Psychological Context.’ Environment and Behavior 27:723–743.
Stern, Paul C., Thomas Dietz, Troy Abel, Gregory A. Guagnano and Linda Kalof.
1999. ‘A Value-Belief-Norm Theory of Support for Social Movements: The Case of
Environmentalism.’ Human Ecology Review 6:91–97.
Stern, Paul C., Thomas Dietz and Linda Kalof. 1993. ‘Value Orientations, Gender, and
Environmental Concern.’ Environment and Behavior 25:322–348.
Vandenbergh, Michael P. 2005. ‘Order Without Social Norms: How Personal Norm
Activation Can Protect the Environment.’ Northwestern University Law Review
99:1101–1166.
Van Liere, Kent D. and Riley E. Dunlap. 1980. ‘The Social Bases of Environmental
Concern: A Review of Hypotheses, Explanations and Empirical Evidence.’ Public
Opinion Quarterly 44:181–199.
Van Liere, Kent D. and Riley E. Dunlap. 1981. ‘Environmental Concern: Does It Make a
Difference How It’s Measured?’ Environment and Behavior 13:651–676.
Xiao, Chenyang and Riley E. Dunlap. 2007. ‘Validating a Comprehensive Model of
Environmental Concern Cross-Nationally: A Canadian-USA Comparison.’ Social
Science Quarterly 88:471–493.
Xiao, Chenyang, Riley E. Dunlap and Dayong Hong. 2013. ‘The Nature and Social Bases
of Environmental Concern in China.’ Social Science Quarterly 94:672–690.
Common abbreviations
ISSP: International Social Survey Programme (www.issp.org)

ISSP 1993; ISSP 2000; ISSP 2010: International Social Survey Programme and
year of implementation of a specific questionnaire module
Acknowledgements
This book is the result of close cooperation with the activities of the European
Sociological Association (ESA) research network RN 12, ‘Environment and
Society’.
The co-authors of the book thank the national teams implementing the Inter
national Social Survey Programme (ISSP) in their countries, for their scientific
rigour and methodological accuracy. In addition, we would like to thank the draft-
ing teams of the ISSP Environment modules, implemented in 1993, 2000 and
2010, for crafting the tools for measuring environmental commitment.
Introduction
How green are Green Europeans?
Matthias Gross and Audronė Telešienė
European environmental behaviour

Every now and then a book comes along which barely requires an introduction
because it is widely acknowledged as filling, at long last, a conspicuous gap in
the literature. We believe this is one such book. Since the book that emerged
from the International Social Survey Programme (Shades of Green, edited by
Alan Frizzell and Jon Pammett) was published almost twenty years ago, it natu-
rally does not cover many more recent debates that are crucial for environmental
sociology and related fields today. In short, there has not been any other col-
lective, broad-based effort to cover European data, and to explore patterns of
environmental behaviour on a European scale. So the book you are holding in
your hands is indeed sorely needed.
In lieu of a standard introduction, then, let us mention a few issues that need
to be borne in mind when leafing through the following pages, including some
comments about the different intellectual arenas to which the various chapters
belong. The key aim of this book is to gather together different approaches that
seek to achieve a better understanding of types of environmental behaviour and of
environmentalism generally in Europe. The original idea arose out of conversa-
tions held with many colleagues during meetings of the European Sociological
Association (ESA), in the course of which we came to feel that a considerable
amount of groundwork is still to be done in European environmental sociology,
most notably comparative research on environmental attitudes, behaviour and
awareness. Oftentimes during presentations, people would ask how this or that
issue is perceived in, say, Poland compared to Denmark, or in Germany compared
to Portugal. More often than not it became clear that although there are case stud-
ies and datasets available (such as the International Social Survey Programme,
or ‘ISSP’), none of the relevant publications are readily accessible and none can
be used to address pressing contemporary issues such as environmental threats,
perceptions of environmental and technological risk, the relationship between
religion and environmental activism and, more generally, the European ‘shades of
green’ in a comparative global perspective, and the disputed notion of a general
‘green’ European identity. This book is intended to fill that gap. More than this,
2 Matthias Gross and Audronė Telešienė
it also seeks to bridge the gap (still considerable, especially in Europe) between
forms of methodologically driven empiricism on the one hand, and rather data-
free theory development on the other. Rather like the 1970s and 1980s in North
America, we still have different research communities spread throughout Europe
addressing public environmental attitudes without drawing in any significant way
on current sociological theory (cf. Buttel and Gijswijt 2004).
The questions addressed by the contributing authors to this volume include the
following:
•• What are the multiple cultural contexts underlying environmental world

views and behaviours within Europe?
•• Do these constitute an identifiably European cultural paradigm within which
environmentalism in Europe is played out?
•• If there is such a paradigm, in what ways is it unique in a global context?
What types of theories are most useful to better understand public environ-
mental attitudes and behaviour?
•• What attitudinal and behavioural patterns are characteristic of European soci-
eties with regard to environmental issues?
•• How are these patterns similar to or different from those in other regions
across the globe?
•• To what extent do European societies contain homogeneous or divergent
environmental world views and behaviours (cross-cultural comparative
approach)?
•• What developmental trends can be identified in relation to environmental
attitudes, risk perception and behaviour in Europe (historical comparative
approach)?
The focus is on attitudes and behaviours that are found among the general (non-
scientific) public, and on the differences that are rooted in diverse cultures and
national societies. One important overall issue addressed is that while many
European countries are widely considered to be ‘green’, or to have an advanced
level of environmental awareness, the question yet to be answered is, ‘Exactly
how green are Green Europeans?’ The book presents a variety of unique data-
driven comparative studies and is divided into three sections (see below). The
empirical comparative nature of the texts is enabled crucially by data from the
ISSP. The ISSP is an academic-led and globally coordinated attempt to conduct
comparative research on relevant topics. Environmental attitudes and behaviours
have been addressed by ISSP questionnaires three times: in 1993, 2000 and 2010.
A more detailed description of the programme can be found at www.issp.org.
Environmental sociology in Europe

In the 1970s Riley Dunlap and Kent van Liere (1978, 1984) developed a theory
that, in the late twentieth century, US society was in the midst of shifting from a
so-called human exemptionalist paradigm (HEP) to a new environmental para-

digm (NEP). There have been numerous attempts to validate this theory using
data generated within and outside the USA. Did European societies go through
the same shift in awareness, or did they follow a different trajectory with regard
to a developing environmental sensibility? In order to shift from the HEP to the
NEP, a society has first to adopt the HEP and then be accelerated by affluence, a
post-modern or an advanced industrialized value system, in order to shift towards
the NEP. But in terms of developmental theory, this is a narrow, uni-linear under-
standing of social change, which assumes that all the different European societies
will somehow follow the same developmental trajectory. Some European soci-
eties have never exhibited the human exemptionalist state of mind. Indeed, it
might well be that some societies are only now beginning to display something
akin to the HEP. This being so, we need to look for social change models other
than HEP–NEP; ones that are grounded instead in empirical data and interpreta-
tions rooted in European cultural contexts. The task of identifying the multiple
faces of European environmentalism is dogged by a lack of comparative studies
based on reliable empirical data. The task is made even more difficult given that
environmental sociology has developed in very different ways in the different
regions of Europe.
Much sociological research on the environment conducted in Europe since the
1970s has focused on conceptualizing the environmental movement as part of
new social movements, coupled with new forms of environmental awareness, as
well as new behaviour and new environmental policy (Brand 2010). This has
often been followed up – especially in the UK, the Netherlands, Scandinavia and
German-speaking countries – by efforts to study lifestyles and consumption pat-
terns. Later, in the 1990s, ecological modernization (originally a label for a policy
strategy in Germany; cf. Huber 1982) was developed by some sociologists as a
new theory of industrial greening (Mol et al. 2009).
As Hellmuth Lange (2011) has argued, France not only has the longest European
tradition in sociology (from Auguste Comte to Emile Durkheim, of course) but
also – thanks to the influence of its Jacobin traditions – a relatively specific politi-
cal culture: on the one hand, individualistic, republican and egalitarian; and on the
other, centralized and elitist. It is in this context that the sociological debate about
the relationship between nature and society started in the late 1960s, among intel-
lectuals such as Serge Moscovici, Henri Lefebvre, Alain Touraine, Edgar Morin
und André Gortz (Lange 2011: 28). Ecological issues, however, were discussed
mainly in response to efforts by the French government of the time to ‘modernize’
French agriculture, which was considered to be too ‘small-scale’. Not surprisingly,
the main input into this debate came from rural sociology (sociologie rurale). The
rather conservative outlook of rural (environmental) sociology in France, how-
ever, did not foster any comparative research on a European scale. The situation in
Germany was not much better (cf. Lange 2002).
In Central and Eastern-European countries (CEEs), where the sociological
imagination developed behind the Iron Curtain of socialist states, voices in the
field of environmental sociology long remained silent. Then, in the late 1980s,
processes of national renaissance and perestroika in CEEs became hot topics
for the international sociological community. The environmental legacies of
communist-state-planned economies, combined with surprisingly high levels
of environmental public awareness, as well as growing numbers of members of
environmental movements, became a focus for environmental sociologists, not
only within CEEs but also in Western and Northern Europe and the USA (e.g.
Waller and Millard, 1992; Taylor, 1995). As Dunlap (2011) notes, these develop-
ments contributed at that time to the revitalization of the field of environmental
sociology. Since the 1990s a range of topics have been elaborated within envi-
ronmental sociology in the CEE region, including research on environmental
movements, environmental behaviours and environmental awareness (e.g. Lang-
Pickvance et al. 1997, Rinkevicius 2000, Glinski 2001, Jehlicka 2001), green
parties and green voting behaviour (e.g. Yanitsky 2001, Podoba 1998), risk per-
ception and acceptability (e.g. Balžekienė 2009) and sustainable development
(e.g. Tysiachniouk 2000). Such studies, along with more recent developments in
environmental sociology in CEEs, are generally based on empirical research, but
there is a lack of comparative approaches (apart from a few exceptions, such as
Domazet and Jerolimov 2014).
Similarly, environmental sociology in Germany in the 1980s was lacking in
contributions to cross-cultural and comparative research, at least in comparison to
North-American approaches. Nevertheless, in 1986 two influential works appeared:
Niklas Luhmann’s Ecological Communication, and Ulrich Beck’s Risk Society.
Luhmann’s systems theoretical approach derived from the sociology of the 1950s
and 1960s, and was based largely on the work of Talcott Parsons (1901–1979).
Parsons (1951, 1977) is very clear in arguing that human action is bounded not just
by cultural constraints and norms, but also by its physical environment. Parsons’s
ideal type model (generally known as the AGIL model) depicts the social system
and its subsystems as a fourfold table consisting of: (1) adaptive function; (2) goal
attainment; (3) integrative function; and (4) latency. The adaptive function is con-
cerned with society’s subsistence production and hence with people’s relationship
to their physical environment, including natural resources and climatic conditions.
In Parsons’s ‘Theory of Action’, human action is unquestionably bounded by the
natural environment, as well as by cultural constraints of other (human) members
of society. Societies continually respond to the limits set by their natural environ-
ment, evolving in complexity and capacity so as to transcend these limits.
Parsons’s model has been criticized on different levels. In particular, it has
been argued that his theory is not able to explain radical changes in social life, and
that its empirical potential is severely limited owing to a high level of abstraction.
However, since the 1970s there has been a revival of interest in Parsonian sociol-
ogy, especially in Germany, and linked generally to the figure of Niklas Luhmann
(1927–1998). Unlike Parsons, Luhmann’s notion of ecological communication
leads us yet further away from actual nature; in fact, it can be said that Luhmann’s
approach is located more within the sociology of environmental communication
than in the field of environmental sociology. Luhmann proceeds from the notion
of society being differentiated according to functional systems: religion, the
economy, the arts, law, and science – but, interestingly enough, not the physical
world. ‘It should be noted,’ Luhmann writes, ‘that [ecological communication]
is a phenomenon that is exclusively internal to society’ (Luhmann 1989: 28). In
his theory, actions are valued in terms of their meaning for the sender and the
receiver. Hence it is systemic communications and their specific codes that are
sociologically important in Luhmann’s view. In the case of the environment this

means that the medium of communication might be either good or bad, sustaina-
ble or unsustainable. The code here is the difference between several subsystems,
especially those of law (justice and injustice) and of the economy (payment and
non-payment); that is to say, Luhmann does not accept that ecological informa-
tion is needed to achieve ecological communication. One can easily see that using
Luhmann’s theory as a framework to analyse environmental attitudes via a theory
of communication is possible, but this highly abstract theory does not inspire any
large-scale empirical research.
A seemingly more comparative approach in contemporary environmental soci-
ology can be found in Ulrich Beck’s Risk Society (1992 [1986]). From given,
objective environmental problems, Beck suggests, it follows that modern society
is turning into a form of society in which the major political conflicts are not class
struggles over the distribution of wealth, but rather non-class-based struggles
over the distribution of environmental and technological risks. Beck suggests that
the driving force in nineteenth-century class society could be summarized in the
phrase ‘I am hungry’, whereas in the risk society the driving force of the ecology
movement could be summarized by the statement ‘I am afraid’ (Beck 1992: 49).
Beck offers hope: if the frightened public can wrest back control from science, it
may lead to what Beck calls reflexive modernization. This new type of reflexive
society that he foresees is one that thinks critically about where technology and
science are going in terms of their exploitation and degradation of the physical
world. In Beck’s view, risk is fundamentally reshaping the politics of our times –
not only in Germany, but also in Europe and around the globe, as his notion of
the ‘world risk society’ assumes (cf. Beck 2006). If he turns out to be right in the
long run, then the twenty-first century is bound to constitute a turning point for
good or ill in the human relationship to the natural environment. With his concept
of a risk society, Beck establishes an original and far-reaching foundation for the
analysis of modern politics. However, in Ulrich Beck’s theoretical approach, the
relationship between nature and society is shaped crucially by the notion of risk
as a guiding principle. Unfortunately, for our purposes, Beck pursued this focus
largely to the exclusion of corresponding empirical research and historical and
cross-cultural perspectives.
The twenty-first century has so far witnessed something of a convergence
between research activities in European environmental sociology, although
there are still major differences (cf. Mol 2006, Lidskog et al. 2015). The issue
of the developmental prospects of rural areas in France, for instance, has been
complemented by the more generalized ones of industrial risks, sustainable con-

sumption and transport.
One characteristic common to the environmental sociological community
in countries such as Greece, France, Portugal and Spain is a strong orientation
towards identifying the economic and broader social causes of environmental
damage, and looking at the social groups affected in their own localities. Thus
a significant proportion of the sociology community (and its work) is character-
ized by a deliberate proximity to environmental groups (Lange 2011), and by

what Frederick Buttel and August Gijswijt once prominently called the ‘partisan
flavour’ of environmental sociology (2004: 49). However, the self-proclaimed ren-
egade status of much of the environmental sociology community in Europe – much
like in North America – has worn thin in many ways. It is now time for a collec-
tion that abstracts from country-specific research and gathers together insights that
unite a variety of unique historical and cross-cultural comparative studies, focused
mainly on European societies. The following chapters discuss the development of
environmental attitudes, risk perceptions and patterns of environmental behaviour,
shedding light on the universal or indeed fragmented nature of what we would
consider a European type of environmentalism – or indeed a European type of
environmental sociology.
Research on environmental behaviour and environmental attitudes (or other
aspects of environmental world views) in Europe has been based largely on non-
comparative methodological approaches. This is owing to the lack of research
projects with an international – and specifically European – comparative focus
(for a detailed account of existing international research programmes, see Haller
et al. 2009). ISSP Environment modules are unique and have been the engine that
has generated data and enabled comparative analyses for several decades. The
existing comparative studies in the field that include European societies gener-
ally rely on ISSP data, e.g. Franzen and Meyer (2010), Hadler and Haller (2011).
Some twenty years ago there was a collective effort to analyse the ISSP Envi-
ronment data in detail: ISSP published the anthology Shades of Green in 1997
(Frizzell & Pammett 1997), but there has been no other collective or broad-based
effort to cover European data. Our hope is that this book will fill that gap.
One significant general issue addressed by this research is that many European
countries are considered to be ‘green’, or to have a high level of environmental
awareness. According to commentators outside Europe, the European Union has a
remarkable ability to agree on recycling, energy consumption and climate-change
issues, to name just three. The assumption implicit here is that environmental
awareness in Europe must also be exceptionally high. But what about the day-to-
day habits of European citizens? Is it possible, for example, that country-specific
attitudes have, for some reason, prompted the emergence of a generalized percep-
tion of how ‘Europeans’ are? After all, Europeans’ apparent ‘love’ for recycling
is very ardent in some regions and rather weak in others.
The contributions to this edited volume are divided into three main sections: envi-
ronmental behaviour, environmental risk perception and environmental attitudes.
These reflect some of the key dimensions of environmentalism, including the major
sets of analytic concepts.
The contributions to this book

Throughout this volume the contributing authors provide competing explanations
of environmental commitment in Europe. Each of the chapters targets one or
more conceptually diverse topics related to environmental world views, behav-

iours or risk perceptions. The authors variously employ classical theories used in
environmental sociology, including rational choice theory, the theory of cognitive
mobilization, the theory of ecological modernization, Stern’s theory of significant
environmental behaviour or the theory of planned behaviour. They consider the
influence of religious beliefs and social capital, the influence of people’s exposure
to environmental degradation, post-materialism hypothesis and ‘objective prob-
lems-subjective values’ approaches, while also considering the ‘democratization’
of different environmental risks. Innovative approaches are also given space.
There are interesting ‘knowledge transfers’ from other disciplines and fields of
sociology, such as degrowth theory, cultural lag theory and the influence of the
pace of societal transformations, hypothesis of hysteresis effect, and widening
use of the societal affluence concept.
Our book is divided into three sections. The first addresses perceptions of
environmental and technological threats and risks. The second deals with envi-
ronmental activism in Europe, and the third discusses environmental attitudes,
environmental concerns and their inherent link to personal pro-environmental
behaviour.
Markus Hadler and Klaus Kraemer open the book with an international com-
parative analysis, contrasting perceptions of environmental threat in a number
of European and non-European countries. Their focus is on environmental
threats perceived to be posed to an individual or their family. The main finding
describes the differentiation between public perceptions of two types of envi-
ronmental issues: climate/nature-induced issues and socially induced issues.
Aistė Balžekienė and Audronė Telešienė analyse perceptions of environmental
and technological risk, and their underlying determinants. The authors identify
regional differences and describe tests for the influence of individual structural
and cognitive factors upon risk perception. They also demonstrate that the influ-
ence of country-specific contextual factors is strong and relate this influence
to the dynamism of socio-economic systems within a country. They conclude
that levels of environmental risk perception are lower among individuals with
a more secure social status and living environment, and a more stable socio-
economic context. André Schaffrin and Alexander Schmidt-Catran conclude
the first part of this book with a longitudinal and multilevel approach. They
conceptualize concerns about environmental threats and risks as being two-
dimensional, and then examine how post-materialism and affluence (among
other factors) influence the different dimensions of these concerns. The authors
show that European populations are not homogeneous with regard to an aware-
ness of environmental threats, environmental efficacy and the willingness to pay
for environmental goods. They thus cast doubts on any such notion of a uniform
‘green’ European identity.
The second part of the book opens with Jean Paul Bozonnet’s look at envi-
ronmental activism in different European cultures. The author views activism
through the lens of national culture and introduces a theory of the ‘cultural hys-
teresis effect’. He shows how cultural and social structures carried over from
the past might influence the current ‘faces’ of environmentalism in Europe.
The chapter also includes a discussion of the relationship between religion and
environmental activism. Eglė Butkevičienė and Vaidas Morkevičius further the
discussion by providing a cross-national comparative and fuzzy-set qualitative
comparative analysis of public involvement in pro-environmental groups in
European countries. They demonstrate that the affluence of a country and high
levels of environmentally friendly behaviour are uniformly frequent features of
environmental group membership in Europe. Iosif Botetzagias, Nikoleta Jones
and Chrysovalantis Malesios analyse the profiles of Europeans who engage in
either one of the three distinct public-sphere pro-environmental behaviours: sign-
ing a petition, contributing money and demonstrating for the environment. They
provide an empirically grounded description of a typical European petitioner,
contributor or protester, and also examine temporal variations between the three
pro-environmental behaviours.
In Chapter 7, which opens section three of the book, Mladen Domazet and
Branko Ančić explore the extent to which environmentally motivated degrowth
potential exists among European populations. The reader will find some support
here for a green European identity. The authors argue convincingly that what
they call ‘prosperity environmentalism’ is a characteristic European phenom-
enon. Yet the reader will also find that regional differences within Europe are
considerable, with Central and Eastern-European populations being less prepared
to make personal sacrifices for the sake of the environment. Jochen Mayerl then
extends the analysis of environmental concern to look at European populations
in a comparative global perspective. In doing so, he also presents testing results
of measurement equivalence, and includes a thorough discussion and validation
of cross-cultural comparisons of environmental concern measures as taken from
ISSP 2010. The chapter will be of huge value to those interested in the correct-
ness and validity of comparative research. Leena Haanpää in Chapter 9 takes a
youth-oriented approach, presenting a cross-national study on environmental con-
cern and behaviours among young people in Europe. The author employs theories
of planned behaviour and post-materialism. In Chapter 10 João Guerra, Luísa
Schmidt and Susana Valente discuss the impact of economic crisis on sustainabil-
ity attitudes. They discuss how competing narratives – those of unlimited growth
and sustainable development – are reflected in the attitudes of Europeans (as well
as other populations around the world) under the pressure of economic scarcity.
This chapter is valuable in its comparative data on different regions of the world:
North America, Latin America, Western Europe, Eastern Europe and Southern
Europe. Special attention is paid to the example of Portugal.
By exploring forms of environmental behaviour and attitudes in a comparative
perspective and, along the way, raising the question of how ‘green’ Europeans
really are in comparison to other parts of the word – and what ‘green’ may mean
in the first place – these chapters constitute an invitation and an encouragement
to join in the long tradition of research on pro-environmental behaviour and prac-
tices, this time with a clear European focus. With Europe undergoing massive
political, cultural, and economic transformation, the ecological aspects of these
changes deserve to be addressed more pivotally by sociologists.
References
Balžekienė, Aistė (2009): Rizikos suvokimas: sociologinė konceptualizacija ir visuomenės
nuomonės tyrimo metodologinės prielaidos [Risk perception: sociological concep-
tualization and methodological approaches to public opinion research]. FilosoFija.
sociologija 20 (4): 217–226.
Beck, Ulrich (1992 [1986]): Risk Society: Towards a New Modernity. London: Sage.
Beck, Ulrich (2006): Living in the World Risk Society. Economy and Society 35 (3): 329–345.
Brand, Karl-Werner (2010): German Environmentalism: Still Feeding on Its Romantic,
Anti-Modern Heritage? Nature + Culture 5 (2): 209–226.
Buttel, Frederick and August Gijswijt (2004): Emerging Trends in Environmental
Sociology. In: Blau, Judith R., ed., The Blackwell Companion to Sociology. New York:
Blackwell, 43–57.
Domazet, Mladen and Dinka Marinović Jerolimov, eds. (2014): Sustainability Perspectives
from the European Semi-periphery. Zagreb: Institute for Social Research in Zagreb.
Dunlap, Riley E. (2011): Aktuelle Entwicklungen in der nordamerikanischen
Umweltsoziologie [Current Developments in North American Environmental Socio
logy]. In: Groß, Matthias, ed., Handbuch Umweltsoziologie. Wiesbaden, Germany: VS,
54–72.
Dunlap, Riley E. and Kent D. van Liere (1978): The ‘New Environmental Paradigm’: A
Proposed Measuring Instrument and Preliminary Results. Journal of Environmental
Education 9 (1): 10–19.
Dunlap, Riley E. and Kent D. van Liere (1984): Commitment to the Dominant Social Para
digm and Concern for Environmental Wuality. Social Science Quarterly 65: 1013–1028.
Franzen, Axel and Reto Meyer (2010): Environmental Attitudes in Cross-National
Perspective: A Multilevel Analysis of the ISSP 1993 and 2000. European Sociological
Review 26 (2): 219–234.
Frizzell, Alan and Jon H. Pammett, eds. (1997): Shades of Green: Environmental Attitudes
in Canada and Around the World. Ottawa, Canada: Carleton University Press.
Glinski, Piotr (2001). The Ecological Movement as the Element of the Civil Society. In
Flam, Helena (ed.) Pink, Purple, Green: Women’s, Religious, Environmental and Gay/
Lesbian Movements in Central Europe Today. Boulder, CO: East European Monographs,
112–119.
Hadler, Markus and Max Haller (2011): Global Activism and Nationally Driven
Recycling: The Influence of World Society and National Contexts on Public and Private
Environmental Behavior. International Sociology 26 (3): 315–345.
Haller, Max, Roger Jowell, and Tom W. Smith, eds. (2009). The International Social
Survey Programme 1984–2009: Charting the Globe. London: Routledge.
Huber, Joseph (1982): Die verlorene Unschuld der Ökologie: Neue Technologien und
superindustrielle Entwicklung [The Lost Innocence of Ecology: New Technologies and
Super-Industrial Development]. Frankfurt am Main, Germany: Fischer.
Jehlicka, Petr (2001). The New Subversives: Czech Environmentalists after 1989. In: Flam,
Helena (ed.) Pink, Purple, Green: Women’s, Religious, Environmental, and Gay/Lesbian
Movements in Central Europe Today. Boulder, CO: East European Monographs, 81–94.
Lange, Hellmuth (2002): Social Science and Nature: A Review of Environmental Sociology
in Germany. Natures, Sciences, Sociétés 10 (1):16–26.
Lange, Hellmuth (2011): Umweltsoziologie in Deutschland und Europa [Environmental
Sociology in Germany and Europe]. In: Groß, Matthias, ed., Handbuch Umweltsoziologie.
Wiesbaden, Germany: VS, 19–53.
Lang-Pickvance, Katy, Nick Manning, and Chris Pickvance eds. (1997): Environmental
and Housing Movements: Grassroots Experience in Hungary, Russia and Estonia.
Aldershot, UK: Avebury.
Lidskog, Rolf; Mol, Arthur P. J, and Peter Oosterveer (2015): Towards a Global
Environmental Sociology? Legacies, Trends and Future Directions. Current Sociology
63 (3): 339–368.
Luhmann, Niklas (1989 [1986]): Ecological Communication. Chicago: University of
Chicago Press.
Mol, Arthur P. J. (2006): From Environmental Sociologies to Environmental Sociology?
A Comparison of U.S. and European Environmental Sociology. Organization &
Environment 19 (1): 5–27.
Mol, Arthur P.J.; David A. Sonnenfeld, and Gert Spaargaren, eds., (2009): The Ecological
Modernisation Reader: Environmental Reform in Theory and Practice. London:
Routledge.
Parsons, Talcott. (1951): The Social System. Glencoe, IL: Free Press.
Parsons, Talcott (1977): Social Systems and the Evolution of Action Theory. Glencoe, IL:
Free Press.
Podoba, Juraj (1998): Rejecting Green Velvet: Transition, Environment and Nationalism in
Slovakia. Environmental Politics 7 (1): 129–144.
Rinkevicius, Leonardus (2000). Transitional Economies – Ecological Modernisation as
Cultural Politics: Transformations of Civic Environmental Activism in Lithuania.
Environmental Politics 9 (1): 171–202.
Taylor, Bron Raymond, ed. (1995) Ecological Resistance Movements: The Global
Emergence of Radical and Popular Environmentalism. Albany: State University of
New York Press
Tysiachniouk, Maria (2000): Sustainable Development through the Efforts of the Third
Sector in Russia. Interdisciplinary Environmental Review 1 (1): 24–67.
Waller Michale and Frances Millard (1992): Environmental Politics in Eastern Europe.
Environmental Politics 1 (2): 159–185.
Yanitsky, Oleg (2001): Russian Greens in a Risk Society. Helsinki: Kikimora.
Part I
threats and risks

Reconsidering environmental

Chapter 1
The perception of environmental

threats in a global and European
perspective
Markus Hadler and Klaus Kraemer
Introduction
The German sociologist Ulrich Beck (1986: 48) posited that some environmental
threats are distributed rather equally within societies and independent of clas-
sic social inequalities such as income and class. What Beck had in mind were
new environmental threats such as nuclear power and air pollution that cannot be
stopped by borders, social inequalities or privileges. His view was corroborated
by the nuclear incident in Chernobyl, which happened shortly after his book was
published. Henceforward, the discussion of the democratization of environmental
risks dominated over approaches focusing on environmental justice, pointing out
that there is a clear connection between the social position and exposure to certain
environmental hazards and threats (Kraemer 2008).
The presence of different environmental hazards and threats can differ
geographically, both within societies and between societies, with some being
related to the natural distribution of environmental qualities and others to social
factors, such as economic development, type of industries and political meas-
ures (see Kraemer 2014; Lutzenhiser & Hackett 1993). At the country level,
for example, we can see that the impact on the environment is usually greater
in more developed countries (York, Eugene & Dietz 2003). However, we also
need to take into account that not all threats increase in the same way with
economic development, and that societies can implement different measures to
offset environmental impacts (Kraemer 2011). Thus, even within countries, we
can observe the presence of different threats, depending on the area of residency
such as suburbs, industrial zones, etc., and more generally between urban and
rural areas (Grant et al. 2010; Lichter & Brown 2011).
In addition, we also need to ask which threats are actually perceived.
Many environmental hazards only influence an individual’s life very indi-
rectly, through complex social and ecological distribution mechanisms that are
far removed from everyday experiences (Kraemer 2008). And even when a
particular threat is assessed, research focusing on the perception of risks has
shown that the perception of the same risk varies between social groups and
experts (Keller et al. 2012; Slimak & Dietz 2006; Slovic 1987). In addition,
14 Markus Hadler and Klaus Kraemer
international comparative research, focusing on environmental attitudes and

behaviours, also points to substantial variations across countries (Franzen &
Meyer 2004; Hadler & Haller 2011; Haller & Hadler 2008).
Our contribution considers both: the influence of the different environmen-
tal hazards, and individual characteristics on the perception of environmental
threats. We focus on a broad range of risks ranging from water shortage to climate
change. Environmental risks in this regard refer to environmental problems that
are perceived as a threat for the respondents’ life, or as a threat for their family.
Environmentally damaging behaviours of our respondents that are often referred
to as ‘environmental risks’ are not addressed.
This introduction is followed by a brief discussion of natural and social aspects
of the distribution of environmental good and bad. Consequently, we discuss the
individual-level determinants of the perception of environmental threats. The
data and methods section introduces the 2010 ISSP data, our dependent and inde-
pendent variables, as well as the analysis strategy. The results section starts with
an overview of threat perception at the country level. This is followed by an
analysis of country-level and individual determinants of environmental-threat
perception. The discussion and conclusion section summarizes our main find-
ings, and concludes that the threat perception depends on both contextual and
individual characteristics.
Distribution of environmental good and bad

Considering the distribution of environmental good and bad, we can distinguish
four dimensions (see Kraemer 2008: 175ff). The first refers to a natural distribu-
tion of environmental characteristics related to differences in climate, geography
and natural resources that are relatively independent from human intervention.
The second dimension refers to human-induced problems that affect the envi-
ronment. Some of these human-induced problems are distributed rather evenly
across all social groups – which is at the centre of Beck’s (1986) democratization
of environmental risks idea – and thus less important when addressing their rela-
tionship to social inequalities.
The third dimension refers to the fact that the social distribution of environ-
mental good and bad is not distributed randomly; an observation that is addressed
within the environmental justice literature (Bolte & Mielck 2004; Elvers, Gross &
Heinrichs 2008). A typical example is pollutants emitted from factories that more
often affect local residents than individuals who live far away. Furthermore, such
public ‘bad’ more often affects social groups in lower social positions and with
other social disadvantages (see, for example, Freeman 1974; Grant et al. 2010).
At the international level, we need to consider that the level of development and
the type of economy affects households differently. Whereas the overall impact
on the environment increases with an increasing economic development (York
et al. 2003), the negative impact of certain pollutants can actually decrease, as
discussed within the ecological modernization literature. The use of fossil fuels
Perception of environmental threats 15
in everyday activities such as cooking, for example, has a major impact on low
household air quality, whereas this form of pollution is of little importance in
more developed societies (Yale University 2014).
The fourth dimension refers to political interventions and the different ways
environmental hazards are addressed. Proponents of ecological modernization
have suggested that increasing development can be used to address environmen-
tal threats and implement more environmentally sustainable means of production
(Harper 2012: 176ff). The empirical evidence, however, is inconclusive, as tech-

nical solutions such as filters for factories can lower the impact of a particular
emission, whereas other hazards such as climate change cannot be addressed as
efficiently (Kraemer 2011). In addition, the implementation of different measures
is also highly dependent on political actors and other stakeholders. The debates
around ecological modernization have thus come to the conclusion that not all
aspects can be addressed, but that, in any case, a minimum level of resources and
affluence is necessary to address the most pressing environmental threats. For
international comparison, we thus can expect that the population of countries,
which are part of a political community such as the European Union, with com-
mon environmental regulations (Hix & Høyland 2011), will be exposed to similar
standards and thus should perceive similar threats.
Individual differences in the perception of

environmental threats
Insights into the determinants of individuals’ threat perceptions can be drawn
from research on risk perception. This type of research was set in motion in
the 1960s by the observation that experts and laymen had contrary views about
the risks of nuclear power (Keller et al. 2012). It was assessed as quite safe by
experts, but considered dangerous by the general public. Slovic, Fischhoff and
Lichtenstein (1981) and Slovic (1987) thus applied a psychometric approach,
and studied the underlying dimensions of how different risks are seen. They were
able to identify three main dimensions: 1) ‘Dread’, which refers to considering
a risk to be potentially catastrophic; 2) ‘Unknown’, which refers to the observ-
ability and familiarity of a risk; and 3) ‘Exposure’, which refers to how many
people can be potentially affected. Nuclear power ranks highly on these dimen-
sions, whereas everyday activities such as using power tools, for example, rank
rather low. These studies show that the perception of very different risks can be
described by similar underlying dimensions, which may also apply to the percep-
tions of environmental threats.
These initial approaches focused on the question of how different risks are
perceived. However, they did not consider why the perception of the same risk
can differ between social groups and individuals. Dake (1992) thus extended these
initial psychometric approaches and included the influence of world views, in
addition to the influence of different cognitive heuristics. More recent approaches
(Keller et al. 2012) also consider place attachment and identities, agency and trust
in institutions. For example, individuals are less likely to perceive a threat if they
have a strong attachment to the area where they live, if they think that they are
able to handle a threat, and when they trust the agencies that are responsible for
dealing with the environmental hazard (see Bickerstaff 2004).
In contrast to their focus on psychological determinants of risk perceptions,
approaches such as that of the sociologist Ulrick Beck (1986) and the anthropolo-
gist Mary Douglas (Douglas 1992) focus rather on the impact of environmental
threats on an entire society (see also Wilkinson 2001). As mentioned before,

Beck considers environmental threats, which are caused by modern forms of pro-
duction and can thus impact all social groups in the same way, as independent
from classic social inequalities such as socio-economic class. Beck, however,
also recognizes that some groups will be better able to avoid some threats. In this
regard, he considers knowledge an important factor. Individuals who are aware
of certain threats and also know how to deal with these problems will more likely
show avoidance strategies.
Mary Douglas (Douglas 1992; Douglas & Wildavsky 1982) likens modern
ecological end-of-the-world scenarios with the doomsday prophecies of various
religious writings, only that the ecological version is more appealing to modern,
rational-thinking individuals. When it comes to differences in the perception of
environmental threats, Douglas refers to her grid and group theory, considering
group embedding and social control that result in four different ways of life:
individualistic, hierarchical, egalitarian and fatalistic. All four forms differ in
the perception of threat, with individuals who live in small, tight-knit groups
tending more to fundamentalist attitudes towards environmental threats than
actors who live in rather loosely integrated groups (Schwarz & Thompson 1990;
Thompson et al. 1990).
In line with these approaches, international comparative research focusing
on environmental attitudes, concerns and threat perception, considers socio-
demographics, values and contextual characteristics in explaining differences in
individual perceptions (see Fairbrother 2013; Hadler & Haller 2011; Haller &
Hadler 2008). A frequently used approach is Stern’s Value-Belief-Norm schema
(Stern 2000) that has been applied specifically to the perception of environmen-
tal risks by Slimak & Dietz (2006). Their results showed that different values
and attitudes are particularly important in determining the risk perceptions of the
general public, whereas the views and opinions of experts were not influenced as
strongly by these traits.
These different approaches lead to the following general hypotheses regard-
ing the perception of environmental threats:1 First, we can expect that these
perceptions are influenced by differences in the natural and social distribution
of different environmental good and bad, by the level of affluence and the way
environmental threats have been addressed by policymakers. At the same time,
we can also expect that differences are present between individuals. A first
difference should emerge regarding the different dimensions of risks – dread,
unknown, and exposure – and a second difference regarding the individual
determinants of the threat perception. Here, our expectation is that the per-
ception of different threats is influenced by environmental efficacy, as well as
different socio-demographic characteristics.
Data and methods

The empirical analysis is based on public opinion data collected by the International
Social Survey Programme (ISSP Research Group 2012). This ISSP Environment
module includes 60 questions on environmental attitudes and behaviours, plus
a comprehensive set of background variables. The data was collected in a total
of 36 countries and is representative of the adult population of each country. A
first cumulative dataset including 32 countries was published by the German
data archive GESIS in 2012. We also added data from Australia, Iceland, the
Netherlands and Portugal – countries where the data was collected or submit-
ted after the official census date of ISSP. These additional four datasets were
published in 2014 and are available at GESIS (Arnalds & Sigurbjörns-Öldudóttir
2014; Evans 2014; Ganzeboom 2014; Vala & Ramos 2014).
Our dependent variable is the response to the following statement: ‘Here is
a list of some different environmental problems. Which problem, if any, affects
you and your family the most? 1) Air pollution, 2) Chemicals and pesticides,
3) Water shortage, 4) Water pollution, 5) Nuclear waste, 6) Domestic waste dis-
posal, 7) Climate change, 8) Genetically modified foods, 9) Using up our natural
resources, 10) None of these.’ Respondents had to choose a single item.
The results section starts with an overview of the perception of these problems
in different countries, based on the unweighted data. Subsequently, the results of a
correspondence analysis of the different responses to these threats in our countries
are presented. These results indicate four distinct patterns of threat perception: a
pattern around ‘water issues’, such as water shortage; a pattern around ‘everyday
problems’, such as domestic waste disposal; a pattern of ‘high risks’, such as
nuclear power; and a distinct pole of ‘None’ of the mentioned threats considered
important. Using these patterns as a guide, the initial list of different threats was
recoded into these four different patterns and used as a dependent variable in the
subsequent analyses.
These analyses start firstly at the county level and consider the correlations
among threat perceptions, societal affluence and EU membership, in order to
test our hypotheses on the association of wealth, political interventions and risk
perceptions. Social affluence is measured by GDP in current USD, (data source:
World Bank 2015), and EU membership is defined as being a member country at
the time of the survey (see Table 1.1).
Secondly, multinomial regressions are used to test our hypotheses on the
influence of environmental efficacy as well as different socio-demographic charac-
teristics. Dependent variables are the four different patterns of threat perceptions.
Independent variables are: gender (female = 1); age (in years), education (years in
school); subjective social position (low to high, on a 10-point scale); marital status
(married or in a relationship versus single); presence of children in the household

(yes = 1); and area of residency (rural to urban, on a 5-point scale). Environmental
self-efficacy was measured by an index based on the five items (out of seven
items on efficacy), which load on a single dimension in a factor analysis: ‘It is
just too difficult for someone like me to do much about the environment’; ‘There
are more important things to do in life than protect the environment’; ‘There is
no point in doing what I can for the environment unless others do the same’;
‘Many of the claims about environmental threats are exaggerated’; ‘I find it hard to
know whether the way I live is helpful or harmful to the environment’. The mean
value across these five items was calculated for each respondent, based on their
responses to the 5-point scales, ranging from strongly agree to strongly disagree.
Perception of environmental threats in an

international comparison
Table 1.1 provides an overview of which threats are perceived in our 36 countries.
Considering the total sample, air pollution is mentioned the most often (19.5%),
followed by climate change (11.5%) and domestic waste disposal (11.5%), chem-
icals and pesticides (10.9%), genetically modified foods (9.5%), water shortage
(7.7%), using up our natural resources (7.2%) and nuclear waste (3.3%). Another
8% of the respondents stated that none of these problems is affecting them or their
family. Comparing the ranking of these threats within the EU countries with the
overall ranking reveals that water shortage is less important and chemicals and
pesticides are more important within the EU countries. Furthermore, the percep-
tion of environmental threats is more similar within the EU countries than in the
remaining set of countries.2
Alongside these commonalities, some countries show very distinct patterns.
Problems of water shortage and water pollution are perceived very often in
Mexico, South Africa, and Israel; countries that are located in rather dry sub-
tropical climates. These perceptions are clearly influenced by the geography and
climate of these countries. Considering the official World Bank (2014a) statistics
on access to clean water, around 12% of the rural population does not have access
to clean water in Mexico, whereas access to clean water is not problematic in
European countries according to this data source. In addition, as in the case of
Israel, the exploitation of the water resources may be important as well. Israel uses
2.7 times more water each year than is replenished naturally during this period,
whereas countries such as Austria – where only 1.4% of the population mentions
water shortage as the main risk – use only 7% of the natural resources each year
(World Bank 2014b).
This interpretation points to an association between the perception of water
problems and geography and climate. This pattern is also partially present regard-
ing air pollution. Air pollution, for example, is mentioned most often in Belgium
(36.9%) and the Netherlands (30.4%), and least often in New Zealand (8%)
and Australia (9.8%). According to an air-quality indicator developed by Yale
Table 1.1 Perception of environmental threats across countries (ISSP 2010)
Air Chemicals Water Water Nuclear Domestic Climate Genetically Using up None of N
pollution and shortage pollution waste waste change modified our natural these
pesticides disposal foods resources
Argentina AR 16.7% 7.4% 11.2% 25.6% 2.1% 14.1% 9.6% 4.0% 7.4% 2.0% 1,063
Australia AU 9.8% 12.7% 16.7% 4.6% 1.3% 9.9% 11.6% 11.6% 9.6% 12.0% 1,764
Austria* AT 16.4% 7.7% 1.4% 6.5% 2.7% 12.1% 20.7% 15.3% 5.6% 11.6% 967
Belgium* BE 36.9% 10.3% 2.7% 7.5% 1.6% 14.6% 9.0% 2.4% 6.5% 8.5% 1,030
Bulgaria* BG 24.3% 10.3% 8.1% 12.5% 2.3% 15.6% 8.1% 10.6% 2.8% 5.3% 941
Canada CA 17.5% 17.6% 2.6% 11.1% 1.6% 10.9% 12.7% 9.3% 9.1% 7.5% 910
Chile CL 27.8% 9.3% 10.0% 10.2% 7.1% 13.1% 8.3% 4.2% 5.7% 4.4% 1,416
Croatia HR 13.4% 11.7% 6.0% 15.3% 4.5% 12.5% 11.2% 12.9% 8.9% 3.7% 1,167
Czech Republic* CZ 29.4% 8.8% 2.9% 5.7% 3.3% 14.3% 10.1% 3.7% 5.8% 16.0% 1,324
Denmark* DK 16.7% 22.8% 2.1% 7.3% 1.0% 8.0% 19.1% 10.4% 8.6% 4.1% 1,049
France* FR 14.3% 24.4% 4.4% 11.1% 5.1% 11.5% 7.4% 9.4% 10.1% 2.4% 2,060
Finland* FI 15.4% 12.9% 3.6% 16.8% 2.6% 13.5% 12.8% 5.8% 7.5% 9.0% 1,069
Germany* DE 15.4% 9.1% 1.6% 3.1% 6.5% 5.1% 20.3% 13.3% 10.8% 14.7% 1,245
Iceland IS 19.4% 8.1% 3.2% 4.2% .6% 13.4% 7.1% 11.9% 8.6% 23.7% 666
Israel IL 24.9% 5.9% 20.7% 10.5% 2.2% 9.9% 5.5% 7.4% 6.2% 6.8% 1,176
Japan JP 12.7% 14.1% 4.1% 6.1% 2.2% 15.9% 28.0% 5.4% 7.7% 3.8% 1,201
South Korea KR 20.8% 12.3% 5.2% 13.2% .8% 21.6% 10.5% 12.3% 3.4% .1% 1,489
Latvia* LV 15.1% 11.8% 2.0% 14.6% 4.4% 9.5% 6.6% 22.4% 5.2% 8.3% 905
Lithuania* LT 20.0% 13.9% 0.4% 10.7% 3.5% 10.2% 7.3% 28.8% 2.2% 3.0% 935
Mexico MX 17.9% 7.5% 19.0% 20.8% 5.0% 11.5% 12.0% 2.8% 2.3% 1.2% 1,596
Netherlands* NL 30.4% 12.2% 2.5% 4.6% 2.4% 7.9% 10.0% 8.7% 10.1% 11.3% 1,259
(continued)
Table 1.1 (continued)
Air Chemicals Water Water Nuclear Domestic Climate Genetically Using up None of N
pollution and shortage pollution waste waste change modified our natural these
pesticides disposal foods resources
New Zealand NZ 8.0% 16.2% 6.2% 12.9% .6% 17.4% 9.0% 7.8% 7.2% 14.8% 1,068
Norway NO 23.3% 6.3% 4.8% 5.0% 2.2% 11.1% 14.7% 5.4% 8.3% 18.8% 1,216
Philippines PH 17.3% 5.8% 13.8% 8.8% 4.0% 13.8% 20.2% 3.8% 8.2% 4.4% 1,163
Portugal* PT 28.7% 10.1% 7.0% 12.5% 1.8% 8.4% 7.3% 8.1% 5.5% 10.6% 978
Russia RU 18.0% 7.0% 5.2% 20.3% 6.3% 10.3% 10.9% 11.4% 8.0% 2.6% 1,532
Slovak Rep.* SK 21.1% 11.7% 3.2% 8.3% 3.1% 19.6% 8.4% 3.5% 4.2% 16.8% 1,090
Slovenia* SI 18.6% 15.3% 2.7% 14.6% 2.4% 15.0% 7.8% 12.6% 2.6% 8.3% 1,036
South-Africa ZA 17.0% 3.4% 25.0% 17.3% 2.1% 11.4% 8.6% 2.5% 2.8% 9.9% 2,965
Spain* ES 19.2% 8.5% 10.8% 9.7% 5.5% 8.4% 15.4% 9.0% 9.6% 4.0% 2,437
Sweden* SE 22.7% 9.8% 2.1% 12.5% 2.7% 7.0% 13.0% 13.3% 7.3% 9.7% 1,011
Switzerland CH 22.7% 7.5% 2.9% 4.8% 12.5% 2.8% 14.6% 6.5% 10.2% 15.3% 1,161
Taiwan TW 23.4% 16.2% 5.6% 19.2% .2% 8.7% 11.0% 4.9% 5.4% 5.5% 2,197
Turkey TR 14.9% 8.7% 7.0% 10.9% 4.3% 6.1% 8.7% 22.3% 12.1% 4.9% 1,566
United Kingdom* GB 14.0% 4.9% 4.5% 2.7% 1.9% 26.9% 11.0% 5.5% 13.0% 15.6% 783
United States US 24.5% 13.0% 5.8% 10.1% 1.2% 5.3% 6.9% 10.8% 12.3% 10.0% 1,224
*2010 EU members 20.9% 12.4% 4.1% 9.4% 3.4% 11.7% 11.6% 10.3% 7.3% 8.8% 20,119
All countries 19.5% 10.9% 7.7% 11.4% 3.3% 11.5% 11.5% 9.0% 7.2% 8.0% 46,659
University (2014), air quality is indeed excellent in Australia and New Zealand,
and rather poor in the Netherlands and Belgium compared to other European
countries. At the same time, however, the poor air quality in South Korea and the
good air quality in Norway are not mirrored in the threat perception of the general
public in these two countries.
To further scrutinize the relationship between the mentioning of different
threats in different countries, we conducted a correspondence analysis using
the data in Table 1.1. This analysis resulted in three distinct dimensions (see
Figures 1.1 and 1.2 for details). Dimension 1 can be described as ‘water issues
versus all other problems’. Dimension 2 distinguishes between respondents who
said that none of the threats affects them or their families and respondents who
mentioned a threat. Dimension 3 differs between high risks such as nuclear
waste, genetically modified food, using up of resources and climate change and
2.0
1.5
Other problems vs. None of these (Dimension 2)
None of these
IS NO
1.0
CZ GB
CH
SK
DE
0.5 NL BE
AT AU ZA
Climate change Air pollution PT PH IL Water shortage
CL
Using up our natural US NZ
0.0 Nuclear waste
SE CA JP FI ES Domestic waste dispo
BG
DK SI MX
TW
AR
−0.5 Chemicals and pestic TR RU
HR
Genetically modified LV FR KR Water pollution
LT
−1.0
−1.5
−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 2.0
Other problems vs. Water issues (Dimension 1)
Figure 1.1 Dimensions 1 and 2 of environmental threat perception and position of

countries, derived from correspondence analysis based on the variables
depicted in Table 1.1. Dimension 1 accounts for 32% of inertia and Dimension
2 for 20%. Symmetrical Normalization (source: ISSP 2010).
2.0
1.5
Everyday problems vs. High risks (Dimension 3)
1.0 TR
CH Nuclear waste
DE AU
Genetically modified Water shortage
0.5 ES
LT Using up our natural
LV Climate change PH IL
ZA
AT
SE US
HR RU MX
IS
0.0 None of these
CL
DK NL FR NO
JP GB PT Water pollution
CA BG
AR
Chemicals and pestic
Air pollution
NZ FI
−0.5 SI
CZ Domestic waste dispo
SK
KR
TW
BE
−1.0
−1.0 −0.5 0.0 0.5 1.0 1.5 2.0
Other problems vs. Water issues (Dimension 1)
Figure 1.2 Dimensions 1 and 3 of environmental threat perception and position of

countries, derived from a correspondence analysis based on the variables
depicted in Table 1.1. Dimension 1 accounts for 32% of inertia, and
Dimension 3 for 13%. Symmetrical Normalization (source: ISSP 2010).
everyday problems such as air pollution, chemicals and pesticides, and domestic
waste disposal.
The first dimension – water issues versus all other threats – resonates well with
our initial distinction between the natural and social distribution of environmental
good and bad, given that water issues are often related to natural circumstances
such as climate zones, whereas threats such as waste disposal and nuclear
power are more related to human interventions and are thus socially produced.
Furthermore, Dimension 3 distinguishes between different socially produced
problems: everyday problems such as waste disposal versus high risks. Using the
dimensions of Slovic et al.’s (1981) psychometric studies, high risks are the risks
that are unknown, dreadful and potentially affect the entire population, whereas
problems such as domestic waste disposal reflect the opposite pole.
Contextual and individual-level determinants

of threat perceptions
After describing the differences in threat perceptions in the previous section, this
part considers the determinants of these perceptions. For this purpose, the separate
threat items were assigned to four different categories, based on the results of the
correspondence analysis: 1) Water shortage and water pollution become ‘water
issues’; 2) air pollution, chemicals and pesticides and domestic waste disposal
become ‘everyday problems’; 3) nuclear waste, climate change, genetically modi-
fied food, and use of natural resources become ‘high risks’; and 4) respondents
who answered ‘none of these problems’ become the residual group.
Table 1.2 presents the correlations among the country-level prevalence of the
substantive threat categories water issues, everyday problems and high risks and
the societal characteristics, affluence and EU membership. It shows that water
issues are perceived less often in more affluent societies and in EU member
countries. Everyday problems are named more often in EU countries, but are
not associated significantly with affluence, whereas the opposite applies to high
risks. It is interesting to see that high risks, such as genetically modified food and
climate change, are perceived more in more affluent societies, whereas EU mem-
ber countries tend to select everyday problems. This is in line with the idea that
EU has stronger regulation regarding many of these high-risk topics than other
equally affluent countries, and this therefore underscores the idea of the influence
of policymakers.
Table 1.3 shows the results of several multinomial regressions that estimate
the effects of various individual-level characteristics on the selection of the four
different threat categories. The model ‘All countries’ covers the entire, pooled
dataset. We then show the results for three countries that present extreme cases
for each dimension of the correspondence analysis: South Africa as the country
that lies next to the water issues pole; Germany as a country close to the high-
risks pole; and Belgium as a country that is close to the endpoint of the everyday
Table 1.2 Correlations among societal affluence, EU membership and threat perceptions*
Water issues Everyday problems High risks
GDP -0.57** 0.21 0.36*

EU -0.44** 0.39* -0.07
*Pearson Correlations between country-level proportion of respondents selecting ‘water issues’,
‘everyday problems’, or ‘high risks’ and country-level GDP and EU Membership. Taiwan excluded
as GDP not available at World Bank (2015). Significance: *0.05; **0.01.
Water Issues = Water shortage and Water pollution; Everyday problems = Air pollution, Chemicals
and pesticides, and Domestic waste disposal; High risks = Nuclear waste, Climate change, Genetically
modified foods, and Using up our natural resources. Association with different dimensions based on
dimension score and contribution to inertia of dimension in correspondence analysis.
Source: ISSP 2010
Table 1.3 Individual-level determinants of environmental threat perceptions*
Water issues Everyday problems High risks
B Sig. Exp. B Sig. Exp. B Sig. Exp.
All Intercept -0.74 0.00 -1.01 0.00 -1.13 0.00

countries Female 0.18 0.00 10.19 0.22 0.00 10.24 0.16 0.00 1.17
Age 0.00 0.00 10.00 0.00 0.00 10.00 0.00 0.53 1.00
Education -0.01 0.18 0.99 0.02 0.01 10.02 0.02 0.00 1.02
Subj. position -0.06 0.00 0.94 -0.04 0.00 0.96 -0.04 0.00 0.96
Partner 0.16 0.00 10.18 0.18 0.00 10.20 0.14 0.00 1.15
Children 0.08 0.15 10.08 0.10 0.04 10.10 0.04 0.38 1.04
Residency -0.01 0.76 0.99 0.15 0.00 10.17 0.10 0.00 1.11
Env. efficacy 0.34 0.00 10.41 0.37 0.00 10.45 0.46 0.00 1.58
Germany Intercept -2.41 0.08 0.08 0.92 0.63 0.37
Female 0.72 0.03 2.05 0.16 0.42 1.17 -0.02 0.91 0.98
Age 0.01 0.64 1.01 0.00 0.87 1.00 -0.01 0.23 0.99
Education -0.06 0.30 0.95 -0.02 0.44 0.98 -0.02 0.38 0.98
Subj. position 0.14 0.20 10.15 -0.03 0.61 0.97 0.02 0.72 1.02
Partner 0.50 0.17 1.65 0.28 0.20 1.32 0.20 0.32 1.22
Children 0.23 0.59 1.26 0.09 0.73 1.09 0.08 0.73 1.08
Residency -0.30 0.06 0.74 -0.04 0.65 0.96 -0.11 0.19 0.90
Env. efficacy 0.06 0.78 1.06 0.28 0.02 1.33 0.44 0.00 1.55
Belgium Intercept -2.66 0.05 -1.85 0.08 -1.01 0.39
Female 0.14 0.63 1.15 0.10 0.67 1.10 -0.04 0.89 0.96
Age 0.03 0.00 1.04 0.04 0.00 1.04 0.03 0.00 1.03
Education 0.04 0.46 1.04 0.10 0.02 1.11 0.04 0.36 1.04
Subj. position 0.02 0.86 1.02 -0.11 0.18 0.89 -0.17 0.08 0.84
Partner 0.15 0.68 1.16 -0.16 0.57 0.85 -0.19 0.55 0.83
Children 0.09 0.81 1.10 0.41 0.15 1.51 0.10 0.76 1.10
Residency -0.16 0.26 0.85 -0.09 0.44 0.92 -0.21 0.09 0.81
Env. efficacy 0.22 0.33 1.25 0.52 0.00 1.68 0.57 0.01 1.77
South Intercept 1.65 0.00 0.82 0.09 -1.15 0.04
Africa Female 0.16 0.26 1.17 -0.07 0.61 0.93 -0.17 0.29 0.85
Age -0.01 0.21 0.99 -0.01 0.24 0.99 -0.01 0.03 0.99
Education 0.03 0.10 1.03 0.04 0.06 1.04 0.05 0.04 1.05
Subj. position -0.11 0.00 0.89 -0.03 0.47 0.97 -0.01 0.79 0.99
Partner 0.18 0.21 1.20 0.28 0.06 1.33 0.40 0.02 1.49
Children -0.11 0.45 0.90 -0.17 0.25 0.84 -0.15 0.36 0.86
Residency -0.23 0.00 0.79 0.02 0.72 1.02 0.16 0.01 1.17
Env. efficacy 0.31 0.00 1.37 0.11 0.26 1.12 0.39 0.00 1.48
*Multinomial logistic regression with ‘none of these problems’ as reference category. Also included in ‘All countries’ model are fixed effects for each country.
Valid cases: 41,122 in all countries; 1,140 in Germany; 1,013 in Belgium; 2,847 in South Africa.
Source: ISSP 2010.
problems. Respondents who answered that none these problems affects them or
their family are used as a reference dimension.
In the entire sample ‘All countries’ water issues are mentioned more often
by women, by individuals who see themselves in lower social positions, and
by individuals with a partner and with a strong environmental self-efficacy.
Everyday problems are named by women, older individuals, better educated
respondents, individuals who classify themselves as socially lower, respond-
ents with a partner and children, as well as urban dwellers and individuals
with a strong self-efficacy. High risks, finally, are perceived more often by
women and respondents who are better educated, see themselves in lower
social positions, have a partner, live in a city and have a strong environmental
self-efficacy. All of these effects need to be interpreted with regard to the ref-
erence dimension of ‘no risk selected’. Therefore, we can summarize that it is
women, respondents with a partner, in subjective lower social positions, and
older individuals that tend to mention a threat. In addition, we observe further
differences as urban dwellers and educated individuals tend to select the two
socially produced issues, while less well educated respondents and rural dwell-
ers choose the water problems.
Consider the effects in our three selected countries. In Germany, water
issues are named by women, while everyday problems, as well as high risks, are
selected by individuals with a strong environmental self-efficacy. In Belgium,
water issues are selected by older respondents; everyday problems are chosen
by older respondents, as well as better educated individuals and those with a
strong self-efficacy; and high risks are named by older respondents and indi-
viduals with a strong self-efficacy. In South Africa, water issues are mentioned
by individuals in (subjectively) lower social positions, rural dwellers and
respondents with a strong self-efficacy; everyday problems are selected evenly
by all respondents; and high risks by respondents who are younger and better
educated, live in urban areas, have a partner and are characterized by a strong
environmental self-efficacy.
Overall, the largest number of significant effects can be found in South Africa,
with the fewest effects in Germany. Only the effects of environmental self-efficacy
regarding the perception of high risks are consistent across all countries, which
points towards the importance of agency. As for the social-demographic vari-
ables, only a few had an impact in the European countries, Germany and Belgium.
This homogeneity suggests that the distribution of environmental good and bad
is more similar in these two European countries than in South Africa, possibly as
a result of better monetary resources and regulations to address environmental
problems. However, considering the explanatory power of these four different
regressions (explained variance, measured by the pseudo R-square Nagelkerke, is
15.3% for the entire sample; 4.3% for Germany; 6.5% for Belgium; and 9.4% for
South Africa), we were able to explain a higher percentage of the variation in the
total sample; a finding that indicates that national characteristics are of substantial
importance in the perception of different threats.
Discussion and conclusion

After presenting and summarizing our results, we can now turn to the implica-
tions of these findings for the different theories and hypotheses discussed at the
beginning of this chapter. We started from the assertion that the perception of
environmental threats is firstly influenced by the presence of different threats,
which are related to geographical differences and socially induced impacts. This
first hypothesis is clearly supported at the country level, as far as natural and geo-
graphic differences are concerned. A correspondence analysis of perceived threats
in 36 countries identified a ‘water issues’ dimension that is related to geographi-
cal characteristics, such as climate zone and water availability, with South Africa
being the country in which this threat is perceived the most often.
Our results were more ambiguous regarding the differences in the presence of
threats owing to the social organization of production and societies in general,
as well as differences in the way such threats are addressed. Climate change, for
example, was perceived very differently across countries, despite being a global
threat. As far as everyday issues such as air pollution are concerned, we saw a
match between the presence of pollutants and individual views in some coun-
tries. The association between the objective levels of pollution and the perceived
threats, however, was far from perfect. A possible explanation for this discrep-
ancy is related to the way the threat question was asked in the ISSP module.
Respondents had to name the most important threat for themselves and their fam-
ily, and thus did not assess the severity of each threat. Consequently, if there is a
particularly pressing issue – such as water shortage, for example – other threats
become less salient and may not be mentioned, even if the objective level of
threat is high.
Despite these limitations, we were able to find an association between the three
categories of threat – ‘water issues’, ‘everyday problems’ and ‘high risks’ – and
the societal characteristics, affluence and membership of the EU. Water issues
were selected less often in wealthier societies and by EU members, which res-
onates with the ecological modernization idea that a certain level of wealth is
necessary to address basic environmental threats. Furthermore, societal affluence
and EU membership had different effects on everyday issues and high risks, with
the former being mentioned more often by EU members and the latter by affluent
societies. This second difference suggests that common environmental and other
regulations may also have an impact on the distribution of environmental threats
and their perception.
As for the individual determinants of the perceptions of these threats, agency
measured as environmental self-efficacy turned out to be particularly important.
First, it is decisive with regard to who perceives threats at all, as it had posi-
tive effects on all three threat categories in almost all countries, and thus clearly
separates respondents, who recognize, from respondents who do not perceive any
threats. Within the different threats, efficacy was important for the perception of
socially induced threats, particularly for high risks, which include nuclear power
and climate change. Considering this environmental self-efficacy as a specific

type of environmental attitude, our results thus resonate with international com-
parative research on environmental attitudes and concerns in general, which also
point to a connection between values, perceptions and attitudes (Fairbrother 2013;
Franzen & Meyer 2004; Haller & Hadler 2008).
Socio-demographic characteristics, such as age, gender, education and others,
had quite diverse effects on the different threats. A possible explanation for these
differences in the explanatory power of various socio-demographic variables is

that negative environmental effects impact individuals in very indirect ways (see
Kraemer 2008: 216ff). Negative impacts in modern industrial societies are often
only recognized when they are already visible and – equally important – also com-
municated. In addition, it is difficult to assess different levels of environmental
threats and to establish a critical level. Given that our analysis is not able to grasp
all these detailed aspects, we can therefore only suggest that the perception of
threats is also influenced by media, politicians, environmental organizations and
other social forces that influence public opinion, as discussed by Keller & Poferl
(1998) and Seifert (2008).
Finally, what can be said regarding the democratization of different environ-
mental threats, as posited by Beck (1986)? Our findings show that the perception
of threats can be influenced by geographical factors such as climate, societal
factors such as affluence and political regimes, as well as various individual
characteristics. For the role of natural and socially induced threats, our analysis
suggests that an increasing societal affluence allows us to address basic environ-
mental threats such as water shortage. The salience of different socially induced
threats, on the other hand, seems to be related to political measures at the county
level. The democratization of threats is thus not a given, but dependent on specific
geographical, social and political circumstances.
Notes
1 We are limiting our general hypotheses to the aspects that can be grasped with the ISSP
data. Bickerstaff (2004), for example, names place attachment, agency and trust in insti-
tutions as potential determinants of risk perceptions. Only agency and, partially, trust in
institutions are captured in ISSP items on environmental efficacy.
2 Based on Table 1.1, the standard deviations for all ten threats were calculated for EU
countries and non-EU countries, and then the average of all standard deviations was
calculated (4.8 for non-EU, and 4.3 for EU countries).
References
Arnalds, ÁA & Sigurbjörns-Öldudóttir, S. 2014, International Social Survey Programme:
Environment III - ISSP 2010 (Iceland). GESIS Data Archive, Cologne. ZA5515 Data
file Version 1.0.0, doi:10.4232/1.11796.
Beck, U. 1986, Risikogesellschaft. Auf dem Weg in eine andere Moderne. Suhrkamp:
Frankfurt a. M.
Bickerstaff, K. 2004, ‘Risk Perception Research: Socio-cultural Perspectives on the Public

Experience of Air Pollution’, Environment International, no. 6, pp. 827–840.
Bolte, G & Mielck, A. 2004, Umweltgerechtigkeit. Die soziale Verteilung von
Umweltbelastungen. Juventa: Weinheim.
Dake, K. 1991, ‘Orienting Dispositions in the Perception of Risk: An Analysis of
Contemporary Worldviews and Cultural Biases’, Journal of Cross-Cultural Psychology,
no. 1, pp. 61–82.
Douglas, M 1992, Risk and Blame: Essays in Cultural Theory. Routledge: London.
Douglas, M & Wildavsky, A. 1982, Risk and Culture: An Essay in the Selection and
Interpretation of Technological and Environmental Dangers. University of California
Press: Berkeley.
Elvers, HD, Gross, M & Heinrichs, H. 2008, ‘The Diversity of Environmental Justice:
Towards a European Approach’, European Societies, no. 5, pp. 835–856.
Evans, A. 2014, International Social Survey Programme: Environment III - ISSP 2010
(Australia). GESIS Data Archive, Cologne. ZA5512 Data file Version 1.0.0,
Fairbrother, M. 2013, ‘Rich People, Poor People, and Environmental Concern: Evidence
across Nations and Time’, European Sociological Review, no. 3, pp. 910–922.
Franzen, A & Meyer, R. 2004, ‘Klimawandel des Umweltbewusstseins? Eine Analyse des
ISSP 2000’, Zeitschrift für Soziologie, no. 2, pp. 119–137.
Freeman, AM. 1974, ‘On Estimating Air Pollution Control Benefits from Land Value
Studies’, Journal of Environmental Economics and Management, no. 1, pp. 74–83.
Ganzeboom, H. 2014, International Social Survey Programme: Environment III - ISSP
2010 (Netherlands). GESIS Data Archive, Cologne. ZA5513 Data file Version 1.0.0,
doi:10.4232/1.11792
Grant, D, Trautner, NM, Downey, L & Thiebaud, L. 2010, ‘Bringing the Polluters Back
In: Environmental Inequality and the Organization of Chemical Production’, American
Sociological Review, no. 4, pp. 479–504.
Hadler, M & Haller, M. 2011, ‘Globally Motivated Activists and Nationally Driven
Recyclers. The Influence of Global Linkages and National Contexts on Individual
Environmental Behavior’, International Sociology, no. 3, pp. 315–345.
Haller, M & Hadler, M. 2008, ‘Dispositions to Act in Favor of the Environment. Fatalism
and Readiness to Make Sacrifices in Cross National Perspective’, Sociological Forum,
no. 2, pp. 282–311.
Harper, C. 2012, Environment and Society. Pearson: Boston.
Hix, S & Høyland, B. 2011, The Political System of the European Union. Palgrave
Macmillan: Houndmills.
ISSP Research Group. 2012, International Social Survey Programme: Environment III - ISSP
2010. GESIS Data Archive, Cologne. ZA5500 Data file Version 2.0.0, doi:10.4232/1.11418
Keller, C, Bostrom, A, Kuttschreuter, M, Savadori, L, Spence, A & White, M. 2012,
‘Bringing Appraisal Theory to Environmental Risk Perception: a Review of Conceptual
Approaches of the Past 40 Years and Suggestions for Future Research’, Journal of Risk
Research, no. 3, pp. 237–256.
Keller, R & Poferl, A. 1998, ‘Vergesellschaftete Natur – Öffentliche Diskurse und sozi-
ale Strukturierung. Eine kritische Auseinandersetzung mit der Cultural Theory,’ in
Soziologie und Natur. Theoretische Perspektiven, ed. KW Brand, Leske & Budrich:
Opladen, pp. 117–142.
Kraemer, K. 2008, Die soziale Konstitution der Umwelt. VS Verlag für Sozialwissens
chaften: Wiesbaden.
Kraemer, K. 2011, ‘Prekärer Wohlstand und nachhaltiger Konsum’, Österreichische

Zeitschrift für Soziologie, no. 2, 35–54.
Kraemer, K. 2014, ‘Ist Prekarität nachhaltig? Nachhaltiger Konsum und die Krise des
wohlfahrts-staatlichen Kapitalismus’, in Klima von unten. Regionale Governance und
gesellschaftlicher Wandel, ed. S Böschen et al., Frankfurt a. M., pp. 169–185.
Lichter, D & Brown, D. 2011, ‘Rural America in an Urban Society: Changing Spatial and
Social Boundaries’, Annual Review of Sociology, pp. 565–592.
Lutzenhiser, L & Hackett, B. 1993, ‘Social Stratification and Environmental Degradation:

Understanding Household CO2 Production’, Social Forces, no. 1, pp. 50–73.
Schwarz, M & Thompson, M. 1990, Divided We Stand: Redefining Politics, Technology
and Social Choice. University of Pennsylvania Press: Philadelphia.
Seifert, F. 2008, ‘Einhellige Abschottung und globalisierungskritische Bauernrevolte –
Widerstand gegen die grüne Gentechnik in Österreich und Frankreich’, SWS-Rundschau,
no. 4, pp. 485–504.
Slimak M & Dietz, T. 2006, ‘Personal Values, Beliefs, and Ecological Risk Perception’,
Risk Analysis, no. 6, pp. 1689–1705.
Slovic, P. 1987, ‘Perception of Risk’, Science, no. 4799, pp. 280–285.
Slovic, P, Fischhoff, B & Lichtenstein, S. 1981, ‘Facts and Fears: Societal Perception of
Risk’, in Societal Risk Assessment. General Motors Research Laboratories, ed. RC
Schwing, & W Albers, Springer, pp. 181–216.
Stern, P. 2000, ‘Toward a Coherent Theory of Environmentally Significant Behavior’,
Journal of Social Issues, no. 3, pp. 407–424.
Thompson, M, Ellis, R, Wildavsky. A, & Trustee Wildsavsky, M. 1990, Cultural Theory,
Westview Press: Boulder.
Vala, J & Ramos, A. 2014, International Social Survey Programme: Environment III -:
ISSP 2010 (Portugal). GESIS Data Archive, Cologne. ZA5516 Data file Version 1.0.0,
doi:10.4232/1.11793.
Wilkinson, I. 2001, ‘Social Theories of Risk Perception: At Once Indispensable and
Insufficient’, Current Sociology, no. 1, pp. 1–22.
World Bank. 2014a, World Bank Open Data, available at: http://data.worldbank.org/indi
cator/ER.H2O.FWTL.ZS, accessed: 15 Aug. 2014.
World Bank. 2014b, World Bank Open Data, available at: http://data.worldbank.org/indi
cator/SH.H2O.SAFE.RU.ZS, accessed: 15 Aug. 2014.
World Bank. 2015, World Bank Open Data, available at: http://data.worldbank.org/indica
tor/NY.GDP.MKTP.CD, accessed: 9 Oct. 2015.
Yale University. 2014, Environmental Performance Index: Air Quality, available at: http://
epi.yale.edu/epi/issue-ranking/air-quality, accessed: 19 Sept. 2014.
York, R, Eugene, R & Dietz, T. 2003, ‘Footprints on the Earth: The Environmental
Consequences of Modernity’, American Sociological Review, no. 2, pp. 279–300.
Chapter 2
Vulnerable and insecure?

Environmental and technological
risk perception in Europe
Aistė Balžekienė and Audronė Telešienė
Introduction
The classical definition provided by Paul Slovic in the journal Science in 1987
states that ‘risk perception is intuitive risk judgement’ (Slovic, 1987:p.236).
This definition was extended by him and his colleagues in 2004 (Slovic et al.,
2004), referring to risk in three fundamental ways: risk as feelings; risk as
analysis; and risk as politics. These dimensions of risk perception are espe-
cially relevant when talking about environmental and technological risks, and
responses to them. Individuals tend to evaluate risks based on intuitive feel-
ings; experts provide analytical knowledge about risks based on statistics, and
the decision-making process about technological development refers to risk
as politics. The term ‘risk perception’ is defined as a ‘process of collecting,
selecting and interpreting signals about uncertain impacts of events, activities,
or technologies’ (Wachinger et al., 2012:p.1049).
Risk-perception empirical research has benefited largely from the international
comparative surveys that provide the possibility to analyse reliable data across
nations. One such survey is the International Social Survey (ISSP), which has
already implemented three waves of the module ‘Environment’ (1993, 2000 and
2010), providing the opportunity to track the changes in the attitudes, perceptions,
values and behaviours at the individual level and across nations. ISSP Environment
data of 1993 and 2000 was explored extensively in academic literature by applying
various theoretical concepts and searching for complex explanatory models. Most
of these models focus on environmental attitudes (e.g., Frizzell & Pammett, 1997;
Franzen, 2003; Franzen & Meyer, 2010); environmental concern (e.g., Olofsson &
Öhman, 2006; Marquart-Pyatt, 2007) or environmental behaviours (e.g., Engel &
Pötschke, 1998; Hayes, 2001; Hunter et al., 2004; Hadler & Haller, 2011). Risk
perception as a focus for investigation based on ISSP data has been analysed less
frequently (Lima et al., 2005; Adeola, 2004).
In recent decades, there have been many efforts in sociological literature to
search for the theoretical models to explain environmental and technological
risks. Wachinger et al. (2012) reviewed the literature on risk perception and con-
cluded that risk perceptions differ depending on the type of risk, the risk context,
32 Aistė Balžekienė and Audronė Telešienė
the personality of the individual and the social context. This chapter aims to
explore these dimensions of environmental and technological risk perception in
Europe based on ISSP data. We examine the risk perception of different environ-
mental and technological issues, then we explore risk perception at the individual
level. We also examine structural and cognitive factors, and the influence of trust
and environmental concern on risk perception. Social context in our research is
defined by a country-level analysis, revealing the differences in risk perception
levels and patterns across countries in Europe.

This chapter seeks to: (1) discuss the differences in environmental and tech-
nological risk perceptions across European countries; (2) explore if and how risk
perception at the individual level is determined by structural and cognitive factors,
environmental concern and trust; and (3) reveal the influence of macrolevel factors
on the perception of environmental and technological risks at the country level.
We have selected European countries for the analysis of environmental and
technological risk perception. We are not limiting our analysis to European Union
countries, as we think perceptions of environmental issues exceed the boundaries
of political unity and, rather, are defined by historical and cultural contexts.
The main research questions addressed in this chapter are: How does environ-
mental and technological risk perception differ across European countries? What
are the main structural and cognitive factors that determine risk perception varia-
tion at the individual level? What macrolevel factors are significant for explaining
risk perception?
Determinants of risk perception

Risk-perception research already spans several decades, thus, recently, in aca-
demic literature, there was an effort to conduct a meta-analysis of risk research
and to summarize the determinant and explanatory models that were used by
various researchers. Keller et al. (2012) provide a comprehensive review of the
literature on risk perception, differentiating between cognitive and affective
risk-perception determinants. Renn and Benighaus (2013) identify psychologi-
cal, social and cultural factors that shape individual risk perceptions. Many
studies integrate the predictors of risk perception at the individual and mac-
rolevels (e.g. Viklund, 2003; Kim et al., 2014). In this chapter, we explore
determinants of risk perception at individual and macrolevels.
Individual-level factors
This analysis starts by defining socio-demographic characteristics such as age,
gender, place of living and self-social-class placement. We call these variables
‘structural determinants.’ As Henwood et al. state:
Social identity positions associated with categories such as age and gender
are key contributors to the ways in which people conceptualize and give
Environmental and technological risk perception 33
meaning to human problems and their solutions. This is certainly the case in
relation to issues of science, technology and risk.
(Henwood et al., 2008:p.662f)
Out of a list of socio-demographic variables, gender is documented by many studies

as significantly influencing risk perceptions (Henwood et al., 2008). Gender is the
most constant predictor of risk perception, as others are not always significant. Men
always tend to express lower levels of concern than women with regard to the envi-
ronment or technology (ibid.)
Lima et al. (2005) conceptualize environmental awareness as a mediator of
the relationships between risk perception and technological diffusion. They argue
that a lower level of environmental awareness is associated with social sensitiv-
ity towards risks. Awareness is closely related to knowledge levels. Keller et al.
(2012) note that, for many environmental hazards, people may lack sufficient
knowledge to make an informed decision about risks, or may simply be unaware
of the issues before forming an opinion.
Therefore, we include cognitive factors of risk perception in our explanatory
model, based mainly on the level of subjective knowledge about environmental
issues. Urban and Hoban (1997) report the significant effect of cognitive-level
factors (education and information) on the perception of risks of biotechnologies.
They use a cognitive causal model for the explanation of risk perception. Some
researchers see knowledge as one of the main factors that can influence risk per-
ception (Kim et al., 2014).
Beck (1992) argues that decreasing trust in science is one of the fundamental
features of a risk society, and allows individuals to cope with the uncertainties
of technological development, increasing the reflexivity of society, especially
towards the possible impacts of new technologies. Many studies show (e.g.,
Gupta et al, 2011; Siegrist and Cvetkovich, 2000) that trust is a very powerful fac-
tor determining environmental and technological risk perceptions. The distrust in
science leads to negative evaluations of technologies and a possible overestima-
tion of risk. Gupta et al. (2011) examined the socio-psychological determinants
of public acceptance of technologies and showed that trust and knowledge are
among the most widely used determinants.
In our operational model to be tested using ISSP data, we include structural var-
iables of risk perception, trust, subjective knowledge (as a cognitive dimension)
and environmental concern.
Contextual (macrolevel factors)

Understanding wider contexts is crucial for a more comprehensive explanation
of environmental and technological risk perceptions. Countries participating in
the ISSP Environment module (also European countries) have divergent levels
of environmental degradation, various levels of economic development, and dif-
ferent sociocultural and political settings. Individuals are exposed to different
contexts that, working through processes of socialization, acculturation, social

learning, etc., shape individual mindset.
Existing research provides evidence for including economic indicators, such as
gross domestic product per capita and economic growth rate, in the analysis. The
economic situation of a country is usually under close scrutiny from the public.
Economic worries are said to compete with environmental concerns and tend to
become dominant during economic crises. Bearing in mind the economic crisis
that began in Europe in late 2008, and lasted through 2010 and 2011 when the
surveys were conducted, there is a sound logic to analyse the relation between the
economic indicators of a country and the average environmental and technological
risk perception. Gross domestic product is more related to the general economic
situation or the affluence characteristic of the country, and the economic growth
rate is related to the experienced crisis at the time of the conducted surveys.
The socio-political contexts of the countries are defined by the Democracy
Index, as determined by the Economist Intelligence Unit, and the Human
Development Index (HDI), measured by the United Nations. The HDI is a sum-
mary measure of average country achievement in key dimensions of human
development: a long and healthy life, being knowledgeable, and having a decent
standard of living (UNDP, 2014). HDI is composed of educational, societal-health
and economic indicators. In order to avoid overlap with the gross domestic prod-
uct indicator described above, the value of non-income HDI is used. Non-income
HDI is computed without economic indicators, from only educational and health
indicators. The Democracy Index provides a snapshot of the state of democracy
in a country. The Economist Intelligence Unit’s Index of Democracy is based
on five categories: electoral process and pluralism; civil liberties; the function-
ing of the government; political participation; and political culture (Economist
Intelligence Unit, 2010).
Another indicator potentially influencing risk perceptions on a macro-to-
microlevel is the pace of societal transformation. This is related to the scope and
speed of social change in a given society. This factor is rarely used in environmen-
tal sociology and thus needs more justification.
Appelbaum (1970) and Smelser (1968), in their classic works, propose that
social change is one of the most powerful factors, having a specific influence
on the collective consciousness of societies. It should be emphasized that the
pace of change is different in contemporary societies compared to earlier types
of societies. Furthermore, rapid change, or the fast pace of change, creates cul-
tural lags between infrastructures and processes in an objective culture on one
side (see Simmel, 1968), while in the subjective culture, the value systems and
norms that have been formed within individual psyches throughout lifetime
experiences create cultural lags on the other side. This individual experience
is essentially a historically embedded experience of previous stages of societal
development; i.e. structures and processes before the occurrence of a specific
instance of change. The thesis of cultural lag comes from equilibrium theory and
describes a situation where one part of a social system is changing faster than
others (Brinkman and Brinkman, 1997). Then, a gap or a so-called cultural lag
occurs; e.g. legislature is changing faster than social norms. Rapid social trans-
formations create cultural lags that, in turn, create dissonance between personal
value and norm systems and external cultural realities, thus creating stress and
raising sensitivity to any kind of risk (risk perceptions tend to be higher).
The Marxist notion of social transformations argues that the structures and
relations of production are faster to change than the ideological structures in a
society. If changes in ideological structures lag behind the changes in forces of

production and changes in productive relations, economic crises commence, lead-
ing to social unrest (Sztompka, 1993:p.172).
Indeed, the situation of rapid societal transformations, whose influence on
risk perception we have explained through the notion of cultural lags, must be
understood as a factor defining macrolevel vulnerability. National resilience and
the ability to absorb and adapt to change has a mitigating impact, which is seen
in strong democracies and affluent welfare states, and is lacking in post-Soviet,
economically unstable countries, or countries with weak/young democracies.
The pace of societal transformations in this chapter is measured by the HDI
derivative indicator of average annual HDI growth, as computed in Human
Development Global Reports.
Together with socio-economic data, such as GDP per capita, the pace of devel-
opment, democracy level, etc., environmental and technological risk perceptions
are related hypothetically to the objective conditions of the natural environment
in which individuals are living. For this measure, we look at Environmental
Performance Index (EPI) scores for 2010, and take the general score for the EPI
and the constituent indices of Environmental Health and Ecosystem Vitality.
These measures describe the national-level environmental data. Environmental
Health measures the protection of human health from environmental harm and
includes indicators for health impacts, air quality, access to drinking water and
sanitation. Ecosystem Vitality measures ecosystem protection and resource man-
agement, and includes indicators regarding water resources, agriculture, forests,
fisheries, biodiversity and habitat, climate and energy (Emerson et al., 2010).
The logic of combining the above shortly described macrolevel factors into
one analytical framework follows the challenge-response explanatory model
(Silbereisen and Chen, 2010). The challenge-response model would argue that
risk perception variations occur as a response to variations in external contexts,
and as a response to external pressures upon an individual, either in a form of eco-
nomic crisis, fast societal transformation or evident environmental degradation.
Data and methods

The analysis is based on data from European countries from the ISSP Environment
module, conducted in 2010 (ISSP Research Group, 2012). Nineteen countries
are included in the analysis: Austria (n = 1,019), Belgium (n = 1,142), Bulgaria
(n = 1,003), Croatia (n = 1,210), Czech Republic (n = 1,428), Denmark (1,305),
Finland (n = 1,211), France (n = 2,253), Germany (n = 1,407), Latvia (n = 1,000),

Lithuania (1,023), Norway (n = 1,382), Russia (n = 1,619), Slovak Republic
(n = 1,159), Slovenia (n = 1,082), Spain (n = 2,560), Sweden (n = 1,181),
Switzerland (n = 1,212) and the United Kingdom (n = 928). The total sample
includes 25,124 respondents.
Two analytical models have been constructed to: (1) test the influence of struc-
tural, cognitive factors, trust and environmental awareness on risk perception at
the individual level (using the total sample); and (2) explore the significance of
macrolevel variables at the national level (using countries as a unit of analysis).
The dependent variable for both models is risk perception, drawn from the ISSP
data. For the individual-level analysis, independent variables are also drawn from
ISSP data, and for the macrolevel analysis, we use country-level data taken from
different sources that are presented in more detail later in this section.
The structure of the dependent variable

The operationalization of risk perception in the ISSP Environment questionnaire
is not straightforward. The question that we use for the dependent variable is: ‘in
general, do you think that <problem> is . . . extremely dangerous to the environ-
ment, very dangerous, somewhat dangerous, not very dangerous or not dangerous
at all for the environment?’ We use seven items measuring the perception of envi-
ronmental and technological risks that are provided in the ISSP questionnaire: air
pollution caused by cars; air pollution caused by industry; pesticides and chemi-
cals used in farming; pollution of the country’s rivers, lakes and streams; a rise in
the world’s temperature caused by climate change; genetic modification of certain
crops; and nuclear power stations.
The analysis is limited to the seven risk perception items available in the ISSP
survey. These items are more focused on environmental risks than on technologi-
cal issues. Therefore, the scope of risks covered by the ISSP is not exhaustive.
Even so, these items of risk perception form an internally consistent index
(Cronbach alpha = 0.806, seven items). Thus, for regression models, we will use
the derivative index of risk perception as a main independent variable.
Risk perception in this chapter is addressed in three ways:
1 At the individual level: the perception of threats from various environmental/

technological issues (seven items).
2 At the individual level: the derivative risk perception index (aggregated mean
score of seven items).
3 At the country level (the macrolevel analysis): mean scores of the risk per-
ception index for each country.
In the ISSP questionnaire, risk perception items are coded as follows: 1. ‘extremely
dangerous for the environment’ . . . 5. ‘not dangerous at all for the environment’.
We think that the results of the data analysis are more comprehensible if higher
values equate to a higher level of risk. Therefore, we take all risk perception
items in a reversed manner and recode them where ‘1’ means not dangerous at
all for the environment and ‘5’ means extremely dangerous for the environment.
From the reverse-coded risk-perception items, we calculate the risk-perception
index (a mean score of the seven risk-perception items), where ‘1’ also means
‘not dangerous at all for the environment’ and ‘5’ means ‘extremely dangerous
for the environment’.
Individual-level independent variables

The set of independent variables for the individual-level analysis is composed
of structural factors, cognitive determinants (subjective knowledge), trust and
environmental concern. All of these factors are approved by various studies in
different risk contexts as having a significant impact on risk perception.
Structural factors include:
•• Gender.
•• Age: for regression models we use the original age variable, and for the com-
parison of risk perception, we calculate age groups.
•• Education (Q: ‘How many full years of schooling or education have you had?’).
•• Self-social class placement (Q: ‘In our society there are groups which tend
to be towards the top and groups which tend to be towards the bottom. Below
is a scale that runs from the top to bottom. Where would you put yourself on
this scale?’ 1. Lowest . . . 10. Highest).
•• Living place (Q: ‘Would you describe the place where you live as . . . 1.A big
city . . . 5. A farm or home in the country?’).
All these variables form a description of the social structure that defines the indi-
vidual and that is more or less constant or stable. We include these structural
variables in regression model 1.
Other items for the analysis describe aspects that are more dynamic and that
can change or be shaped over time. They include trust items, environmental
concern and cognitive aspects, such as knowledge level. The influence of these
variables upon risk perception is tested in regression model 2.
Trust in our analysis is measured by three items:
•• Trust in science (Q: ‘Modern science will solve our environmental problems
with little change to our way of life’. Values from 1. ‘strongly agree’ to 5.
‘strongly disagree’).
•• Social trust (Q: ‘Would you say that most people can be trusted, or that you
can’t be too careful in dealing with people?’ Values from 1. ‘ strongly agree’
to 5. ‘strongly disagree’).
•• Political trust (Q: ‘Most of the time we can trust people in government to do
what is right.’ Values from 1. ‘ strongly agree’ to 5. ‘strongly disagree’ ).
The cognitive dimension includes two items measuring subjective knowledge:
•• Causes of sorts of environmental problems (Q: ‘How much do you feel you
know about the causes of these sorts of environmental problems?’ Values
from 1. ‘know nothing at all’ to 5. ‘know a great deal’).
•• Solutions to environmental problems (Q: ‘How much do you feel you know
about solutions to these sorts of environmental problems?’ Values from 1.
‘know nothing at all’ to 5. ‘know a great deal’).
Finally, environmental concern is measured by a single item (Q: ‘How concerned

are you about environmental issues?’ Values from 1. ‘not at all concerned’ to 5.
‘very concerned’).
Macrolevel (contextual) independent variables

For the macrolevel analysis, the dependent variable is the risk-perception
index (the mean score of risk-perception items). As we indicated in the section
‘Determinants of risk perception’, the main macro-determinants of risk percep-
tion at the country level are economic situation, the developmental level of a
country, democracy index, the pace of societal development and conditions of
the environment.
The economic situation, or the affluence of a country, is measured by gross
domestic product (GDP) per capita (current US$) in mid-2010, as taken from
the World Development Indicators database.1 Economic growth rate is meas-
ured by GDP annual growth (%) in 2010 compared to 2009, as taken from the
World Development Indicators database.2 The developmental level of a country
is measured by the non-income HDI value for 2010, as taken from the Human
Development Report (UNDP 2010). The Democracy Index is taken from a report
from the Economist Intelligence Unit (2010). The pace of societal transformation
is measured by average annual HDI growth (%) from 2000–2010, as taken from
the Human Development Report (UNDP 2010). Conditions of the environment
are measured using the EPI and its constituent parts of Environmental Health and
Ecosystem Vitality (Emerson et al., 2010). The summary of macrolevel determi-
nants is presented in Table 2.1.
We use multiple linear regression models to test the factors influencing
risk perception at the individual level. These models use the total sample of
European countries (n = 25,124). For the analysis of the relations between mac-
rolevel indicators and risk perception, we use linear regression at the country
level (number of countries, n = 19). The dependent variable in this analysis is the
risk-perception index (mean scores at the country level), and the set of predictors
is from secondary sources.
Where applicable, we use non-parametric tests for the significance of mean
differences and correlation analyses. Data are weighted using sample weights
provided in the dataset.
Table 2.1 List of macrolevel measures, indicators and data sources.
Measure Indicator Data source
Economic GDP per capita (current US$) in World Development

situation midyear 2010 Indicators database
GDP annual growth (%) in 2010
compared to 2009
Socio-political Non-income HDI value for 2010 Human Development

contexts Report, 2010
Democracy index from 2010 Economist Intelligence
Unit, 2010
Pace of societal Average annual HDI growth (%) Human Development
transformations from 2000–2010 Report, 2010
Objective EPI scores for 2010 Emerson et al., 2010;
conditions of EPI: Environmental Health scores Yale Center for
the natural for 2010 Environmental Law and
environment Policy
EPI: Ecosystem Vitality scores
for 2010
Note: GDP: Gross Domestic Product; HDI: Human Development Index; EPI: Environmental
Performance Index
Further, we present results that are structured according to the research ques-
tions, starting with the risk-perception pattern across European countries. We then
discuss the perception of different types of risks, followed by an analysis of risk-
perception determinants at the individual level, and macrolevel determinants at
the country level.
Results
Risk perception in European countries
There were significant differences in environmental and technological risk per-
ception across countries in Europe (Figure 2.1). The results suggested that
risk-perception patterns were dependent on cultural and national contexts, spe-
cific to certain regions in Europe, and not so much on objective conditions in
the physical environment. The general tendency was that, in all countries that
participated in the survey, the risks to the environment from various issues were
considered moderately high. In two countries (Russia and Croatia), the average
of the risk-perception index was above four, which showed that most people
regarded environmental and technological issues as very dangerous or extremely
dangerous for the environment.
Results indicated that there were some regional tendencies, and people in
Eastern and Central Europe were more concerned with environmental and
RU-Russia 4.14
HR-Croatia 4.03
BG-Bulgaria 3.94
ES-Spain 3.91
LT-Lithuania 3.90
DE-Germany
3.86
AT-Austria 3.85
SK-Slovak Republic 3.84
SI-Slovenia 3.84
FR-France 3.74
CH -Switzerland 3.63
LV-Latvia 3.61
CZ-Czech Republic 3.58
SE-Sweden 3.48
DK-Denmark 3.47
FI-Finland 3.43
BE-Belgium 3.42 Total mean = 3.72
GB-Great Britain 3.35
NO-Norway 3.35
2.0 2.5 3.0 3.5 4.0 4.5 5.0
Figure 2.1 Perception of risk across the European countries’ means of risk index, where 1
is ‘not at all dangerous’ and 5 is ‘extremely dangerous’ (ISSP 2010, N = 25,133).
t echnological risks than people from the rest of Europe. Later in this chapter,
we will explore in more detail the influence of economic development and other
socio-political indicators upon risk-perception differences across countries.
In addition, considering the analysis of risk perception of different environ-
mental and technological issues (Figure 2.2), there were significant differences
comparing the perceptions of different environmental issues and technologies
(F test. p<0.00).
The risks from air pollution from industry and pesticides were perceived as the
highest, and risks from pollution from cars and risks from the genetically modi-
fied crops were perceived as the lowest. Urban and Hoban (1997) indicated that
individuals had higher risk perceptions if they thought risks were involuntary.
This explanation was consistent with our results, as pollution from industry and
chemicals in farming were involuntary risks, and genetically modified organisms
(GMOs) were a voluntary risk (one could choose to buy GMOs, but could not
choose to avoid the polluted air from industry).
A surprising finding was that the level of nuclear power and GMO risk percep-
tion was lower compared to all other risks (except car pollution). Many studies
on risk perception reported that nuclear-power risks were usually regarded as
Air pollution from

cars
5.00
Air pollution from 3.53 Modifying genes of

3.94 4.00
industry crops
3.54
3.00
3.87
2.00
Chemicals and
Climate change
pesticides in farming 3.67
3.69
3.83
Pollution of rivers, Nuclear power
lakes and steams stations
Figure 2.2 Perceptions of different types of risk at individual level. Mean scores (ISSP 2010;
N = 25,133), where 1 is ‘not at all dangerous’ and 5 is ‘extremely dangerous’.
high compared to other types of risks. In the classic book, The Perception of Risk
Slovic (2000) applied the psychometric paradigm and reported that the risk per-
ception of nuclear power was the highest among many unwanted environmental
features, considering such risk factors as fatality, threat for future generations or
catastrophic potential.
Our research results showed that well-known, empirically observable and
(in terms of the psychometric paradigm) ‘old’ risks (pollution from industry,
water pollution) were perceived as more dangerous to the environment as those
that were uncertain or not directly perceivable (e.g., GMO, nuclear power or
climate change). These results, in a way, contradicted the ideas of Slovic (1987,
2000), who stated that risks that had catastrophic potential and were uncontrol-
lable or not observable usually received higher scores of risk perception. To
explain these results, we considered the macrolevel indicators, measuring, for
example, the real levels of air and water pollution, and analysing if these objec-
tive indicators correlated with risk perceptions at the country level. Obviously, if
individuals were heavily exposed to low-quality living environments, this would
capture their attention.
Later in this chapter, we focus on the environmental and technological risk
perception in general, using the risk-perception index, which was derived from
seven risk-perception items.
The individual level: structural determinants

An analysis of the relations between socio-demographic variables and the risk-
perception index indicated significant differences in risk perception (Figure 2.3).
3.20 3.40 3.60 3.80 4.00

Male 3.63
Gender
Female 3.81
15–24 3.70
25–39 3.78
Age
40–54 3.75
55+ 3.67
Highest, Top 08–10 3.62

Self Social-class Middle, 04–07 3.73
placement Lowest, 01–03 3.87
A big city 3.82

The suburbs or outskirts of a big city 3.63
Living
A town or a small city 3.72
place
A country village 3.72
A farm or home in the country 3.50 Total mean =3.72
Figure 2.3 Risk perception across socio-demographic characteristics. Mean scores

(ISSP 2010), where risk-perception mean varies from 1. ‘not dangerous at all’
to 5. ‘very dangerous’.
The tests for the differences in means (Mann-Whitney U test for risk percep-
tion means by gender, and Kruskal–Wallis test for risk perception means by age
groups, self-social class placement and living place) were all significant at the
0.01 level. Women, individuals who positioned themselves in the lowest social
class, individuals aged from 25 to 39 and those who lived in big cities showed
higher (than average) levels of risk perception. Therefore, it seemed that risk
perception was closely related to personal vulnerability. Those in vulnerable situ-
ations, owing to either gender-specific roles, work status, social-class placement
or living environment, had higher levels of risk perception.
The lowest social-class members are the most vulnerable with regard to
environmental risks (everyday experiences, water shortage, and living close to
industry). People in big cities are also most vulnerable regarding, for example, air
or water quality. Those in a more secure social status expressed a lower level of
risk perception. A more secure social status is related to working age and higher
social-class placement.
A multiple linear regression analysis was conducted to predict risk perception
from structural factors. For this analysis, the derivative risk-perception index was
used as a dependent variable. Independent variables in this model were gender,
age, education, self-social-class placement and living place (urban–rural). All

these variables showed significant correlations with the risk-perception index.
According to the model, gender was the most significant factor. Many studies
also revealed that women perceived risks to be higher than men did (e.g., Slovic,
1999; Slovic et al., 2004; Henwood et al., 2008). Henwood et al. (2008) mentioned
an explanation common to the academic literature that gender differences in risk
perception could occur as a result of biological or psychological factors (so-called
‘gender effects’). However, authors argued that the socio-political dimension of

gender was important and suggested an approach called ‘effects made by gender’.
They analysed the ways in which gender regulated the social and cultural experi-
ences of individuals, how people engaged in the world and how they conducted
themselves in everyday life (Henwood et al. 2008:p.664). The authors discussed
different discourses on how men and women articulated the values of technolo-
gies, and the use and impact of technologies in everyday life (ibid:p.673). These
different socially shaped discourses on technologies probably led to different pat-
terns of risk perceptions.
Self-placement on the social-class scale was also significantly correlated with
risk perception (Spearman rho = -0.124, p = 0.000). The lower the self-social-
class placement, the higher the level of risk perception.
The weakest predictive power, as seen from the regression model, was from
variables of age and education. This result could be explained by the fact that the
relation between age and risk perception was not linear. As results in Figure 2.3
indicated, higher means of risk perception were in the age groups of 25–39 and
40–54 years. Furthermore, the youngest and oldest people saw risks as lower.
Even though most socio-demographic characteristics were significantly corre-
lated with risk perception, structural factors appeared to have very low explanatory
Table 2.2 Multiple linear regression analysis of structural factors as determinants of

risk perception.
Model 1 Unstandardized Standardized

Coefficients Coefficients
B Std. Error Beta Sig.
(Constant) 3.966 0.026 0.000

Gender (1 = male) 0.173 0.009 0.133 0.000
Age -0.003 0.000 -0.070 0.000
Education: years of schooling 0.000 0.000 -0.028 0.000
Self-social-class placement -0.046 0.002 -0.123 0.000
(1 = lowest)
Living place: urban–rural -0.039 0.003 -0.075 0.000
(1 = a big city)
Adjusted R Square 0.042

Dependent Variable: risk-perception index, from 1. ‘not dangerous at all’ to 5. ‘very dangerous’
power. The model factors only explained 4.2% of risk perception variation. There
fore, the set of explanatory variables should be extended.
Cognitive dimension, trust and environmental concern

In addition, we developed an explanatory model of risk perception with variables
that represented individual characteristics that were more dynamic. Some were
related primarily to understanding and perceiving environmental issues (environ-

mental concern, knowledge about causes and solutions to environmental problems
and trust in science’s ability to solve environmental problems). Items of social and
political trust revealed broader attitudinal standpoints that could mediate risk per-
ceptions. The results of the regression analysis are presented in Table 2.3.
From the variables included in model 2, environmental concern is the strongest
predictor of risk perception, and the least significant are the items of subjec-
tive knowledge. This model explains better risk perception than the model with
structural determinants. However, the variance that is explained is still quite low
(Adjusted R2 = 0.115)
Environmental concern had a significant impact on environmental and techno-
logical risk perceptions (Spearman rho = 0.273, p<0.00). The more a person was
concerned with environmental conditions, the higher they perceived environmental
risks. However, the direction of the relation for these variables was not clear; high risk
perception could have caused a high level of environmental concern, or vice versa.
The correlations between risk perception and trust items were significant,
but weak. Trust in science (Spearman rho = 0.068, p<0.00) and political trust
Table 2.3 Multiple linear regression analysis of cognitive factors, trust and environmental
concern as determinants of risk perception.

(Constant) 3.050 0.027 0.000

Environmental concern 0.182 0.004 0.300 0.000
(1 = not at all concerned)
Subjective knowledge: causes of -0.016 0.006 -0.025 0.006
environmental problems
Subjective knowledge: solutions -0.014 0.006 -0.022 0.013
to environmental problems
Trust in science (1 = high) 0.023 0.004 0.037 0.000
Social trust (1 = high) -0.060 0.003 -0.116 0.000
Political trust (1 = high) 0.061 0.004 0.100 0.000

(Spearman rho = 0.132, p<0.00) influenced risk perception in a different direc-

tion than social trust (Spearman rho = -0.126, p<0.00). The lower the trust in
science, the higher the risk people attributed to environmental problems and
technologies. However, social trust worked in an opposite manner. Those who
revealed a higher level of social trust also indicated risks to be higher, compared
to those who expressed a lower level of trust. We attributed this differentiation
in trust in relation to risk perception to broader risk contexts. Social trust was
not directly related to the attitudes towards the environment. However, people
that trusted others probably were more often actively involved in environmental
activities, thus having common goals (for example, environmental protection).
Possibly, collective action and collective awareness would be more characteris-
tic of those with higher levels of social trust.
Cognitive determinants that were included in the explanatory model had sig-
nificant, but very low correlations with risk perception. The risk-perception index
correlated both with knowledge about the causes of environmental problems
(Spearman rho = 0.03, p = 0.00) and knowledge about the solutions to environ-
mental problems (Spearman rho = 0.018, p = 0.00). However, this relation could
be of secondary importance in explaining risk perception. In general, Europeans
did not know a lot about solutions to various environmental problems. Only 21.6%
indicated that they knew a lot or a great deal about these solutions.
Comparing regression models 1 and 2, it appeared that risk perception was
better explained by trust, environmental concern and cognitive indicators than by
structural determinants.
Finally, we combined all variables from the first and second models into one
regression to see how they explained risk perception at the individual level and
which of the variables had the strongest influence (see Table 2.4).
All sets of indicators in the regression model accounted for 14% of the total
risk-perception variance, which was low. From the joint regression model, gen-
der, self-social-class placement and environmental concern were the strongest
predictors of risk perception. The influence of other indicators on risk perception
was very weak. Cognitive determinants in the aggregated model appeared to be
insignificant.
The analysis of individual-level determinants of risk perception suggested a
constructivist nature of risk perception. It was very weakly determined purely by
social structures, but rather, was an outcome of world views or dynamic aspects
(e.g., living place) that create different risk contexts.
The differences of risk perception at the country level (presented in Figure 2.1)
indicated the need for contextual macrolevel factors that could explain these
variations.
Country-level contextual determinants

The analysis of relationships among contextual determinants used in this chapter
and their relation to the main dependent variable – environmental risk-perception
Table 2.4 Multiple linear regression analysis for all predictors of risk perception at
individual level.


(Constant) 3.314 0.037 0.000

Gender (1 = male) 0.146 0.009 0.113 0.000
Age of respondent –0.003 0.000 –0.070 0.000
Education I: years of schooling 0.000 0.000 –0.027 0.000
Self-social-class placement –0.039 0.003 –0.103 0.000

(1 = lowest)
Living place: urban–rural –0.038 0.003 –0.075 0.000
(1 = a big city)
Environmental concern 0.169 0.004 0.281 0.000
(1 = not at all concerned)
Subjective knowledge: causes of –0.006 0.006 –0.009 0.323
environmental problems
Subjective knowledge: solutions –0.004 0.006 –0.007 0.466
to environmental problems
Trust in science (1 = high) 0.018 0.004 0.029 0.000
Social trust(1 = high) –0.050 0.004 0.098 0.000
Political trust(1 = high) 0.049 0.004 0.082 0.000

national average – is presented in this section of the chapter. Table 2.5 provides
the macro level data used in the analysis.
The correlation matrix shown in Table 2.6 provides arguments to exclude
three variables from the explanatory model: GDP growth, EPI and Ecosystem
Vitality. These variables did not have statistically significant correlations with
environmental risk-perception national averages. There was a strong posi-
tive relationship between the risk-perception index and HDI annual average
growth. There was also a strong but negative relationship between the risk-
perception index and GDP per capita, non-income HDI, Democracy Index and
Environmental Health Index.
The five significant context determinants of risk perception are different in
nature. GDP per capita, non-income HDI and the Democracy Index relate to the
socio-economic and political situations of a country. These might be a result of
rapid or gradual changes within a society. Instead, HDI annual average growth
shows the dynamics; i.e. the average speed of changes in recent decades.
Moreover, the environmental health determinant relates to the quality of a living
environment.
Table 2.5 Macrolevel data and risk-perception index values (mean score) for European countries.
Country Risk GDP per GDP Non-income Democracy HDI EPI7 ENV ECO
perception1 capita2 growth3 HDI4 index5 growth6 HEALTH8 SYSTEM9
Russia (RU) 4.136 10,709.77 4.50 0.729 4.26 0.82 61.2 68.59 53.83
Croatia (HR) 4.025 13,500.85 -1.70 0.798 6.81 0.63 68.7 82.47 54.83
Bulgaria (BG) 3.935 6,580.81 2.50 0.795 6.84 0.69 62.5 73.21 51.85
Spain (ES) 3.914 30,736.00 0.01 0.897 8.16 0.42 70.6 88.71 52.52
Lithuania (LT) 3.901 11,852.17 1.33 0.832 7.24 0.71 68.3 74.34 62.32
Germany (DE) 3.857 41,723.37 4.09 0.915 8.38 0.43 73.2 90.75 55.70
Austria (AT) 3.85 46,444.18 1.88 0.859 8.49 0.3 78.1 89.47 66.80
Slovakia (SK) 3.837 16,509.90 4.83 0.854 7.35 0.69 74.5 84.50 64.41
Slovenia (SI) 3.835 23,417.64 1.22 0.853 7.69 0.59 65 84.98 44.93
France (FR) 3.744 40,706.08 1.97 0.898 7.77 0.45 78.2 90.75 65.69
Switzerland (CH) 3.634 74,276.72 2.95 0.889 9.09 0.18 89.1 92.29 85.90
Latvia (LV) 3.608 11,446.51 -0.34 0.822 7.05 0.81 72.5 90.18 54.90
Czech Republic (CZ) 3.579 19,764.02 2.30 0.886 8.19 0.5 71.6 86.87 56.37
Sweden (SE) 3.478 52,076.26 5.99 0.911 9.5 -0.04 86 92.77 79.32
Denmark (DK) 3.473 57,647.93 1.63 0.883 9.52 0.27 69.2 89.82 48.49
Finland (FI) 3.434 46,202.42 2.99 0.897 9.19 0.54 74.7 90.75 58.74
Belgium (BE) 3.42 44358.26 2.50 0.888 8.05 0.05 58.1 89.05 27.09
Great Britain (GB) 3.35 38,363.44 1.91 0.86 8.16 0.31 74.2 89.82 58.66
Norway (NO) 3.346 86,096.14 0.48 0.954 9.8 0.34 81.1 90.75 71.52
1 Risk-perception index (aggregate index from risk perception items means, from 1. ‘not dangerous at all’ to 5. ‘extremely dangerous’)
2 GDP per capita (current US$) in mid-2010
3 GDP annual growth (%) in 2010 compared to 2009
4 Non-income human development index value for 2010
5 Democracy index 2010
6 Environmental performance index (EPI) scores for 2010
7 EPI: Environmental Health scores for 2010
8 EPI: Ecosystem Vitality scores for 2010
Table 2.6 Correlation matrix of dependent variable of risk-perception index and

independent context variables.
Risk- GDP GDP Non-income Democracy HDI

perception per growth HDI index growth
index capita
Risk-perception
index
GDP per capita -0.649
GDP growth — —
Non-income HDI -0.669 0.766 —
Democracy index -0.740 0.805 — 0.903
HDI growth 0.615 -0.757 — -0.679 -0.731
EPI — 0.637 — — 0.622 —
EPI: Environmental -0.690 0.688 — 0.827 0.818 -0.670
Health
EPI: Ecosystem — — — — — —
Vitality
Note: Table only includes correlations where p<0.01
GDP: Gross Domestic Product; HDI: Human Development Index; EPI: Environmental Performance
Index
Before analysing further the influence of the five context factors on environ-
mental risk perception, we tested if there were significant differences among
countries in national averages for risk perception. We ran an ANOVA test for dif-
ferences across the 19 European countries chosen for the analysis. Results showed
significant differences: F = 198.585, p = 0.000<0.01. To explain the differences,
we examined the effects of multiple contextual predictors on environmental risk
perception using simple linear regression and scatter-plotting aggregate data. The
scatter plots are presented in Figures 2.4–2.8.
First, we examined the effect of the economic affluence indicator (i.e., GDP per
capita, 2010, US$) on the risk-perception index at the national level via the scat-
ter-plotting of aggregated data. As seen from the regression line in Figure 2.4, the
higher the GDP per capita, the lower the level of risk perception. Second, Figure 2.5
revealed that the higher the developmental level of a country, the lower the level of
environmental risk perception. Thirdly, the regression line in Figure 2.6 showed that
a higher democracy index decreased environmental risk perception.
The linear regression results of free determinants describing the socio-
economic and political state of a country implied that, with high living standards,
welfare and democracy, the risk perception tended to be lower. In other words,
socio-economic stability and comfort were related to an increased sense of secu-
rity, and democratic conditions were related to a sense of possibility to influence
and control the risks.
Then, the impact of pace of societal changes, as measured by the average
annual HDI growth (%) from 2000–2010, was analysed. The regression line in
$100,000
$90,000 y = −60,380x + 258,972
NO R2 = 0.4216
$80,000
CH
$70,000
$60,000
DK
$50,000 FI SE
AT
BE DE
$40,000 FR
GB
$30,000 ES
SI
$20,000 CZ
SK
$10,000 LV LT RU
HR
BG
$0
3 3.2 3.4 3.6 3.8 4 4.2 4.4
Figure 2.4 Impact of affluence on risk perception: x-axis: average aggregate risk-

perception index (2010, N = 24,927); y-axis: GDP per capita (2010, US$).
0.95 NO
FI SE DE
0.9 CZ FR ES
DK CH
BE AT
GB
0.85 SK
SI LT
LV
0.8 BG HR
y = −0.1442x + 1.3982
0.75 R2 = 0.4478
RU
0.7
3.2 3.4 3.6 3.8 4 4.2
Figure 2.5 Impact of human development level on risk perception: x-axis: average aggregate
risk-perception index (2010, N = 24,927); y-axis: non-income Human Development
Index 2010.
Figure 2.7 showed that the higher pace of changes was related to higher levels
of risk perception. This might be explained through the notion of cultural lag.
Cultural lags often occur in cases of rapid transformation and are related to the dis-
continuity of social structures, and social and political trust relationships. These,
12
10 SE
NO
DK
FI CH AT
FR DE
8 GB BE CZ ES
SI
LT
LV SK HR
BG
6
4 RU
y = −3.9343x + 22.544
R2 = 0.5469
2
0
3.2 3.4 3.6 3.8 4 4.2
Figure 2.6 Impact of democracy on risk perception: x-axis: average aggregate risk-

perception index (2010, N = 24,927); y-axis: Democracy Index 2010.
0.9
y = 0.6134x − 1.8121 RU
0.8 LV
R2 = 0.3817 LT
0.7 SK BG
HR
0.6 SI
FI
0.5 CZ
DE
FR
0.4 ES
NO
0.3 GB AT
DK
0.2 CH
0.1
BE SE
0
3.2 3.4 3.6 3.8 4 4.2
Figure 2.7 Impact of pace of societal transformations on risk perception: x-axis: average

aggregate risk-perception index (2010, N = 24,927); y-axis: average annual
Human Development Index growth (%) in 2000–2010.
in turn, c reate a sense of insecurity and vulnerability, thus raising sensitivity and
awareness of various risks. As seen from the figure, there was some regional
specificity. Post-socialist countries, like Russia, Croatia, Bulgaria, Lithuania,
Slovakia and Slovenia, showed high levels of environmental risk perception and
a high pace of developmental change. Western and Northern welfare societies,
100.00
95.00
FI SE CH
NO DE
90.00 LV FR
DK AT ES
GB
BE CZ
85.00 SI
SK
HR
80.00
75.00 LT
BG
70.00
y = −19.846x + 159.81 RU
65.00 R2 = 0.4756
60.00
3.2 3.4 3.6 3.8 4 4.2
Figure 2.8 Impact of environmental health on risk perception: x-axis: average aggregate

risk-perception index (2010, N = 24,927); y-axis: Environmental Health Index
2010.
such as Norway, Great Britain, Denmark, Belgium or Sweden, demonstrated a

low pace of transformation and low levels of environmental risk perception.
Lastly, the relation between environmental conditions and national average
environmental risk perception was analysed. The regression results showed
that the poorer the environmental health, the higher the level of risk perception.
Environmental health was related to air quality, access to drinking water and sani-
tation, and health impacts. Thus, negative experiences of these tended to raise risk
perceptions.
Conclusions and discussion

Environmental and technological risk perception in Europe is not homogeneous.
Higher levels of risk perception in Europe are related to experienced potential
vulnerabilities at both individual and national levels. At the individual level,
environmental and technological risk perception varies depending on the rela-
tive strength/weakness of the social structural position of an individual in a given
society. At the macrolevel, the risk perception is divided between stronger and
more stable democracies of Western and Nordic European countries on one side,
and Central, Eastern and Southern European countries on the other.
Environmental and technological risk perception is related to individual struc-
tural vulnerability factors in two ways: through gender and social class. Gender and
self-social-class placement are the strongest predictors at this level. Women and
those who place themselves in a lower social class have a higher risk perception.
Gendered differences in risk perception have often been explained by psy-

chological differences. Sociologically, it might also be explained through
gender-specific roles, which place women in situations of care (of either children
or other family members), thus, they are more likely to consider risks. Lower
social-class positions are related to a less secure employment status and a proba-
bility of increased exposure to vulnerable environments. Thus, personal structural
vulnerability produces higher risk perception.
Still, gender and self-social-class placement are not the strongest predictors
when we take into account a wider explanatory model. Trust, environmental
concern and cognitive dimensions have a higher explanatory power of risk per-
ceptions than structural factors. Environmental concern is the strongest predictor
of risk perception at the individual level. Environmental concern is of a dynamic
nature, as it can quickly change over time as a function of changing environmental
conditions or other contexts.
Regarding macrolevel variables, GDP per capita, non-income HDI and the
Democracy Index have a significant negative impact on environmental risk per-
ception. The pace of societal development has a significant positive influence on
environmental risk perception. This indicates that environmental risk perception,
as analysed at a macrolevel (national level), is sensitive to the general socio-
economic-political system’s stability or dynamics. The Environmental Health
Index is significantly negatively related to risk perception. This indicates that
objective negative situations in the surrounding natural environment transform
into higher levels of environmental risk perception.
Generally, a more secure individual social status and living environment, and
more stable socio-economic contexts lead to lower levels of environmental risk
perception, and vice versa.
It is a significant finding that national-level context variables have a consid-
erable influence on average environmental risk perceptions. This implies that
changes in risk perception are largely dependent on general societal transforma-
tion rather than solely on individual factors.
The country-level analysis supports the idea that risk perception is a collective
phenomenon and is socioculturally determined by national and regional contexts.
Countries fall into groups regarding risk-perception patterns based not on physi-
cal environments (neighbouring countries could have significantly different risk
perception means), but rather on socio-political contexts.
It is also possible to indicate two dimensions that are common for individ-
ual- and macrolevel risk perception: vulnerability and dynamics. Vulnerability,
defined both by biographical situations and by national socio-political situations,
leads to a higher feeling of insecurity, resulting in higher levels of risk percep-
tion. In addition, dynamics experienced in the form of rapid societal change or
individual value change lead to higher sensitivity towards the surrounding envi-
ronment. Trust is also a dynamic concept, as it can be built or lost (Slovic, 1993,
as cited by Keller et al., 2012:p.242). Our data show a significant influence of trust
on risk perception.
Coming back to the question raised in the title of this chapter – ‘Vulnerable
and insecure?’ – it seems that vulnerabilities at personal and national levels lead
to higher levels of risk perception, or the feeling of insecurity, in many European
countries.
Acknowledgements
This chapter was prepared with the support of the research project ‘International
social survey programme: citizenship, work orientations and social welfare in
Lithuania (ISSP LT-CIWO)’, funded by a grant (No. MIP-082/2014) from the
Research Council of Lithuania. Period of implementation 2014–2016, coor-
dinated by Kaunas University of Technology, Institute of Public Policy and
Administration.
Notes
1 World Development Indicators database. Gross domestic product (GDP) per capita
(current US$) in mid-2010. Available from: http://data.worldbank.org/indicator/
NY.GDP.PCAP.CD
2 World Development Indicators database. GDP annual growth (%) in 2010 compared
to 2009. Available from: http://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG/
countries
References
Adeola, F. O. (2004) Environmentalism and risk perception: empirical analysis of black and
white differentials and convergence. Society & Natural Resources, 17 (10), 911–939.
Appelbaum, R. P. (1970) Theories of social change. Chicago, Markham.
Beck, U. (1992) Risk society: towards a new modernity. London, Sage.
Brinkman, R. L. & Brinkman, J. E. (1997) Cultural lag: conception and theory. International
Journal of Social Economics, 24 (6), 609–627.
Economist Intelligence Unit. (2010) Democracy index 2010: Democracy in Retreat.
London, The Economist. Available from: https://graphics.eiu.com/PDF/Democracy_
Index_2010_web.pdf
Emerson, J., Esty, D. C., Levy, M. A., Kim, C. H., Mara, V., de Sherbinin, A. &
Srebotnjak, T. (2010) Environmental performance index. New Haven, Yale Center for
Environmental Law and Policy.
Engel, U. & Pötschke, M. (1998) Willingness to pay for the environment: social structure,
value orientations and environmental behaviour in a multilevel perspective. Innovation:
The European Journal of Social Science Research, 11 (3), 315–332.
Finucane, M. L., Slovic, P., Mertz, C. K., Flynn, J. & Satterfield, T. A. (2000) Gender,
race, and perceived risk: the ‘white male’ effect. Health, Risk & Society, 2 (2), 159–172.
Franzen, A. (2003) Environmental attitudes in international comparison: an analysis of the
ISSP surveys 1993 and 2000*. Social Science Quarterly, 84 (2), 297–308.
Franzen, A. & Meyer, R. (2010) Environmental attitudes in cross-national perspective: a
multilevel analysis of the ISSP 1993 and 2000. European Sociological Review, 26 (2),
219–234.
Frizzell, A. & Pammett, J. H. (Eds.). (1997) Shades of green: environmental attitudes in

Canada and around the world. Vol. 2. Ottawa, McGill-Queen’s Press-MQUP.
Gupta, N., Fischer, A. R. & Frewer, L. J. (2011) Socio-psychological determinants of
public acceptance of technologies: a review. Public Understanding of Science, 21 (7),
782–795.
Hadler, M. & Haller, M. (2011) Global activism and nationally driven recycling: the
influence of world society and national contexts on public and private environmental
behavior. International Sociology, 26 (3), 315–345.

Hayes, B. C. (2001) Gender, scientific knowledge, and attitudes toward the environment: a
cross-national analysis. Political Research Quarterly, 54 (3), 657–671.
Henwood, K. L., Parkhill, K. A. & Pidgeon, N. F. (2008) Science, technology and risk
perception: from gender differences to the effects made by gender. Equal Opportunities
International, 27 (8), 662–676.
Hunter, L. M., Hatch, A. & Johnson, A. (2004) Cross-national gender variation in environ-
mental behaviors. Social Science Quarterly, 85 (3), 677–694.
ISSP Research Group (2012) International Social Survey Programme: Environment
III - ISSP 2010. GESIS Data Archive, Cologne. ZA5500 Data file Version 2.0.0,
doi:10.4232/1.11418 Available from: https://dbk.gesis.org/dbksearch/sdesc2.asp?no =
5500.
Keller, C., Bostrom, A., Kuttschreuter, M., Savadori, L., Spence, A., & White, M. (2012).
Bringing appraisal theory to environmental risk perception: a review of conceptual
approaches of the past 40 years and suggestions for future research. Journal of Risk
Research, 15 (3), 237–256.
Kim, S., Choi, S. O. & Wang, J. (2013) Individual perception vs. structural context: search-
ing for multilevel determinants of social acceptance of new science and technology
across 34 countries. Science and Public Policy, 41, 44–57.
Lima, M. L., Barnett, J. & Vala, J. (2005) Risk perception and technological development
at a societal level. Risk Analysis, 25 (5), 1229–1239.
Marquart-Pyatt, S. T. (2007) Concern for the environment among general publics: a cross-
national study. Society & Natural Resources, 20 (10), 883–898.
Olofsson, A. & Öhman, S. (2006) General beliefs and environmental concern transatlantic
comparisons. Environment and Behavior, 38 (6), 768–790.
Renn, O., & Benighaus, C. (2013). Perception of technological risk: insights from research
and lessons for risk communication and management. Journal of Risk Research, 16
(3–4), 293–313.
Siegrist, M., & Cvetkovich, G. (2000). Perception of hazards: the role of social trust and
knowledge. Risk Analysis, 20 (5), 713–720
Silbereisen, R. K. & Chen, X. (Eds.). (2010) Social change and human development:
concept and results. London, Sage Publications.
Simmel, G. (1968) The conflict in modern culture and other essays. New York, Teachers’
College Press.
Slovic, P. (1987) Perception of risk. Science, 236, 280–285.
Slovic, P. (1999) Trust, emotion, sex, politics, and science: surveying the risk-assessment
battlefield. Risk Analysis, 19 (4), 689–701.
Slovic, P. (2000) The perception of risk. London, Earthscan.
Slovic, P., Finucane, M. L., Peters, E. & MacGregor, D. G. (2004) Risk as analysis and
risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Analysis,
24 (2), 311–322.
Smelser, N. J. (1968) Toward a general theory of social change. In: Smelser, N. J. (1968)
Essays in sociological explanation. Englewood Cliffs, Prentice Hall.
Sztompka, P. (1993) The sociology of social change. Oxford, Blackwell.
UNDP Human Development Report Office. (2010) Human development report 2010: the
real wealth of nations: pathways to human development. New York, United Nations
Development Programme. Available from: http://hdr.undp.org/sites/default/files/reports/
270/hdr_2010_en_complete_reprint.pdf (accessed 16 March, 2016).
UNDP Human Development Report Office. (2014) Human Development Report 2014:
Sustaining Human Progress: Reducing Vulnerabilities and Building Resilience. New
York, United Nations Development Programme. Available from: http://hdr. undp. org/
sites/default/files/hdr14-report-en-1. Pdf (accessed 23 April, 2015).
Urban, D. & Hoban, T. J. (1997) Cognitive determinants of risk perceptions associated
with biotechnology. Scientometrics, 40 (2), 299–331.
Viklund, M. J. (2003) Trust and risk perception in Western Europe: a cross-national study.
Risk Analysis, 23 (4), 727–738.
Yale Center for Environmental Law & Policy (2010) 2010 Environmental performance
index. Available from: http://www.ciesin.org/documents/EPI_2010_report.pdf (accessed
7 May, 2015).
Wachinger, G., Renn, O., Begg, C. & Kuhlicke, C. (2012) The risk perception paradox—
implications for governance and communication of natural hazards. Risk Analysis, 33
(6), 1049–1065.
World Bank (2014a) GDP per capita. World Development Indicators, Available from:
http://data.worldbank.org/indicator/NY.GDP.PCAP.CD (accessed 29 April, 2015).
World Bank (2014b) GDP growth. World Development Indicators. Available from: http://
data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG/countries (accessed 29 April,
2015).
Chapter 3
Changing concern about

environmental threats and risks
A longitudinal and multilevel perspective
on the relationship between values
and interests
André Schaffrin and Alexander Schmidt-Catran
Introduction
A number of recent insightful studies in environmental sociology have provided
a more thorough understanding of environmental values, beliefs, attitudes and
behaviours (see Dunlap and Jones, 2002; Gifford and Nilsson, 2014; Schaffrin,
2011a for a review). Theories from the early 1990s seeking to understand cross-
national variation in environmental concern argue that economic wealth leads
to wider support for environmental protection in the highly developed, affluent
nations, either via post-materialist values or as a consequence of an increasing
demand for environmental goods (Fairbrother, 2013; Franzen, 2003; Franzen and
Meyer, 2010; Franzen and Vogl, 2013a; Givens and Jorgenson, 2011; Knight and
Messer, 2012; Pampel, 2014). Another group of scholars argues that economic
development is seen as only one explanation as opposed to environmental deg-
radation which substantially alters environmental concern at the individual and
national level (Adeola, 2004; Brechin and Kempton, 1994; Marquart-Pyatt, 2012;
Uyeki and Holland, 2000). Empirical results from this early wave of compara-
tive environmental sociology used cross-sectional data and mainly correlative
measures (Diekmann and Franzen, 1999; Franzen, 2003; Inglehart, 1995; Kidd
and Lee, 1997). In a second wave of studies, the improved quality of interna-
tional surveys, the availability of repeated cross-sectional data, and the statistical
methods used simultaneously to account for national- as well as individual-level
determinants of environmental concern stimulated a number of additional stud-
ies on post-materialism, affluence, and exposure to environmental degradation
(Franzen and Meyer, 2010; Franzen and Vogl, 2013a, 2013b; Gelissen, 2007;
Kemmelmeier et al., 2002).
However, results from both waves of research are inconclusive. For example,
a number of studies which argue for the importance of environmental degradation
report no or even a negative effect of affluence on measures of environmen-
tal concern (Brechin, 1999; Dunlap and Mertig, 1997; Dunlap and York, 2008;
Escobar, 2006; Fairbrother, 2013; Givens and Jorgenson, 2011; Israel, 2004;
Knight and Messer, 2012; Pampel, 2014; Sarigoellue, 2009). Findings from
Changing concern about threats and risks 57
Brechin et al. (2005), Dunlap and York (2008), and Yuchtman-Ya’ar (2002)
also challenge the affluence approach by revealing that global measures together
with local ones, as well as individual post-materialist values, are strong even in
less developed countries. In contrast, proponents of the affluence approach and
the post-materialism thesis report only insignificant effects of indicators of envi-
ronmental degradation (Franzen and Meyer, 2010; Franzen and Vogl, 2013a,
2013b; Gelissen, 2007).
According to Dietz et al. (2005), ‘Reaching consensus about the relationship

between affluence, values, and environmentalism requires better integration of
these theories, clearer specification about the implications of each, and compara-
tive tests of these implications’ (p. 360). We argue that the majority of studies
fail to agree on at least three dimensions of environmental concern. Firstly, while
earlier studies rely chiefly on bivariate correlations or cross-sectional data, only
recently have scholars started to analysis multiple levels (national and indi-
vidual) in combination with changes in local and global conditions over time
(Fairbrother, 2013; Nawrotzki and Pampel, 2013; Pampel and Hunter, 2012).
Inconclusive results might be resolved by using multilevel and longitudinal data
and techniques.
Secondly, current research rarely considers the two-dimensional value basis of
environmental utilization and environmental preservation, as suggested in the lit-
erature on environmental psychology (Hansla et al., 2013; Kaiser and Scheuthle,
2003; Milfont et al., 2006; Milfont and Duckitt, 2010; Milfont and Gouveia, 2006;
Wiseman and Bogner, 2003). Both dimensions regard personal interest factors
and values as the main drivers. Concerning values, the debate has been dominated
by the discussion of post-materialism as one possible determinant. However,
Dietz et al. argue that it would be of particular value to investigate ‘the full spec-
trum of ideas about values [ . . . ] including both the values Inglehart has invoked
in his materialist/post-materialist distinction and the consideration of altruism
and self-interest that dominates the broader literature’ (2005, p. 360). Factors of
personal interest have so far been considered only as either negative exposures
to environmental threats in developing countries, or as the individual demand
for environmental goods in developed societies. Taking a more holistic approach
to individual self-interest as a determinant of the environmental concern would
also mean analysing negative contextual conditions, such as poverty risks, in the
Global North and positive contextual conditions like benefits from eco-tourism in
the less affluent Global South.
Thirdly, only a minority of empirical studies consider the intra-attitudinal
dimensions (Diekmann and Franzen, 1999; Dunlap and York, 2008; Marquart-
Pyatt, 2012; Schaffrin, 2011b). These studies, however, clearly demonstrate a
variation in the effect of interest factors and indicators of personal values on the
different components of environmental concern.
Our study begins to fill this gap by examining how post-materialism, affluence,
and a wider range of factors could influence different components of environ-
mental concern. The study has three objectives: (1) to develop a more thorough
58 André Schaffrin and Alexander Schmidt-Catran
understanding of the variation of environmental concern cross-nationally and

within societies, using more general value approaches (Dietz et al., 2005; Milfont
et al., 2006; Schultz and Zelezny, 1998) and the distinction between environmen-
tal utilization and environmental preservation (Hansla et al., 2013; Kaiser and
Scheuthle, 2003; Milfont et al., 2006; Milfont and Duckitt, 2010; Milfont and
Gouveia, 2006; Wiseman and Bogner, 2003); (2) to identify and discuss positive
and negative conditions arising from personal interests and vulnerabilities, which
influence the level of concern (Gifford, 2011; Marshall et al., 2005; Schaffrin,
2011b; Whitmarsh, 2009); and (3) to demonstrate that determinants affect com-
ponents of environmental concern across countries differently (Diekmann and
Franzen, 1999; Dunlap and York, 2008; Marquart-Pyatt, 2012; Schaffrin, 2011b).
In doing so, we seek to move the debate from theoretical competition towards
integrating the most commonly applied approaches of post-materialism, affluence
and environmental degradation.
This research uses data from the Environment modules of the International
Social Survey Programme (ISSP Research Group, 2012, 2003, 1995), which
were surveyed in 1993, 2000 and 2010. Combining these datasets allows for a
multilevel, comparative analysis considering national variation over time. The
paper begins by reviewing cross-national research by discussing in detail the three
dimensions of environmental concern: the local versus global; the values ver-
sus interest; and the intra-attitudinal. Throughout the theory section, the paper
expresses the importance of considering multilevel approaches, by using a range
of values and interest factors, and investigates the effects of these determinants
on different components and measures. The ISSP datasets are used to examine a
multilevel model, including individual-level and country-level variables to pre-
dict three distinct dimensions of environmental concerns: individual awareness
of environmental threats; environmental efficacy; and willingness to pay. A series
of macrolevel explanatory factors describing positive and negative interests are
included, alongside post-materialist values and affluence measures. The results
demonstrate the assumed variation in these determinants across the three compo-
nents of environmental concern, and stress the importance of dimensionality for
future cross-national scholarship.
Theory: three dimensions of environmental concern

The local versus global dimension
Different theories have emerged during the last two decades of environmental
sociological research, focusing on the role of economic wealth and environmental
degradation to explain national and individual differences in environmental con-
cern. In the following, we will briefly discuss the basic theories and review the
empirical support for both approaches.
One commonly stated argument is that economic wealth leads to wider sup-
port for environmental protection in the highly developed, affluent nations
(Fairbrother, 2013; Franzen, 2003; Franzen and Meyer, 2010; Franzen and Vogl,
2013a; Givens and Jorgenson, 2011; Knight and Messer, 2012; Pampel, 2014).
The mechanisms linking economic wealth and environmental concern are based
on either more general value-based explanations or economic theory. Literature
on the value-based explanation identified post-materialist values as a higher
order need, which is a crucial determinant of the level of concern (Franzen and
Meyer, 2010; Inglehart, 1995; Inglehart and Abramson, 1999). Whether people
hold (post)materialist values depends upon their current economic situations and
levels of social security, but even more upon their economic situations during
their socialization. Abramson and Inglehart (1999) define materialism as a state
where individual concerns are basically oriented towards daily survival, which
implies the avoidance of material scarcity and the pursuit of economic and social
security. Post-materialists, in contrast, strive for a higher quality of life and self-
actualization, including concern for environmental issues which are beyond direct
relevance to their own material situation (Inglehart, 2008, 2006, 2002).
Scholars advancing the affluence approach argue that in developed nations it is
not necessarily post-materialist values which increase concern but the demand for
environmental goods (Fairbrother, 2013; Franzen and Meyer, 2010). Environmental
quality is seen as an amenity or public good which can be more easily accessed.
Hence, demand rises with increasing levels of wealth independently of value
orientations (Diekmann and Franzen, 1999; Diekmann and Preisendorfer, 2003;
Franzen, 2003; Franzen and Meyer, 2010; Kemmelmeier et al., 2002; Meyer and
Liebe, 2010). In this line of argument, developing economies have to pass a point
where the impact of economic growth on the environment reaches a high level,
in order to gain sufficient resources and cognition for environmental protection
(Grossman and Krueger, 1995; Spaargarden and Mol, 1992; Zahran et al., 2007,
see also Jorgenson and Clark, 2012). Here, development and modernization are
‘creating opportunities for both reduced environmental harm and increased envi-
ronmental concern’ (Givens and Jorgenson, 2013, p. 421).
Both the post-materialism hypothesis and the affluence approach have
stimulated a large amount of research on this topic. Most debate is stimulated
by the macrolevel assumption of a linear (or curvilinear) relationship between
economic development and environmental concern. Early studies showed post-
materialism had a positive effect on environmental concern (Inglehart, 1995;
Kidd and Lee, 1997). Support for the affluence approach comes from a num-
ber of contributions reporting a positional effect of economic factors operating
independently from post-materialist values (Best, 2009; Best and Mayerl, 2013a;
Diekmann and Franzen, 1999; Franzen, 2003; Franzen and Meyer, 2010; Franzen
and Vogl, 2013a, 2013b; Gelissen, 2007; Kemmelmeier et al., 2002). Franzen
and Vogler (2013a), for example, found that taking national variations of acqui-
escence into account would reveal a positive effect of economic wealth on the
willingness to protect the environment, at both the individual as well as the coun-
try levels. However, a number of studies reveal a negative or insignificant effect
contingent upon socio-economic status or economic wealth (Escobar, 2006;
Sarigoellue, 2009; Givens and Jorgenson, 2011; Fairbrother, 2013; Knight and
Messer, 2012; Pampel, 2014). One prominent critique reporting that economic
wealth has a negative effect comes from Dunlap and colleagues (Brechin, 1999;
Dunlap and Mertig, 1997; Dunlap and York, 2008). These scholars argue that the
globalization of environmental concern in developing and highly industrialized
countries alike means that economic affluence and post-materialism do not pro-
vide a consistent explanation (Dunlap and Mertig, 1997; Dunlap and York, 2008;
Gifford et al., 2009).

The ‘objective problems-subjective-values’ approach (OPSV) brought forward
by Inglehart (1995) emerged as a way to explain the mixed evidence on the influ-
ence of affluence and post-materialist values (see Brechin, 1999; Brechin and
Kempton, 1997; Kidd and Lee, 1997). Inglehart argues that environmental con-
cern is derived from two sources; namely, the level or post-materialist values in
highly industrialized countries and the individual or national exposure to envi-
ronmental degradation predominantly found in developing nations. ‘[R]elatively
advantaged groups and nations can better afford to protect themselves from myriad
environmental insults’ (Pampel and Hunter, 2012, p. 423), whereas disadvantaged
groups or developing countries are more likely to experience greater exposure
to local environmental degradation (Adeola, 2004; Brechin and Kempton, 1994;
Marquart-Pyatt, 2012; Uyeki and Holland, 2000). This approach claims that
while post-materialist values target more general, global and symbolic environ-
mental concerns, exposure to environmental pollution and potentially harmful
degradation stimulates specific attitudes to changing local problems among the
communities and citizens of less developed countries (Brechin, 1999). In fact, the
broad majority of studies presenting critical results on the affluence approach and
the post-materialism thesis discussed above used different measures from schol-
ars of the affluence thesis, including measures of environmental degradation and
concern for local rather than global environmental problems (Fairbrother, 2013;
Givens and Jorgenson, 2011; Knight and Messer, 2012; Mostafa, 2013). However,
findings are inconclusive since analyses by Brechin et al. (2005) and Dunlap and
York (2008) also reveal global measures of environmental concern together with
local ones to be strong even among less developed countries.
The values versus interest dimension

As discussed above, there is national variation, not only in terms of individual
concern on global environmental issues such as climate change or biodiversity
loss, but there is also strong concern about local environmental degradation both
in the Global South and the Global North (Givens and Jorgenson, 2011; Goksen
et al., 2002). These and other findings indicate that environmental concern oper-
ates at different levels according to environmental degradation, post-materialism
and affluence. For instance, it is not clear why considering only post-materialist
and no other basic values would be a determinant. Furthermore, now that we
know about the significance of environmental degradation, why not analyse other
related factors likely to influence environmental attitudes and behaviour? In the

following, we try to integrate the discussion of affluence, values and degradation
on more general grounds of environmental values.
As Schultz and Zelezny state, ‘Values are distinct from attitudes or beliefs
because they function as an organized system and are typically viewed as deter-
minants of attitudes and behaviors’ (1998, p. 256). This statement is in line with
the commonly cited cognitive hierarchy model proposed by Homer and Kahle
(1988), which claims that values influence behaviour indirectly through attitudes.
This general assumption about the determining role of values on attitudes and
behaviour has also received wide empirical support (Best and Mayerl, 2013b;
Milfont et al., 2010a).
Scholars distinguish between three basic environmental values: egoistic, altru-
istic and biospheric (Dietz et al., 2005; Milfont et al., 2006; Schultz and Zelezny,
1998). Biospheric values constitute a general understanding that nature is sacred
and worthy of protection, and as such is independent of the implicit dangers for
humankind. The altruistic value orientation sees environmental protection as
a means to preserve nature for individual good (e.g. health) or the survival of
humankind (e.g. in the context of natural resources and climate change). Egoistic
value orientation, in contrast, motivates individuals to look after their own inter-
ests in order to avoid perils, and to be able to afford environmental goods (Soyez
et al., 2009).
What can these values tell us about the multidimensional structure of envi-
ronmental concern? First, value orientations focus strongly on geographical and
spatial issues. Egoistic values should promote a local and short-term perspective
on environmental concern, whereas altruistic and biospheric values might stim-
ulate more concern about global environmental problems on a long-term scale
(Groot and Steg, 2007; Milfont et al., 2010b). Second, analyses of environmen-
tal psychology (Milfont and Duckitt, 2004; Milfont and Gouveia, 2006) reveal
a second-order structure of environmental concern with two dominant, slightly
negatively correlated factors: a dimension of environmental preservation, derived
from a general biospheric or altruistic value orientation; and an environmental uti-
lization dimension which reflects personal interests supported by egoistic values
(Hansla et al., 2013; Kaiser and Scheuthle, 2003; Milfont et al., 2006; Milfont
and Duckitt, 2010; Milfont and Gouveia, 2006; Wiseman and Bogner, 2003). The
dimension of preservation describes individuals who argue for environmental
protection not only on a local scale, but who also support concrete actions for
global environmental threats, even if the consequences of fighting them impose
personal costs. On the dimension of utilization, individuals are concerned about
either their own vulnerability in the face of environmental degradation or the ben-
efits to be derived from environmental quality (Dietz et al., 2005; Hansla et al.,
2013; Milfont and Gouveia, 2006).
This basic distinction is linked strongly to the discussion of local problems and
global values. It seems that scholars of post-materialism address only the dimen-
sion of environmental preservation, whereas explanations of national affluence or
environmental degradation predominantly include the dimension of utilization.

The fact that a number of studies find that affluence has an impact on environmen-
tal concern independently from individual or national levels of post-materialism
is only one of a number of results supporting this argument. Given these thoughts,
we argue that post-materialism can be interpreted, along with other values
such as collectivism, as a trigger for concern about environmental preservation
(Dietz et al., 2005; Hansla et al., 2013; Milfont and Gouveia, 2006). Similarly,
degradation and affluence describe two special cases along the dimension of envi-
ronmental utilization, where short-term and highly localized personal interests
are addressed: vulnerability towards negative exposures to environmental threats;
and positive opportunities to gain individual benefits from environmental goods.
Vulnerability and opportunities both arise from contextual factors in the respec-
tive countries and the personal environment of the respondents. Therefore, we
will discuss personal interests supporting environmental utilization which arise
from the specific contexts discussed in the following.
Individual interests for environmental preservation arise where either being
vulnerable because of environmental degradation or finding opportunities to gain
personal benefit out of environmental protection are dominant (e.g., Marshall
et al., 2005). In other words, there are a number of positive and negative con-
ditions influencing individual interests for environmental preservation. This
understanding allows us to take a more general perspective on environmental
interest than just considering exposure to environmental degradation in the Global
South, and the demand for environmental goods in the Global North, as proposed
by the OPSV approach. We argue that studies of environmental concern should
include a wider range of positive and negative interest factors across countries
and (within societies) between social groups.
In highly developed countries, positive conditions can be rather straightforward.
In line with the affluence approach, individuals might display environmental behav-
iour such as cycling, insulating or recycling simply because they benefit directly
from these in terms of better health or increasing financial returns (Gifford, 2011;
Whitmarsh, 2009). In the more recent developments in combating climate change,
a number of additional positive interests arise where domestic climate policies,
for instance, subsidize privately owned renewable energy installations and thus
create a direct benefit for specific socio-economic groups (Faiers et al., 2007;
Sardianou, 2007; Schelly, 2010; Walker, 2008; Welsch and Kuehling, 2009).
What has hardly been considered by the affluence approach is that the same
policies also produce negative conditions for certain socio-economic groups. For
example, sectoral changes towards low-carbon economies create new risks in
the job market for employees in the industrial sector (Bowen and Fankhauser,
2011; Carfi and Schiliro, 2012; Jaenicke, 2012; Kikuchi, 2011). As O’Connor
et al. show, ‘Economic circumstances and anxieties are not important predictors,
but the belief that environmental protection efforts do not threaten jobs for peo-
ple like the respondent, limit personal freedoms, and hurt the economy’ (2002,
p. 1). The debate about energy poverty further indicates that support for renewable
energy in developed countries most often increases energy prices for households,
and might cause an existential threat to vulnerable low-income groups of soci-
ety (e.g., Schaffrin, 2014; Schaffrin and Reibling, 2015). Several studies also
demonstrate that location and direct environmental threats, such as rising sea
levels via climate change, hurricane activity, or other potential environmental
hazards, influence environmental concern in the highly developed countries
(Drori and Yuchtman-Ya’ar, 2002; Gifford and Nilsson, 2014; Marshall et al.,
2005; Peacock et al., 2005; Stedman, 2003). For example, using Geographic
Information Systems, Brody et al. (2008) show that proximity to the coastline,
or cheap dwellings at lower elevations and other areas at high risk of sea-level
rise explains concern about climate change. Potential direct threats to individual
health in the developed world also arise from new technologies such as fracking
or nuclear power. In fact, one of the main triggers of the local protest movements
against nuclear power in Germany has been the local fear of direct exposure near
nuclear waste-disposal sites, particularly after the Chernobyl disaster in 1986
(see also Marshall et al., 2005 for the development of the environmental move-
ments in the USA during the 1960s).
In developing countries, negative attitudes arise from the need to avoid poten-
tially harmful situations. As stated by proponents of the OPSV approach, aware-
ness of local environmental problems grows out of an interest in minimizing health
risks or even death. What has rarely been considered is that positive conditions
exist independently of higher levels of affluence in developing countries. One pos-
itive factor is the proportion of natural habitats on the total land available. Nature
conservation sites exist not only in highly industrialized countries but more so in
the developing world. Conservation most often goes hand in hand with tourism in
developing countries, as a major benefit for local jobs and as a source of income
for the country. Furthermore, in developing countries, a higher percentage of the
population is employed in the agricultural sector. Farmers are more directly inter-
ested in keeping natural resources intact for their own benefit and see themselves
as increasingly vulnerable to industrial development (Pampel, 2014). The high
number of employees in the agricultural sector is not necessarily a direct influence
on environmental concern but increases the chances for an individual to build
social ties with someone in that business. As Macias and Nelson point out, ‘place
of residence is less significant in explaining utilitarian attitudes towards the envi-
ronment than familial or other social ties one has to farming’ (2011, p. 563; see
also Sharp and Smith, 2003; Wachenheim and Rathge, 2002). In other words, we
might find a stronger exposure to environmental knowledge and concern through
social ties and social capital in the agricultural sector in the developing countries
than in the developed world.
Given these arguments we conclude that environmental concern depends
strongly on where individuals live, and is independent of ad-hoc distinctions
between more or less affluent countries. This perspective stresses the impor-
tance of considering the relative position of the respondents or communities
within a society, or their relative exposure to environmental degradation rather
than economic development per se (Knight and Messer, 2012; Marshall et al.,
2005; Martínez-Alier, 2002; Pellow and Brulle, 2005). In other words, poten-
tial environmental threats to personal health caused by radioactive waste-disposal
sites might be as concerning for individuals in Western societies as real exposure
to intensive air or water pollution for citizens of developing countries. Higher
income allows individuals or groups to avoid extensive exposure to potential or
real environmental degradation and to gain benefits from positive conditions, in
economically poor and rich countries alike. Their subjective concern about envi-
ronmental conditions and opportunities thus depends on their relative position in
their respective societies.
Another relative issue is when the individual context changes over time
(Conroy and Emerson, 2014; Fairbrother, 2013; Givens and Jorgenson, 2011;
Israel, 2004; Knight and Messer, 2012). As discussed above, changes in job
opportunities and personal incomes might prove as threatening to individuals
as increasing environmental pollution. What is important is that the change in
personal conditions relative to others in the society produces a stronger impact
on environmental concern than differences in the actual level between these
groups. This is indicated by a number of studies which find that economic
growth has a stronger influence on concern for the environment rather than the
absolute level of national income (e.g., Conroy and Emerson, 2014; Givens and
Jorgenson, 2011).
The intra-attitudinal dimension

Individual concern for the environment is also characterized by an intra-
attitudinal structure. It occurs where a person who is aware of environmental
problems supports public or political actions to prevent them, and/or shows will-
ingness to take action (Dunlap and Jones, 2002; Schaffrin, 2011a). This definition
is based on the classical tripartite attitude theory, which identifies three compo-
nents of environmental concern: affective, cognitive and conative (Dunlap and
Jones, 2002; Maloney et al., 1975; Maloney and Ward, 1973; Rosenberg and
Hovland, 1960). The affective component refers to evaluative stages measured
mostly by individual awareness and emotional reaction to environmental prob-
lems. The cognitive component includes aspects of knowledge, beliefs or norms,
and basically describes personal rational insights into environmental issues.
Here individual efficacy is used for purposes of operationalization, in order to
measure the importance of addressing environmental problems. The conative
component is an expression of behavioural intentions most often measured as
willingness to pay.
These components are widely accepted in the mainstream literature of this
field (e.g., Best and Mayerl, 2013b; Franzen and Vogl, 2013a, 2013b; Marquart-
Pyatt, 2012). A more recent development, however, is the argument that the
effect of determinants of environmental concern, such as post-materialism,
affluence and positive or negative context conditions, varies across the measures
of these components and across countries (Marquart-Pyatt, 2012; Pampel and

Hunter, 2012; Schaffrin, 2011a). For example, Diekmann and Franzen (1999)
find that environmental awareness is negatively correlated with affluence,
whereas willingness-to-pay measures are positively correlated. Variation in the
relationship to different measures of environmental concern has been reported
for national indicators (e.g., Dunlap and York, 2008), as well as determinants at
the individual level (Marquart-Pyatt, 2008, 2007; Xiao and Dunlap, 2007). More
recently, Marquart-Pyatt (2012) demonstrated this variation effect for measures

of environmental awareness, efficacy and willingness to pay in a multilevel
approach where simultaneity accounts for macrolevel effects and individual-
level variance.
However, none of these studies makes explicit claims about how post-
materialism and affluence, or in more general terms personal values and (positive
and negative) conditions, influence each of the three components of the intra-
dimension of environmental concern. First, we argue, as personal values are
relatively stable and dominate cognitive concepts which rarely change over time,
they should have quite separate effects on all three components of awareness
(affective), efficacy (cognitive) and willingness to pay (conative). Personal values
predefine and structure environmental attitudes in a consistent and dominant way,
even under changing circumstances. This assumption is supported by the finding
that respondents in an Istanbul sample reported not only their willingness to pay
for local but also for global environmental problems (Goksen et al., 2002, see also
Schaffrin, 2011b).
Second, positive and negative context conditions such as environmental deg-
radation, job loss, or increasing returns from renewable energy policies might
directly or indirectly affect individual citizens in the medium term. Direct expo-
sure to low air quality or immediate and increasing costs for household energy,
owing to the rising support for renewable energy, directly and immediately
increase awareness. In contrast, it needs time and substantial cognitive resources
in order to become informed about either the fundamental causes and mechanisms
of environmental degradation, and the negative impact of environmental policies,
or else the technical and financial investment for people to benefit from renewable
energy programmes. Empirical studies support this assumption where awareness
is stronger than willingness to pay for individuals with greater exposure to envi-
ronmental degradation (Arbuthnott et al., 2012)
Third, affluence and personal income should have the strongest effect on
measures of behavioural intentions such as willingness to pay. Even smaller
changes in personal income or the economic condition of society are felt directly
and measured in the same currency: personal financial means. Economic down-
turn does not necessarily impact on the personal job situation but increases fears
of future risks and so might affect individual consumption and savings behaviour
(Conroy and Emerson, 2014). The reason is that the conative component –
willingness to act – already includes an evaluation of available resources and
opportunity costs. In fact, most of the literature on the influence of economic
wealth on environmental concern use willingness-to-pay measures (e.g., Conroy

and Emerson, 2014; Fairbrother, 2013; Kemmelmeier et al., 2002), or at least
exclude the dimension of personal awareness of environmental threats from the
analysis of environmental concern (e.g., Franzen and Vogl, 2013b; Nawrotzki
and Pampel, 2013).
Given these sorts of discussion, cross-national research on environmental con-
cern is limited in at least four ways. Firstly, as more recent studies increasingly
apply multilevel techniques (Franzen, 2003; Franzen and Meyer, 2010; Franzen
and Vogl, 2013a, 2013b; Kemmelmeier et al., 2002; Marquart-Pyatt, 2012, 2007;
Mostafa, 2013), analyses of potential changes over time at both the individual and
the national level are rare (see Fairbrother, 2013 as one exception from the rule).
Most of the earlier results, however, lack this multilevel time-change perspective
(see Knight and Messer, 2012 for a discussion). Secondly, as discussed above,
studies should include individual values and factors describing personal interests
(e.g., Milfont and Duckitt, 2004). Here, the more general value basis of environ-
mental concern should be considered. Thirdly, interest factors include positive as
well as negative conditions in both developing and developed countries. Fourthly,
values and interests might affect certain intra-attitudinal dimensions of environ-
mental concern differently, and thus should be analysed separately, as suggested
by Marquardt-Pyatt (2012).
To address these gaps in the literature, we apply techniques which simulta-
neously consider changes over time at the individual and national levels. We
investigate the effect of post-materialist values, as well as factors describing per-
sonal interests at both levels. Here, we consider negative context conditions in the
highly developed countries, as well as positive effects of natural environments
in the developing societies. Furthermore, this research examines whether levels
of three different dimensions – awareness of environmental threats, environmen-
tal efficacy and willingness to pay – are affected by post-materialism, economic
wealth and positive and negative context conditions, according to the assumptions
proposed above.
Methods, Data, Variables

Data
In our analysis we estimate models for three different dependent variables which
are derived as factor scores from a confirmatory factor analysis (CFA): (1) Enviro
nmental threat awareness; (2) environmental efficacy; and (3) willingness to pay.
We use all ISSP waves that include the environmental module (1993, 2000 and
2010). Therefore, we restrict the choice of items to those that are included in each
of the three waves. Table 3.1 shows this selection of items and their assignment to
the three latent constructs we aim to measure. In general we follow Marquart-Pyatt
(2012) in the selection items to measure environmental threat awareness, efficacy
and willingness to pay. However, two items that have been used as an indicators of
efficacy by Marquart-Pyatt (2012) have not been considered in all waves (‘There
are more important things in life than to protect the environment’ and ‘There is no
point in doing what I can for the environment unless others do the same’; neither
item was asked in 1993) and were therefore excluded from our list of items. On
the other hand, we added an item into the factor analysis that has been used by
Franzen and Vogel (2013b) and Nawrotzki and Pampel (2013) but ignored by
Marquatt-Pyatt (2012): ‘In order to protect the environment [COUNTRY] needs
economic growth’. For each wave, we performed confirmatory factor analyses

in which we analysed the three latent constructs simultaneously. The item ‘I do
what is right for the environment whether it costs more money or takes more time’
had to be taken out of the model because the factor loadings were very low in all
three waves (about 0.30). The final CFA models of the remaining items show suf-
ficiently good fit indices (compare notes below Table 1), convincingly high factor
loadings and a remarkable stability of the loadings across time. Using the model
shown in Table 1, we predicted factor scores which we used as the dependent
variables in the subsequent analyses. The factor scores of the latent constructs
environmental threat awareness and willingness to pay have been reversed
because the original scales measured awareness and willingness negatively. All
dependent variables have been z-standardized to allow direct comparison of the
effect sizes of independent variables.
As independent variables at the individual level we use years of schooling (top
coded to a maximum of 18 years), age, relative household income (household
income z-standardized using country-year-specific means and standard devia-
tions), gender (male = 0, female = 1), residency (big city = 1, all other = 0) and a
measure of post-materialism (following Franzen and Vogl, 2013b). Unfortunately,
the ISSP provides no other value measures for inclusion in the models.
In our theoretical conceptualization we differentiate between positive and
negative interest factors at the contextual level. As negative interest factors we
measure the two most essential human needs, breathing and drinking:
1 We measure the percentage of the population that has access to improved

drinking water sources (Source: Millennium Development Goals Data Base,
United Nations). Drinking water is an essential need which most affects
well-being. It is therefore a central indicator of individuals’ exposure to
environmental threats. Higher values on this variable indicate less exposure
to environmental degradation and, as a consequence, the absence of nega-
tive interests. Therefore we expect the variable to have a negative effect (the
higher the proportion of the population using improved drinking water, the
lower the environmental concern). Furthermore, we measure;
2 The regulation of the ‘dirty dozen’ persistent pollutants (Source: Johnson,
2013). As direct measures of air quality are hardly available for all countries,
we use the regulation of pollutants as a proxy. We expect this variable to have
a negative effect because it indicates less exposure to air pollution and thus
the absence of negative interest factors.
Table 3.1 Confirmatory factor analysis of awareness, efficacy and willingness.
Latent Item-ID Item-Text Factor Loadings

Construct
1993 2000 2010
Awareness In general, do you think . . . ?

1.1 . . . air pollution caused by cars . . . 1.00 fixed 1.00 fixed 1.00 fixed
1.2 . . . air pollution caused by industry . . . 1.21 *** 1.08 *** 1.02 ***
1.3 . . . pollution of country’s rivers, lakes and 1.18 *** 1.10 *** 1.01 ***
streams . . .
1.4 . . . pesticides and chemicals used in farming . . . 1.27 *** 1.14 *** 1.02 ***
1.5 . . . the rise in the world’s temperature caused by 1.25 *** 1.13 *** 1.10 ***
the ‘greenhouse effect’ [‘climate change’ since
2010] . . .
. . . is [extremely/very/somewhat/not very/not at all]
dangerous to the environment
Efficacy How much do you agree or disagree with each of these
statements?
2.1 We worry too much about the future of the 1.00 fixed 1.00 fixed 1.00 fixed
environment and not enough about prices and
jobs today
2.2 People worry too much about human progress 0.90 *** 0.88 *** 0.85 ***
harming the environment
2.3 In order to protect the environment [COUNTRY] 0.59 *** 0.65 *** 0.56 ***
needs economic growth
2.4 It is just too difficult for someone like me to do 0.71 *** 0.70 *** 0.69 ***
much about the environment
2.5 Modern science will solve our environmental 0.54 *** 0.64 *** 0.53 ***
problems with little change to our way of life
[Agree strongly/Agree/Neither agree nor disagree/
Disagree/Disagree strongly]
Willingness How willing would you be to . . . ?
3.1 . . . pay much higher prices . . . 1.00 fixed 1.00 fixed 1.00 fixed
3.2 . . . pay much higher taxes . . . 1.15 *** 1.03 *** 0.99 ***
3.3 . . . accept cuts in your standard of living . . . 0.97 *** 0.82 *** 0.80 ***
. . . in order to protect the environment? [Very willing
/fairly willing/neither willing nor unwilling/fairly
unwilling/very unwilling]
How much do you agree or disagree with each of these
statements?
3.4 I do what is right for the environment whether it 0.00 dropped 0.00 dropped 0.00 dropped
costs more money or takes more time . . .
[Agree strongly/Agree/Neither agree nor disagree/
Disagree/Disagree strongly]
Notes: *** p<.001; error correlations between items 1.1 and 1.2; 1.3 and 1.4 (2010 and 2000 only). Model fit statistics: 2010: RMSEA = 0.04, CFI = 0.97, TLI = 0.96,
SRMR = 0.03, CD = 0.99; 2000: RMSEA = 0.04, CFI = 0.97, TLI = 0.96, SRMR = 0.03, CD = 0.99; 1993: RMSEA = 0.04, CFI = 0.97, TLI = 0.96, SRMR = 0.03, CD = 0.99.
As positive interest factors we measure:
1 The percentage of protected terrestrial and marine areas (of the total ter-
ritorial area) (source: Millennium Development Goals Data Base, United
Nations); we expect this variable to have a positive effect in terms of envi-
ronmental quality and tourism.
2 The percentage of agricultural land use (of all land area) (source: World
Development Indicators, World Bank) and also expect this variable to have a
positive effect.
3 The percentage of renewable energy use of all energy consumption and we
expect a positive effect of this variable as well (source: authors’ calculations
from the Energy Use Data Base, United Nations).
As a control variable we include GDP per capita (at current prices) in US$1,000
in our model. We also tested for GDP growth as an additional control variable but
excluded it as it was not significant.
We derived our final dataset by list-wise deletion. Table 3.2 shows the coun-
tries, years and number of individuals included in the multilevel analysis, which
we present in the next section. In total we have data from 35 countries, 59 country
years and 54,086 individual respondents. Owing to a lack of information on many
country-level indicators for 1993, our sample contains only very few countries in
this first wave. However, a sufficiently large number of countries is included in
both 2000 and 2010 to allow not only for a cross-sectional but also a longitudinal
analysis, which investigates how changes over time affect the dependent variables
(see Methods).
Methods
Our data have a three-level structure with individuals at level 1, nested in coun-
try years (samples) at level 2, which in turn are nested in countries at level 3
(Schmidt-Catran and Fairbrother, 2016). This data structure constitutes a quasi-
panel at the level of countries and allows estimating multilevel models which
break down the effects of country-level variables into their cross-sectional and
longitudinal components. As individuals are not observed repeatedly, individ-
ual-level effects can only be estimated cross-sectionally. Malcolm Fairbrother
proposed the modelling approach that we use here (see Fairbrother, 2014 for a
detailed treatment of our approach) and also used it for the analysis of environ-
mental attitudes (2013).
As our dependent variables are factor scores, we treat them as continuous and
normally distributed variables and therefore use linear hierarchical regression
modelling. Basically, our model is a simple three-level model (with the structure
described above). However, it breaks down the effects of time-varying country-
level variables into cross-sectional (between) and longitudinal (within) effects
by means of a variable transformation. For each time-varying country-level
variable, the model contains two variables: one being the average value across
Table 3.2 Number of respondents per country–year used in statistical analysis (after
list-wise deletion).
Respondents
Country 1993 2000 2010
Hungary 1,122
Netherlands 1,480
Portugal 971
Argentina 671
Australia 1,419
Austria 941 612
Belgium 749
Bulgaria 1,033 849 511
Canada 882 532
Chile 1,451 736
Croatia 530
Czech Republic 1,112 693
Denmark 841
Finland 1,207 577
France 1,186
Germany 1,442 786
Israel 495
Japan 970 615
Korea (South) 1,200
Latvia 881 356
Lithuania 452
Mexico 1,166 456
New Zealand 956 704
Norway 1,257 851
Philippines 1,163 996
Russia 1,735 1,580 808
Slovak Republic 719
Slovenia 864 992 377
South Africa 1,606
Spain 705 1,284
Sweden 907 711
Switzerland 581 767
Turkey 1,019
United Kingdom 816 517
United States 1,339 1,033 875
Source: ISSP 1993, 2000, 2010.
waves (i.e. the between effect, capturing enduring cross-sectional differences

between countries); and the other being the original variable minus the cross-
time mean (i.e. the within effect, capturing changes over time, independently
of the cross-sectional differences in level). From an econometric perspective,
within effects should provide better estimates of causal effects because they are
based only on the variation that is observed within units over time. Note that the
described transformation to derive within components is equivalent to the fixed
effects transformation used in panel-data analysis (Andress et al., 2013, p. 126ff).
Therefore, within effects cannot be biased by unobserved heterogeneity between
the countries.
Results
Decription
Table 3.3 presents four regression models for each of three dependent variables.
We also estimated empty models which break down the total variances into their
components at each of the three levels. For dependent variable awareness we
observe about 11% of the variance at the level of countries and about 2% at
the level of country years. For dependent variable efficacy the values are very
similar; namely, about 12% and 2%, respectively. Thus in both cases there is
indeed a reasonable amount of between-country variance that can potentially
be explained. The variance observed at the level of country years is rather small
but still significant, which indicates that the models also have the potential to
explain changes over time. The variable ‘willingness to pay’ shows slightly dif-
ferent values. We find about 5% of the variance between countries and about 3%
between country years. Thus, compared with the other two dependent variables,
enduring differences between countries are rather small, while there are stronger
changes over time.
In the first set of models (M1), we include only a time trend but no individual-
or country-level variables. These models have a rather descriptive purpose and
reveal that there is no significant trend over time with regard to the awareness
of environmental threats and willingness to pay. Environmental efficacy, on the
other hand, shows a significant positive trend.1 In the second set of models (M2),
we included all individual-level variables. As we have standardized all dependent
variables, the effect size of each independent variable can be compared across
the three dependent variables. Education has a highly significant positive impact
on awareness, efficacy and willingness to pay but the effect appears to be much
stronger on the latter two. Age has a negative and significant effect on awareness
and efficacy, but a positive one on willingness to pay. The positive effect of age,
however, is relatively weak and disappears in the later models.
The relative income position within a country has no effect on awareness of
environmental threats but a significant positive effect on efficacy and willingness
to pay, with the latter effect being almost twice as strong as the former. Women
show significantly higher awareness and efficacy, but no significant difference
with men regarding willingness to pay. Respondents living in urban areas show
higher levels of awareness and also willingness to pay, but no significant difference
from residents in non-urban areas when it comes to efficacy. Employment status
affects perceived efficacy and willingness to pay. With regard to the latter, we find
that unemployed respondents are less willing to pay compared with respondents
who are employed or who are not in the labour force. On the other hand, employed
respondents show higher perceived efficacy compared with unemployed and non-
working respondents. Finally, post-materialism has a positive effect on all three
dependent variables. The effect is strongest on the willingness to pay and weakest
on awareness.
The individual-level variables included in the model M2 have quite good
explanatory power with regard to efficacy and willingness to pay. Of the variance
in efficacy, compared with the M1 models, the M2 models reduce the unexplained
variance (the variance components) by about 8% of the individual level, about
28% of the country level and about 3% of the country-year level. With regard to
willingness to pay they explain about 5%, 12% and 19%, respectively, while the
figures for awareness are about 2%, 1% and 0%.
In the M3 models, we add GDP/C (in US$1,000). Note that our modelling
approach yields two estimates for each variable, a (longitudinal) within effect
(WE) and a (cross-sectional) between effect (BE). Economic wealth has a sig-
nificant positive effect on the awareness of environmental threats. We are able to
identify this effect cross-sectionally (-0.0145***) and longitudinally (-0.0064*),
which we see as a good indicator of the actual presence of this effect. Efficacy is
positively influenced by economic wealth but this effect can only be identified
cross-sectionally (0.0167***) and not longitudinally.
With regard to the willingness to pay we find a positive effect of economic
wealth, which again can be identified cross-sectionally (0.0086***) and longi-
tudinally (0.0059*). The amount of explained variance at the country level is
about 41% for awareness (increase compared with M1: 46%); 73%, for effi-
cacy (increase compared with M1: 44%); and about 35% for willingness to pay
(increase compared with M1: 23%). At the country-year level the explained vari-
ances are 2% for awareness (increase compared with M1: 0.5%), 10% for efficacy
(increase compared with M1: 7%) and 30% for willingness-to-pay (increase com-
pared with M1: 11%).
In the last set of models (M4) we enter the independent country-level variables
of interest. In general, we find only a very limited number of significant effects.
Threat awareness increases with a rise in renewable energy use. This effect can
be found only within countries (1.65*) but not between them. Thus, within a
given country, awareness increases when the percentage of renewable energy use
increases but this does not pertain between countries. Between countries we find a
significant negative effect of pesticide regulation on efficacy (-0.009*). We find a
positive effect of access to improved drinking water, which can be identified cross-
sectionally (0.0415**) and longitudinally (0.0347*). Finally, we find a significant
positive effect of within-country changes in protected terrestrial and marine areas
(0.005*). As for willingness to pay, we find a significant positive longitudinal
effect of pesticide regulations. In this last set of models, the explained variance
between countries does not increase by much but the variance explained at the
level of country years does: 10 percentage points for awareness; 16 for efficacy;
Table 3.3 Multilevel models of awareness, efficacy and willingness with breakdown of cross-sectional and longitudinal effects.
M1 M2 M3 M4
Aware Efficacy Willingness Aware Efficacy Willingness Aware Efficacy Willingness Aware Efficacy Willingness
Individual-level
variables
Years of 0.0123*** 0.0500*** 0.0453*** 0.0123*** 0.0499*** 0.0451*** 0.0123*** 0.0498*** 0.0451***
schooling
Age -0.0022*** -0.0037*** 0.0005* -0.0022*** -0.0037*** 0.0004 -0.0022*** -0.0038*** 0.0005
Household -0.0092 0.0453*** 0.0785*** -0.0089 0.0458*** 0.0784*** -0.0090 0.0464*** 0.0782***
Income
(relative)
Gender 0.1550*** 0.0947*** 0.0044 0.1549*** 0.0947*** 0.0044 0.1550*** 0.0946*** 0.0044
(women = 1)
Residency (big 0.1052*** -0.0081 0.0326*** 0.1048*** -0.0078 0.0333*** 0.1047*** -0.0073 0.0334***
city = 1)
Employment
status
Not in labor Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref.
force
Employed -0.0019 0.0658*** 0.0118 -0.0019 0.0656*** 0.0117 -0.0019 0.0653*** 0.0116
Unemployed -0.0014 -0.0157 -0.0401* -0.0018 -0.0156 -0.0399* -0.0021 -0.0156 -0.0402*
Post- 0.0686*** 0.1646*** 0.1931*** 0.0689*** 0.1645*** 0.1928*** 0.0689*** 0.1646*** 0.1928***
materialism
Year
1993 Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref.
2000 0.0554 0.1114* -0.1103 0.0438 0.1194* -0.1137 0.0718 0.0938* -0.1278* 0.0789 0.0546 -0.1839**
2010 -0.0279 0.2467 *** -0.0213 -0.0163 0.1418** -0.1787* 0.1117 0.1592** -0.2785*** 0.0696 0.0460 -0.3128***
Country-level
variables
GDP/C (in -0.0145*** 0.0167*** 0.0086*** -0.0129** 0.0143*** 0.0109**
US$1,000)
[BE]
GDP/C (in -0.0064* -0.0032 0.0059* -0.0061* -0.0021 0.0050*
US$1,000)
[WE]
Pesticide -0.0085 -0.0090* -0.0057
regulation
[BE]
Pesticide 0.0015 0.0085 0.0258**
regulation
[WE]
Improved -0.0150 0.0415** -0.0086
drinking
water [BE]
Improved 0.0120 0.0347* -0.0039
drinking
water [WE]
(continued)
M1 M2 M3 M4
Aware Efficacy Willingness Aware Efficacy Willingness Aware Efficacy Willingness Aware Efficacy Willingness
Protected 0.0005 -0.0049 -0.0011

terrestrial
and marine
areas [BE]
Protected 0.0028 0.0050* -0.0042
terrestrial
and marine
areas [WE]
Agricultural -0.0020 -0.0021 0.0000
land (% of all
land) [BE]
Agricultural 0.0155 0.0086 0.0072
land (% of all
land) [WE]
Renewable 0.0047 0.0095 0.0083
energy (% of
all energy)
[BE]
Renewable 0.0165* 0.0060 0.0041
energy (% of
all energy)
[WE]
Constant -0.0222 -0.2366** -0.0105 -0.2452** -0.7821*** -0.6032*** -0.0022 -1.1396*** -0.7361*** 1.6573 -4.8199** 0.1823
Variance
components
Var(Country) 0.1144*** 0.1286*** 0.0622*** 0.1205*** 0.0923*** 0.0549*** 0.0671*** 0.0351*** 0.0402*** 0.0656*** 0.0263*** 0.0456***
Var(Year) 0.0186*** 0.0091*** 0.0213*** 0.0183*** 0.0088*** 0.0173*** 0.0182*** 0.0082*** 0.0149*** 0.0148*** 0.0049*** 0.0096***
Var(Individual) 0.8656*** 0.8450*** 0.9322*** 0.8508*** 0.7798*** 0.8823*** 0.8508*** 0.7798*** 0.8823*** 0.8508*** 0.7798*** 0.8823***
Statistics
N (Country) 35 35 35 35 35 35 35 35 35 35 35 35
N (Country- 59 59 59 59 59 59 59 59 59 59 59 59
year)
N(Individual) 54,086 54,086 54,086 54,086 54,086 54,086 54,086 54,086 54,086 5,4086 54,086 54,086
AIC 145,950.42 144,635.33 14,9942.14 145,034.74 140,299.88 146,974.33 145,020.74 140,271.92 146,965.14 145,034.53 140,270.06 146,976.72
BIC 1460,03.81 144,688.72 149,995.53 145,159.32 140,424.45 147,098.91 145,163.11 140,414.3 147,107.52 145,265.88 140,501.41 147,208.08
Notes: * p<.05, ** p<.01, *** p<.001 (one-sided tests). [BE] = between-country effect, [WE] = within-country effect
Source: ISSP 1993, 2000, 2010.
and 25 for willingness to pay. Although we do not find many significant effects,
this result might indicate that the contextual-level effect on environmental attitudes
might be better explained and investigated by means of longitudinal analysis within
countries rather than by cross-sectional comparisons among them. One reason
might be that the unobserved heterogeneity between countries is simply too large.
Longitudinal analyses within countries, on the other hand, provide effects that hold
any time-constant country-level characteristics constant.
Discussion
The results allow an interpretation of individual determinants of environmental
concern and national indicators of positive and negative context conditions at the
country level. First, as suggested by Inglehart (2008, 2002, 1971), post-materialists
consistently show more concern for the environment than respondents who are
more materially oriented, net of individual and national affluence and other condi-
tions. This finding underlines the argument for a strong value basis of individual
environmental concern, which is significantly independent of economic changes
and interest factors. Not in line with our assumptions is the finding that the effect
of post-materialism is stronger for efficacy and the willingness-to-pay measure
than for individual awareness of environmental threats. One possible explanation
for this is that awareness depends much more on external factors such as the actual
physical threat or media coverage of environmental issues. Post-materialists play
a positive role in the awareness of these issues, but they are even more relevant
when it comes to decisions on behaviour intentions.
Figure 3.1 supports this idea when looking at the national level. With respect
to awareness, it seems that non-Western countries from South America, Oceania
or Asia have the highest levels of awareness but low post-materialist values.
In fact, these are the regions with the greatest vulnerability to environmental
threats and massive environmental changes, such as the depletion of rainforests
(South America), air pollution (Asia) or risk of flooding (Oceania). However,
high population means of post-materialism play a positive role for environmental
efficacy and willingness to pay, whereas Eastern Europe clearly sticks out from
the European pattern (all filled symbols). This finding very much supports the
idea of a strong influence of personal values developed over generations in the
wealthier countries, whereas Eastern Europe has only had very recent centuries
to follow that path.
Second, our results support the hypothesis that positive and negative con-
text conditions influence environmental concern at the individual level. Higher
education equips a person with the knowledge about how to avoid exposure to
environmental threats, or how to discover opportunities for financial investment
returns from renewable energy. Also, in line with the interest approach, is the find-
ing that younger respondents are more aware of environmental issues and engage
more with environmental efficacy. Nowadays, more and more information is
provided via new media and so younger cohorts have a major advantage in being
able to exercise individual control about what concerns their environment and the
0.4
0.2
Awareness
0
−0.2
−0.4
0.4 0.6 0.8 1 1.2 Northern Europe

Eastern Europe
0.4 Southern Europe
Western Europe
0.2
North America
Efficacy
0 South America
Oceania
−0.2 Asia
South Africa
−0.4 Israel
Turkey
0.4 0.6 0.8 1 1.2
Fitted values
0.5
Willingness to pay
−0.5
0.4 0.6 0.8 1 1.2

Post-materialism
Figure 3.1 Post-materialism and awareness, efficacy, and willingness-to-pay by geographical

regions. Source: ISSP 2010 (weighted data).
Notes: Northern Europe = Denmark, Finland, Norway, Sweden; Eastern Europe = Hungary, Poland,
Bulgaria, Croatia, Czech Republic, Latvia, Lithuania, Russia, Slovakia, Slovenia; Southern Europe =
Italy, Portugal, Spain; Western Europe = Ireland, Netherlands, Austria, Belgium, France, Germany,
Switzerland, Great Britain; North America = USA; Canada; South America = Argentina, Chile,
Mexico, Oceania = Australia, New Zealand; Asia = Taiwan, Japan, South Korea, Philippines.
means to do something about it. Negative interests arise out of individual exposure
to environmental threats and risks induced by environmental policies in the labour
market. In line with our assumptions, respondents living in urban areas with higher
population densities, lower air quality and increasingly high noise levels are more
aware of environmental problems, and are willing to pay more than rural residents.
On the dimension of efficacy, the motivation of urban people to be informed and
become active in changing their personal situation might rate equally with citizens’
knowledge and attachment to nature in rural areas. Furthermore, awareness is
equally high independent of the respondents’ employment status, but unemployed
citizens are more concerned about potential risks to the labour market induced by
environmental policies, and this shows up in lower levels of efficacy and willing-
ness to pay (net of household income).

At this national level, our results for interest factors are mixed. The posi-
tive effect of changes in investment returns from renewable energy and
benefits from protected terrestrial and marine areas, on awareness and efficacy,
respectively, lends some support to our assumption that positive interests in
developing, as well as developed, countries influence environmental concern.
Arguably, this interpretation is rather weak when considering that the high vis-
ibility and need for decentralized local agency of small-scale wind-power and
photovoltaic installations influence respondents’ awareness rather than positive
interests from vague investment opportunities. Furthermore, the evidence of
negative interests by measures of environmental degradation shows the oppo-
site of what we would expect, with a higher concern in terms of environmental
efficacy when the quality of drinking water rises. Support for the influence of
environmental degradation on environmental concern is demonstrated by the
indicator for air quality, where higher levels of regulation decrease environmental
efficacy. Again, the positive sign of pesticide regulation on personal willing-
ness to pay might also be interpreted as similar to our discussion on renewable
energy. Respondents might wish to see public actors as responsible for pro-
viding environmental goods before they themselves act. They see progress in
renewable energy installations and pesticide regulation as ‘good examples’ and
want to contribute their share.
Third, affluence at the individual level as well as the national level consistently
raises environmental efficacy and willingness to pay when income rises and GDP
per capita grows. This finding is in line with the affluence theory stating that eco-
nomic wealth should have a positive effect on environmental concern. However,
the same measures also show a negative influence on environmental awareness.
As expected, we find the strongest effect size of family income on environmental
concern for willingness to pay, and the smallest for awareness of environmental
threats. At the national level, GDP per capita shows only minimal differences in
the effect size across the components of environmental concern, with the strongest
effect for environmental efficacy. This and the finding that the effect size of fam-
ily income remains stable even when introducing GDP per capita, indicates that
personal income is relatively independent of actual changes in national econo-
mies. Furthermore, the decreasing effect size of GDP per capita when introducing
positive and negative context conditions indicates that part of the effect of afflu-
ence is explained by environmental degradation and opportunities for positive
interests within the countries.
Our results clearly demonstrate the importance of measuring environmental

concern as a multidimensional concept with three components: awareness, effi-
cacy, and willingness to pay. The finding that affluence measured for both levels
of investigation is positively related for two elements, but has a negative effect on
awareness, is in line with the most recent results from Marquardt-Pyatt (2012). It
also brings into question findings of a consistently positive association between
affluence and environmental concern, measured only two-dimensionally by will-
ingness to pay and efficacy (e.g., Conroy and Emerson, 2014; Fairbrother, 2013;
Franzen and Vogl, 2013b; Kemmelmeier et al., 2002; Nawrotzki and Pampel,
2013). Proponents of this approach generally argue that willingness to pay is
the most reliable measure of environmental concern since it takes account of the
intra-personal conflict between environmental behavioural intentions and per-
sonal resources (Gelissen, 2007; Kidd and Lee, 1997; Meyer and Liebe, 2010).
However, this argument seems especially suitable for higher-income respondents
and accords with the economic idea of an increasing demand for environmental
goods, as proposed by the affluence approach. Yet scholars stressing the impor-
tance of environmental degradation as a trigger for environmental concern argue
that awareness better reflects the environmental concerns of disadvantaged groups
and nations (Arbuthnott et al., 2012; Brechin, 1999). The importance of consider-
ing the tripartite conceptualization of environmental concern is also underlined by
the finding that post-materialist values show no consistent effects across the three
measures and that, even more commonly, studies of demographic variables such
as gender or age (see Gifford and Nilsson, 2014 for a review) reveal no significant
influence on the personal willingness to pay for environmental goods.
Conclusion
This article contributes to the literature on environmental sociology by providing
a discussion and empirical analyses of different dimensions of environmental con-
cern. It addresses the need: (1) to develop a more thorough understanding of the
variation in environmental concern cross-nationally and within societies, using
more general value approaches (Dietz et al., 2005; Milfont et al., 2006; Schultz
and Zelezny, 1998) and more recent findings from environmental psychology
(Hansla et al., 2013; Kaiser and Scheuthle, 2003; Milfont et al., 2006; Milfont
and Duckitt, 2010; Milfont and Gouveia, 2006; Wiseman and Bogner, 2003);
(2) to identify and discuss the context of positive and negative context conditions
as sources of personal interests and vulnerabilities influencing environmental con-
cern (Gifford, 2011; Marshall et al., 2005; Schaffrin, 2011b; Whitmarsh, 2009);
and (3) to demonstrate that the determinants have different effects on different
components of environmental concern across countries (Diekmann and Franzen,
1999; Dunlap and York, 2008; Marquart-Pyatt, 2012; Schaffrin, 2011b).
This research examined environmental concern using the ISSP Environment
(1993, 2000 and 2003) datasets for a multilevel, comparative analysis consider-
ing national variations over time. The results confirm the basic assumptions of
the OPSV approach stressing that post-materialist values, as well as a variety of

interest factors derived from both positive and negative context conditions, influ-
ence environmental concern. They also reveal a positive relationship between
national wealth or family income, with environmental efficacy and willingness-
to-pay measures, as anticipated by the affluence approach. For the individual
awareness of environmental threats, however, we found a negative effect of
affluence measures. In general, the analyses stress the importance of applying
multilevel analyses as well as longitudinal data, in order to account for changes

in contextual conditions. They also make a case for stronger consideration of
more general values and a variety of contextual factors stimulating positive and
negative interests. What has become clear is that environmental concern is a
multidimensional concept which should be operationalized via at least three
components: awareness, efficacy, and willingness to pay. As we have demon-
strated, environmental concern is not a unique Western-European phenomenon
but is widespread across a number of countries. In fact, European populations
are highly heterogeneous with regard to the awareness of environmental threats,
environmental efficacy and their willingness to pay for environmental goods.
The empirical application so far is limited to the available number of countries
and waves of the ISSP. A larger number of countries and a longer time series would
allow the estimation of more robust results. Further studies using similar survey and
longitudinal data should apply concepts which are comparable to ours in order to
elaborate on the role of contextual factors. With the limited number of context fac-
tors applied in this study, we are only able to grasp whether, but not how, regional
exposure or interests in environmental conditions would influence environmental
concern. Furthermore, the analysis would certainly benefit from additional meas-
ures of personal-value dimensions, such as the collectivism scale, in order to address
the strong heterogeneity of environmental concern across European populations.
The interest factors we used on the individual, as well as the national level, were
rather crude indicators but should measure vulnerabilities, exposure and benefits
more directly. In general, it seems that individual-level interests are much more
influential with respect to environmental concern than macrolevel changes in eco-
nomic or environmental conditions. This is a finding that has also been substantiated
by Marshall et al. (2005), who stressed the role of community-level exposure rather
than national changes. This should be given further attention in future studies.
Unfortunately, our analysis had to work within the limitations of available questions
and the data of the ISSP. However, since these first findings reveal a positive result,
we encourage researchers and practitioners to apply our approach to comparative
studies of environmental concern on a broader and more thorough empirical basis.
Note
1 Please note that our sample is only of limited value for the estimation of a general time
trend because the set of countries included is both selective and different for each wave.
References
Adeola, F., 2004. Environmentalism and Risk Perception: Empirical Analysis of Black and
White Differentials And Convergence. Soc. Nat. Resour. 17, 911–939. doi:10.1080/
08941920490505329
Andress, H.-J., Golsch, K., Schmidt, A., 2013. Applied Panel Data Analysis for Economic
and Social Surveys. Springer, Berlin, Heidelberg.
Arbuthnott, K.D., Devoe, D., Lawrie, T., 2012. Economic Development and Environmental
Intentions: A Comment on Liu and Sibley (2011). Anal. Soc. Issues Public Policy 12,
209–215. doi:10.1111/j.1530-2415.2012.01289.x
Best, H., 2009. Eating Comes First and Then Morality? A Field Experiment on the Low-
Cost Hypothesis and the Model of Frame Selection. Z. Soziol. 38, 131–151.
Best, H., Mayerl, J., 2013a. Values, Beliefs, Attitudes: An Empirical Study on the Structure
of Environmental Concern and Recycling Participation. Soc. Sci. Q. 94, 691–714.
doi:10.1111/ssqu.12010
Best, H., Mayerl, J., 2013b. Values, Beliefs, Attitudes: An Empirical Study on the Structure
of Environmental Concern and Recycling Participation. Soc. Sci. Q. 94, 691–714.
doi:10.1111/ssqu.12010
Bowen, A., Fankhauser, S., 2011. The Green Growth Narrative: Paradigm Shift or Just
Spin? Glob. Environ. Change-Hum. Policy Dimens. 21, 1157–1159. doi:10.1016/
j.gloenvcha.2011.07.007
Brechin, S.R., 1999. Objective Problems, Subjective Values, and Global Environmentalism:
Evaluating the Postmaterialist Argument and Challenging a New Explanation. Soc. Sci.
Q. 80, 793–809.
Brechin, S.R., Kempton, W., 1997. Beyond Postmaterialist Values: National Versus
Individual Explanations of Global Environmentalism. Soc. Sci. Q. 78, 16–20.
Brechin, S.R., Kempton, W., 1994. Global Environmentalism: A Challenge to the
Postmaterialism Thesis. Soc. Sci. Q. 75, 245–269.
Brechin, S.R., Mohai, P., Simones, S., 2005. Environmentalism and Postmaterialism: A
Dead End? Presented at the Annual Meeting of the American Sociological Association,
Philadephia, PA.
Brody, S.D., Zahran, S., Vedlitz, A., Grover, H., 2008. Examining the Relationship
Between Physical Vulnerability and Public Perceptions of Global Climate Change in
the United States. Environ. Behav. 40, 72–95. doi:10.1177/0013916506298800
Carfi, D., Schiliro, D., 2012. A Coopetitive Model for the Green Economy. Econ. Model.
29, 1215–1219. doi:10.1016/j.econmod.2012.04.005
Conroy, S.J., Emerson, T.L.N., 2014. A Tale of Trade-Offs: The Impact of Macroeconomic
Factors on Environmental Concern. J. Environ. Manage. 145, 88–93. doi:10.1016/
j.jenvman.2014.05.033
Diekmann, A., Franzen, A., 1999. The Wealth of Nations and Environmental Concern.
Environ. Behav. 31, 540–549. doi:10.1177/00139169921972227
Diekmann, A., Preisendörfer, P., 2003. Green and Greenback: The Behavioral Effects of
PF Environmental Attitudes in Low-Cost and High-Cost Situations. Ration. Soc. 15,
441–472. doi:10.1177/1043463103154002
Dietz, T., Fitzgerald, A., Shwom, R., 2005. Environmental Values. Annu. Rev. Environ.
Resour., Annual Review of Environment and Resources 30, 335–372. doi:10.1146/
annurev.energy.30.050504.144444
Drori, I., Yuchtman-Ya’ar, E., 2002. Environmental Vulnerability in Public Perceptions and
Attitudes: The Case of Israel’s Urban Centers. Soc. Sci. Q. 83, 53–63. doi:10.1111/1540-
6237.00070
Dunlap, R.E., Jones, R., 2002. Environmental Concern: Conceptual and Measurement
Issues, in: Dunlap, R.E., Michelson, W. (Eds.), Handbook of Environmental Sociology.
Greenwood Press, Westport, London, pp. 485–524.
Dunlap, R.E., Mertig, A., 1997. Global Environmental Concern: An Anomaly for Postma
terialism. Soc. Sci. Q. 78, 24–29.

Dunlap, R.E., York, R., 2008. The Globalization of Environmental Concern and the Limits
of the Postmaterialist Values Explanation: Evidence from Four Multinational Surveys.
Sociol. Q. 49, 529–563. doi:10.1111/j.1533-8525.2008.00127.x
Escobar, A., 2006. An Ecology of Difference: Equality and Conflict in a Globalized World.
Eur. J. Anthropol. 47, 120–137.
Faiers, A., Cook, M., Neame, C., 2007. Towards a Contemporary Approach for Under
standing Consumer Behaviour in the Context of Domestic Energy Use. ENERGY
POLICY 35, 4381–4390. doi:10.1016/j.enpol.2007.01.003
Fairbrother, M., 2014. Two Multilevel Modeling Techniques for Analyzing Comparative
Longitudinal Survey Datasets. Polit. Sci. Res. Methods 2, 119–140.
Fairbrother, M., 2013. Rich People, Poor People, and Environmental Concern: Evidence
across Nations and Time. Eur. Sociol. Rev. 29, 910–922. doi:10.1093/esr/jcs068
Franzen, A., 2003. Environmental Attitudes in International Comparison: An Analysis of
the ISSP Surveys 1993 and 2000. Soc. Sci. Q. 84, 297–308. doi:10.1111/1540-6237.
8402005
Franzen, A., Meyer, R., 2010. Environmental Attitudes in Cross-National Perspective:
A Multilevel Analysis of the ISSP 1993 and 2000. Eur. Sociol. Rev. 26, 219–234.
doi:10.1093/esr/jcp018
Franzen, A., Vogl, D., 2013a. Acquiescence and the Willingness to Pay for Environmental
Protection: A Comparison of the ISSP, WVS, and EVS. Soc. Sci. Q. 94, 637–659.
doi:10.1111/j.1540-6237.2012.00903.x
Franzen, A., Vogl, D., 2013b. Two Decades of measuring Environmental Attitudes: A
Comparative Analysis of 33 Countries. Glob. Environ. Change-Hum. Policy Dimens.
23, 1001–1008. doi:10.1016/j.gloenvcha.2013.03.009
Gelissen, J., 2007. Explaining Popular Support for Environmental Protection: A Multi
level Analysis of 50 Nations. Environ. Behav. 39, 392–415. doi:10.1177/001391650
6292014
Gifford, R., 2011. The Dragons of Inaction Psychological Barriers that Limit Climate
Change Mitigation and Adaptation. Am. Psychol. 66, 290–302. doi:10.1037/a0023566
Gifford, R., Nilsson, A., 2014. Personal and Social Factors that Influence Pro-Environmental
Concern and Behaviour: A review. Int. J. Psychol. 49, 141–157. doi:10.1002/ijop.12034
Gifford, R., Scannell, L., Kormos, C., Smolova, L., Biel, A., Boncu, S., Corral, V.,
Guentherf, H., Hanyu, K., Hine, D., Kaiser, F.G., Korpela, K., Lima, L.M., Mertig,
A.G., Garcia Mira, R., Möser, G., Passafaro, P., Pinheiro, J.Q., Saini, S., Sako, T.,
Sautkina, E., Savina, Y., Schmuck, P., Schultz, W., Sobeck, K., Sundblad, E.-L., Uzzell,
D., 2009. Temporal Pessimism and Spatial Optimism in Environmental Assessments:
An 18-Nation Study. J. Environ. Psychol. 29, 1–12. doi:10.1016/j.jenvp.2008.06.001
Givens, J.E., Jorgenson, A.K., 2013. Individual Environmental Concern in the World
Polity: A Multilevel Analysis. Soc. Sci. Res. 42, 418–431. doi:10.1016/j.ssresearch.
2012.10.005
Givens, J.E., Jorgenson, A.K., 2011. The Effects of Affluence, Economic Development,
and Environmental Degradation on Environmental Concern: A Multilevel Analysis.
Organ. Environ. 24, 74–91. doi:10.1177/1086026611406030
Goksen, F., Adaman, F., Zenginobuz, E., 2002. On Environmental Concern, Willingness to
Pay, and Postmaterialist Values: Evidence from Istanbul. Environ. Behav. 34, 616–633.
doi:10.1177/0013916502034005003
Groot, J.I.M. de, Steg, L., 2007. Value Orientations and Environmental Beliefs in five
Countries: Validity of an Instrument to Measure Egoistic, Altruistic and Biospheric Value

Orientations. J. Cross-Cult. Psychol. 38, 318–332. doi:10.1177/0022022107300278
Grossman, G., Krueger, A., 1995. Economic Growth and the Environment. Q. J. Econ. 110,
353–377. doi:10.2307/2118443
Hansla, A., Gaerling, T., Biel, A., 2013. Attitude toward Environmental Policy Measures
Related to Value Orientation. J. Appl. Soc. Psychol. 43, 582–590. doi:10.1111/j.1559-
1816.2013.01038.x
Homer, P., Kahle, L., 1988. A Structural Equation Test of the Value Attitudes Behavior
Hierarchy. J. Pers. Soc. Psychol. 54, 638–646. doi:10.1037//0022-3514.54.4.638
Inglehart, R., 2008. Changing Values among Western Publics from 1970 to 2006. WEST
Eur. Polit. 31, 130–146. doi:10.1080/01402380701834747
Inglehart, R., 2002. Human Values and Social Change: Findings from the Values
Surveys. Comp. Sociol. 1 (Introduction to Special Issue), 231–233. doi:10.1163/1569
13302100418529
Inglehart, R., 1995. Public Support for Environmental Protection: Objective Problems
and Subjective Values in 43 Societies. PS-Polit. Sci. Polit. 28, 57–72. doi:10.2307/
420583
Inglehart, R., 1971. The Silent Revolution in Europe: Intergenerational Change in Post-
Industrial Societies, Am. Pol. Sci. Rev. 65 (December), 991–1017. doi:10.1017/S000
3055406392568
Inglehart, R., Abramson, P., 1999. Measuring Postmaterialism. Am. Polit. Sci. Rev. 93,
665–677. doi:10.2307/2585581
Israel, D., 2004. International Support for Environmental Protection. Environ. Dev. Econ.
9, 757–780. doi:10.1017/S1355770X0400169X
ISSP Research Group, 2012. International Social Survey Programme: Environment III -
ISSP 2010. GESIS Data Arch. Cologne ZA5500 data file version 2.0.0. doi:10.4232/
1.11418
ISSP Research Group, 2003. International Social Survey Programme: Environment II - ISSP
2000. GESIS Data Arch. Cologne ZA3440 Datenfile Version 1.0.0. doi:10.4232/1.3440
ISSP Research Group, 1995. International Social Survey Programme: Environment I -
ISSP 1993 GESIS Data Archive, Cologne. doi:10.4232/1.2450
Jaenicke, M., 2012. ‘Green Growth’: From a Growing Eco-Industry to Economic
Sustainability. Energy Policy 48, 13–21. doi:10.1016/j.enpol.2012.04.045
Johnson, L., 2013. National Regulatory Actions on the ‘Dirty Dozen’ POPs through the
Stockholm Convention. Yale Center for Environmental Law and Policy, New Haven.
Available at: http://uneplive.unep.org/media/docs/global/gl/gl_agriculture_2014.xls
Jorgenson, A.K., Clark, B., 2012. Are the Economy and the Environment Decoupling? A
Comparative International Study, 1960–2005. Am. J. Sociol. 118, 1–44.
Kaiser, F., Scheuthle, H., 2003. Two Challenges to a Moral Extension of the Theory of
Planned Behavior: Moral Norms and Just World Beliefs in Conservationism. Personal.
Individ. Differ. 35, 1033–1048. doi:10.1016/S0191-8869(02)00316-1
Kemmelmeier, M., Krol, G., Kim, Y., 2002. Values, Economics, and Pro-Environmental
Attitudes in 22 Societies. CROSS-Cult. Res. 36, 256–285. doi:10.1177/106971020
36003004
Kidd, Q., Lee, A., 1997. Postmaterialist Values and the Environment: A Critique and
Reappraisal. Soc. Sci. Q. 78, 1–15.
Kikuchi, R., 2011. Environmental and Socio-Economic Factors in Carbon Offsets: An
Approach to Sustainable Management and Planning in Climate Change Strategy.
J. Environ. Plan. Manag. 54, 355–367. doi:10.1080/09640568.2010.506084

Knight, K.W., Messer, B.L., 2012. Environmental Concern in Cross-National Perspective:
The Effects of Affluence, Environmental Degradation, and World Society. Soc. Sci. Q. 93,
521–537. doi:10.1111/j.1540-6237.2012.00846.x
Macias, T., Nelson, E., 2011. A Social Capital Basis for Environmental Concern: Evidence
from Northern New England. RURAL Sociol. 76, 562–581. doi:10.1111/j.1549-
0831.2011.00063.x
Maloney, M., Ward, M., 1973. Ecology:– Let’s Hear from the People: Objective Scale
for Measurement of Ecological Attitudes and Knowledge. Am. Psychol. 28, 583–586.
doi:10.1037/h0034936
Maloney, M., Ward, M., Braucht, G., 1975. Psychology in Action: Revised Scale for
Measurement of Ecological Attitudes and Knowledge. Am. Psychol. 30, 787–790.
doi:10.1037/h0084394
Marquart-Pyatt, S.T., 2012. Contextual Influences on Environmental Concerns Cross-
Nationally: A Multilevel Investigation. Soc. Sci. Res. 41, 1085–1099. doi:10.1016/j.
ssresearch.2012.04.003
Marquart-Pyatt, S.T., 2008. Are there Similar Sources of Environmental Concern?
Comparing Industrialized Countries. Soc. Sci. Q. 89, 1312–1335. doi:10.1111/j.1540-
6237.2008.00567.x
Marquart-Pyatt, S.T., 2007. Concern for the Environment among General Publics: A Cross-
National Study. Soc. Nat. Resour. 20, 883–898. doi:10.1080/08941920701460341
Marshall, B., Picou, J., Bevc, C., 2005. Ecological Disaster as Contextual Transformation:
Environmental Values in a Renewable Resource Community. Environ. Behav. 37,
706–728. doi:10.1177/0013916505275310
Martínez-Alier, J., 2002. The Environmentalism of the Poor: A Study of Ecological
Conflicts and Valuation. Edward Elgar, Cheltenham, UK.
Meyer, R., Liebe, U., 2010. Are the Affluent Prepared to Pay for the Planet? Explaining
Willingness To Pay for Public and Quasi-Private Environmental Goods in Switzerland.
Popul. Environ. 32, 42–65. doi:10.1007/s11111-010-0116-y
Milfont, T.L., Duckitt, J., 2010. The Environmental Attitudes Inventory: A Valid and
Reliable Measure to Assess the Structure of Environmental Attitudes. J. Environ.
Psychol. 30, 80–94. doi:10.1016/j.jenvp.2009.09.001
Milfont, T.L., Duckitt, J., 2004. The Structure of Environmental Attitudes: A First- and
Second-Order Confirmatory Factor Analysis. J. Environ. Psychol. 24, 289–303.
doi:10.1016/j.jenvp.2004.09.001
Milfont, T.L., Duckitt, J., Cameron, L.D., 2006. A Cross-Cultural Study of Environmental
Motive Concerns and their Implications for Pro-Environmental Behavior. Environ.
Behav. 38, 745–767. doi:10.1177/0013916505285933
Milfont, T.L., Duckitt, J., Wagner, C., 2010a. A Cross-Cultural Test of the Value-Attitude-
Behavior Hierarchy. J. Appl. Soc. Psychol. 40, 2791–2813. doi:10.1111/j.1559-1816.
2010.00681.x
Milfont, T.L., Gouveia, V.V., 2006. Time Perspective and Values: An Exploratory
Study of their Relations to Environmental Attitudes. J. Environ. Psychol. 26, 72–82.
doi:10.1016/j.jenvp.2006.03.001
Milfont, T.L., Sibley, C.G., Duckitt, J., 2010b. Testing the Moderating Role of the Com
ponents of Norm Activation on the Relationship Between Values and Environmental
Behavior. J. CROSS-Cult. Psychol. 41, 124–131. doi:10.1177/0022022109350506
Mostafa, M.M., 2013. Wealth, Post-Materialism and Consumers’ Pro-Environmental
Intentions: A Multilevel Analysis across 25 Nations. Sustain. Dev. 21, 385–399.

doi:10.1002/sd.517
Nawrotzki, R.J., Pampel, F.C., 2013. Cohort Change and the Diffusion of Environmental
Concern: A Cross-National Analysis. Popul. Environ. 35, 1–25. doi:10.1007/s11111-
012-0182-4
O’Connor, R., Bord, R., Yarnal, B., Wiefek, N., 2002. Who Wants to Reduce Greenhouse
Gas Emissions? Soc. Sci. Q. 83, 1–17. doi:10.1111/1540-6237.00067
Pampel, F.C., 2014. The Varied Influence of SES on Environmental Concern. Soc. Sci. Q.
95, 57–75. doi:10.1111/ssqu.12045
Pampel, F.C., Hunter, L.M., 2012. Cohort Change, Diffusion, and Support for Environmental
Spending in the United States. Am. J. Sociol. 118, 420–448. doi:10.1086/666506
Peacock, W., Brody, S., Highfield, W., 2005. Hurricane Risk Perceptions among Florida’s
Single Family Homeowners. Landsc. URBAN Plan. 73, 120–135. doi:10.1016/j.
landurbplan.2004.11.004
Pellow, D.N., Brulle, R.J. (Eds.), 2005. Power, Justice, and the Environment: A Critical
Appraisal of the Environmental Justice Movement. MIT Press, Cambridge, MA.
Rosenberg, M.J., Hovland, C.I., 1960. Cognitive, Affective and Behavioral Components of
Attitudes, in: Hovland, C.I., Rosenberg, M.J. (Eds.), Attitude Organization and Change.
Yale University Press, New Haven, CT, pp. 1–14.
Sardianou, E., 2007. Estimating Energy Conservation Patterns of Greek Households.
Energy Policy 35, 3778–3791. doi:10.1016/j.enpol.2007.01.020
Sarigoellue, E., 2009. A Cross-Country Exploration of Environmental Attitudes. Environ.
Behav. 41, 365–386. doi:10.1177/0013916507313920
Schaffrin, A., 2014. The New Social Risks and Opportunities of Climate Change, in:
Fitzpatrick, T. (Ed.), International Handbook on Social Policy & the Environment.
Edward Elgar, pp. 3–61.
Schaffrin, A., 2011a. No Measure without Concept. A Critical Review on the
Conceptualization and Measurement of Environmental Concern. Int. Rev. Soc. Res. 1,
11–31.
Schaffrin, A., 2011b. Environmental Concern for Climate Change: A Welfare State
Perspective 8/2011. Soclife Work. Pap. Ser. 8.
Schaffrin, A., Reibling, N., 2015. Household Energy and climate Mitigation Policies:
Investigating Energy Practices in the Housing Sector. Energy Policy 77, 1–15.
doi:10.1016/j.enpol.2014.12.002.
Schelly, C., 2010. Testing Residential Solar Thermal Adoption. Environ. Behav. 42,
151–170. doi:10.1177/0013916508327867
Schmidt-Catran, A.W., Fairbrother, M., 2016. The Random Effects in Multilevel Models:
Getting Them Wrong and Getting Them Right. Eur. Sociol. Rev., 32(1), 23–38.
doi:10.1093/esr/jcv090
Schultz, P., Zelezny, L., 1998. Values and Pro-Environmental Behavior: A Five-Country
Survey. J. CROSS-Cult. Psychol. 29, 540–558. doi:10.1177/0022022198294003
Sharp, J., Smith, M., 2003. Social Capital and Farming at the Rural-Urban Interface: The
Importance of Nonfarmer and Farmer Relations. Agric. Syst. 76, 913–927. doi:10.1016/
S0883-2927(02)00083-5
Soyez, K., Hoffmann, S., Wuenschmann, S., Gelbrich, K., 2009. Pro-Environmental Value
Orientation across Cultures Development of a German and Russian Scale. Soc. Psychol.
40, 222–233. doi:10.1027/1864-9335.40.4.222
Spaargarden, G., Mol, A.P.J., 1992. Sociology, environment, And Modernity: Ecological
Modernization as a Theory of Social Change. Soc. Nat. Resour. 5, 323–344. doi:10.

1080/08941929209380797
Stedman, R. (2003). Is it Really Just a Social Construction? The Contribution of the
Physical Environment to Sense of Place. Soc. & Nat. Res. 16(8), 671–685. http://doi.
org/10.1080/08941920390217627
Uyeki, E., & Holland, L. (2000). Diffusion of Pro-Environment Attitudes? American
Behavioral Scientist 43(4), 646–662. http://doi.org/10.1177/00027640021955478
Wachenheim, C.J., Rathge, R., 2002. Residence and Farm Experience Influence Perception
of Agriculture. Rural Am. 16, 18–29.
Walker, G., 2008. Decentralised Systems and Fuel Poverty: Are There Any Links or Risks?
Energy Policy 36, 4514–4517. doi:10.1016/j.enpol.2008.09.020
Welsch, H., Kuehling, J., 2009. Determinants of Pro-Environmental Consumption:
The Role of Reference Groups and Routine Behavior. Ecol. Econ. 69, 166–176.
doi:10.1016/j.ecolecon.2009.08.009
Whitmarsh, L., 2009. Behavioural Responses to Climate Change: Asymmetry of Intentions
and Impacts. J. Environ. Psychol. 29, 13–23. doi:10.1016/j.jenvp.2008.05.003
Wiseman, M., & Bogner, F. (2003). A Higher-Order Model of Ecological Values and Its
Relationship to Personality. Pers. and Ind. Diff. 34(5), 783–794. http://doi.org/10.1016/
S0191-8869(02)00071-5
Xiao, C., & Dunlap, R. E. (2007). Validating a Comprehensive Model of Environmental
Concern Cross-Nationally: A US-Canadian Comparison. Soc. Sci. Quar. 88(2),
471–493. http://doi.org/10.1111/j.1540-6237.2007.00467.x
Yuchtman-Ya’ar, E. (2002). Value priorities in Israeli society: An examination of
Inglehart’s theory of modernization and cultural variation. Comparative Sociology
1(3–4), 347–67. doi:10.1163/156913302100418637
Zahran, S., Kim, E., Chen, X., & Lubell, M. (2007). Ecological Development and Global
Climate Change: A Cross-National Study of Kyoto Protocol Ratification. Soc. & Nat.
Res. 20(1), 37–55. http://doi.org/10.1080/08941920600981355
Part II
in Europe
Personal activist profiles

Chapter 4
Explaining environmental
activism by national cultures
The hypothesis of hysteresis
Jean-Paul Bozonnet
Common sense and literature tell a story where some nations in Europe are green
and virtuous and others are not. The eternal ‘spirit of the people’ would lead some
countries to environmental commitment and others to artificial worlds; to a delib-
erate and even joyful destruction of nature. Such are the differences described
between Protestants and Catholics: the French artificial alignment of Versailles
pitted against the blurred natural British gardens; or the Mediterranean taste for
vibrant city crowds contrasted with the romantic German solitude of the forest;
not to mention the Slavic soul and its infinite nostalgia for nature. This tradi-
tional literature probably encourages the current commonplaces about the green
German and Nordic models, as opposed to the environmental laxness imagined in
Mediterranean countries.
This chapter deals with the different levels of environmental activism in
European countries. The objective is to answer the question: What has this envi-
ronmental commitment to do with national cultures in Europe? Specifically, after
describing how activism is spread over the continent, we will identify the factors
that explain this distribution. Could these factors be a product of various forms of
a universal rational choice, or do they reflect specific elements of each country?
In short, is the concept of national culture relevant when analysing environmental
mobilization or is it a purely imaginary construct?
Defining the concepts

The concept of culture is difficult to define, not because of the lack of definitions,
but rather because there are too many. We retain the concept in the ethnological
sense; that is to say, regarding the specific characteristics of a society or a group.
We will build on the theory of Max Weber, who set the sciences of nature
against the sciences of culture (Weber, 1965). Culture is a pattern of values that
inspire the goals of human action and impose historical choices on societies.
According to Weber, culture pervades not just the purposes of action, but also the
means used to achieve these goals, insofar as they are guided by values. More gen-
erally, culture is not a specific entity but can be found everywhere, in everything
from technical objects to moral ideas, provided that they are an objectification
92 Jean-Paul Bozonnet
of the value system. Culture is wider than artistic activity, because it also affects
politics, education, economy, technology etc. It cannot be reduced to ideology
since it also includes physical activity and the things themselves through their
design. This definition coincides with that of most ethnologists since the nine-
teenth century (Tylor, 1871).
But this definition of culture is not completely adequate for our purposes. In
fact, it includes both instrumental rational goals, which are universal, and value-
rational goals, which are related to specific groups or societies (Weber, 1965).
We will put aside the goals related to rational choice and only retain those guided
by those values that characterize national cultures studied here. Two additional
features need to be added to the definition. First, cultures are transmitted through
socialization, and whatever their origins, acquired values quickly become super-
personal and anonymous (Kroeber, 1952). They are printed in cultural patterns
and applied to entities of different sizes, from large civilizations or nations to
small groups that can be analysed as subcultures. Thus, cultural anthropology
argues that, although inherited from known historical circumstances, cultures
remain largely unconscious for individuals. Second, every culture is specific to
a collective entity, and the operational criteria to identify it are the significant
differences between these entities. Conversely, what is universal is deemed to be
outside of cultures (Lévi-Strauss, 2003).
Our definition has at least two consequences. First, it is quite far from that
set out by Mary Douglas, Shalom Schwartz and Ronald Inglehart, who insist
much more on the psychological aspect and universality of cultural values
(Milfont, 2012), while here we focus on the specificity of a society. Second,
it is based on socialization, and this will lead us to the concept of hysteresis
(Bourdieu, 1979; 2002). This helps us understand more precisely the function-
ing of the socialization that is at the root of national cultures, which foster or
hamper environmental activism.
In addition, this conception of cultures clarifies the meaning of several other
concepts: political culture, national culture and environmental culture.
As far as culture is concerned, the concept applies to very different objects. It
can be considered from the perspective of the political choices made by a soci-
ety or a social group, which then represent a political culture. Cultural values
permeate societies in their relationships to the State: they leave their mark not
just on the political aims, but also on the means used to achieve them; namely,
on the political regimes, public policies and various forms of participation. For
example, choosing the repertoire of the demonstration or the association reflects
a different political culture than waiting for the State to take action. Scholars have
shown that political cultures vary greatly according to social class or religion
(Bréchon, 2004).
Culture may be regarded from the nation’s point of view. We then distinguish
different national cultures, in that they propose specific goals and means, which
characterize the relationship to the State, business, religion, and all kinds of beliefs
and customs specific to a country.
Explaining environmental activism by culture 93
At last we can apply the concept of culture to different areas of social life. So,
environmentalism is part of the culture of a country, since it lays down the pro-
tection of nature and the environment as goals and values for its citizens. Thus,
environmental cultures are found in attitudes and opinions, as well as in different
behaviours that reflect environmental commitment, such as petition, demonstra-
tion, organization membership or donation.
Environmental activism is a less controversial concept than culture. Two
main types of practices can be distinguished: ‘impact-oriented behaviours’

that are environmentally friendly because of their consequences, and ‘intent-
oriented behaviours’ that are directly motivated by environmental purposes
(Stern, 2000).
‘Impact-oriented behaviours’ are not necessarily driven by ecological goals.
They may be based on the aim of health, as, for example, the case of organic
food, or financial incentives such as saving water or electricity. They take place
mainly in the private or domestic sphere. In contrast, ‘intent-oriented behaviours’
are structured by environmental purposes: for example, a demonstration or a peti-
tion is motivated directly by the values of
protecting nature and the environment.
They may be analysed as means oriented towards achieving environmental goals.
Environmental activism does not seek to change the environment directly, but to
influence the choices of public authorities in such a way that they implement envi-
ronmental policies, so, ‘intent-oriented behaviours’ pertain to the public sphere.
However, their intentions frequently go unheeded (Stern, 2000): protests do not
always reach their proclaimed goals, and, unfortunately, the quality of the envi-
ronment is not necessarily affected by these actions.
When they are studied by historians of social movements, ‘intent-oriented
behaviours’ are often considered to constitute a repertoire requiring skills
acquired in the course of history by a long learning process, and providing specific
resources to social actors (Tilly, 1984). In the field of political science, they are
analysed as forms of collective action or political participation, which correspond
to the input (politics) that puts pressure on the political system, in order to obtain
an output; the implementation of public policies (Easton, 1965; Dompnier, 2014).
Data and methods

Our objective is to provide evidence of a correlation between environmental
activism and national cultures in Europe, and to suggest a hypothesis to explain
this correlation; more specifically, the concept of hysteresis. To test this hypothe-
sis, crossing activism with European countries is not enough, but it is necessary to
take in account the influence of other variables often considered to be at its origin
in the academic literature – in particular, economic causality, post-materialism,
cognitive mobilization and social capital. In addition, we will consider religious
affiliation and attendance. We will cross environmental activism, regarded here
as a dependent variable, with the classical predictors of this social commitment,
tested here as independent variables. For each of these latter variables, we discuss
the respective parts of an explanation via universal rational choice, and via the
national cultures specific to each country.
The data come from the 2010 ISSP.1 The environmental activism indicator
is a cumulative index constructed on the basis of four questions with respect to:
protesting, environmental group membership, petitioning on environmental issues
and contributing to environmental organizations in last five years. This index is
reduced to a binary variable with two values: ‘at least one action; or ‘no action’.
The construction of the indicators for the independent variables and their value
labels:
•• Age ⇒ 18–25, 26–35, 36–55, 56–70, 71+ years.

•• Household income ⇒ quartiles.
•• Years of schooling ⇒ primary = 0 to 8 years; secondary = 9 to 13 years;
higher education = 14+ years and still studying.
•• The construction of an index of post-materialism with the two priorities
(Q3a and Q3b) ⇒ 1 = materialism; 2 = mixed with priority to materialism;
3 = mixed with priority to post-materialism; 4 = post-materialism.
•• Social trust ⇒ quartiles of the cumulative index of ‘general trust in people’
and ‘Most people try to take advantage of you’ (Q4a and Q4b).
•• Subjective social-class placement ⇒ quartiles.
•• Religious affiliation ⇒ Roman Catholic; Christian Orthodox; Protestant
(or other Christian); Muslim; no religion, agnostic or atheist.
•• Religious attendance ⇒ Several times a week or more often; once a week to
once a month; once to 3 times a year; less frequently or never.
All the European countries in the ISSP-2010 survey were selected here up to a
total of 20. We also included Russia and Turkey because it is justifiable here,
even though the question of whether they belong to Europe is sometimes debated:
the high rate of Orthodox Christians and Muslims in these countries provides
meaningful data for these religions; a key variable of our hypothesis. The list
of these countries can be seen in Table 4.1. Moreover, in order to introduce the
national dimension in the logistic regression, we constructed a new variable –
‘European areas’ – with a reduced number of only 10 modalities. The grouping
was constructed on the basis of three criteria: country size, geographic proximity
and comparable percentages of environmental activism. Where the latter criterion
was absent we kept the country separate; for example, Switzerland.
Regarding weighting in the tables, we used the national weight for each coun-
try considered separately. In addition, the totals in the tables and the European
areas are weighted by the population size of each country.
Results
After presenting the large variations in environmental activism in European coun-
tries, we will review the main factors by giving the percentages and results of the
logistic regression.
Table 4.1 Percentages of environmental activism (ISSP 2010).
European Environmental Petitioning on Contributing Protesting At least

countries group environmental to 1 activist
membership issues environmental behaviour or
organizations membership of
environmental
group
Austria 7.4% 20.7% 20.4% 3.5% 33.0%

Belgium 7.6% 21.2% 13.1% 3.8% 27.9%
Bulgaria 1.7% 9.8% 1.3% 4.4% 12.4%
Croatia 2.6% 15.7% 4.1% 1.9% 18.3%
Czech 3.0% 13.5% 7.9% 4.6% 18.9%
Republic
Denmark 9.8% 17.4% 21.7% 2.7% 29.5%
Finland 4.9% 14.6% 25.9% 1.8% 32.5%
France 5.8% 29.3% 9.4% 10.8% 34.3%
Germany 5.7% 22.4% 15.3% 4.3% 28.3%
Latvia 1.2% 4.6% 5.3% 2.6% 9.8%
Lithuania 1.8% 6.6% 1.2% 2.7% 8.6%
Norway 4.8% 16.3% 22.4% 3.3% 30.6%
Russia 1.7% 5.1% 3.2% 3.0% 8.1%
Slovak 2.0% 13.8% 9.2% 2.3% 19.9%
Republic
Slovenia 3.6% 10.9% 6.9% 2.8% 17.0%
Spain 3.4% 16.2% 8.0% 10.6% 23.4%
Sweden 5.4% 19.7% 22.6% 2.9% 34.1%
Switzerland 13.5% 26.6% 33.8% 4.5% 43.2%
Turkey 1.7% 4.6% 3.3% 2.3% 7.7%
United 5.7% 22.9% 17.0% 2.9% 29.5%
Kingdom
Total 4.0% 14.9% 9.6% 4.5% 20.1%
Different levels of mobilization in European countries

In the whole survey, the most common environmental action is to petition (14.9%
of citizens), followed by giving a donation (9.6%), participating in a demon-
stration (4.5%) and being a member of an organization (4%). But beyond these
averages, each type of activism is more or less dependent on the country.
Membership of an organization is highest in two countries: Switzerland
(13.5%) and Denmark (9.8%). Petitioning is especially popular in France (29.3%),
Switzerland (26.6%), and throughout Western Europe, but far less in the North of
the continent. Switzerland is still the leader when it comes to giving donations
(33.8%), closely followed by Finland (25.9%), and far ahead of Belgium (13.1%)
and France (9.4%). Finally, participating in an environmental demonstration is
most common in France (10. 8%) and Spain (10.6%), and far less common in
other countries (little more than 4%). Interestingly, despite their strong reputa-
tion for having a green attitude, people in Germany and the Nordic countries do
not seem to be more involved in environmental activism than people in France or
Belgium.
Despite this seemingly chaotic patchwork, there is a general logic in terms
of European geography. Overall, environmental activism is most widespread
in Western and Northern Europe: 28% to more than 40% of citizens, according
to the countries, perform at least one action. The percentages (about 20%) are
lower in Southern and Central Europe (Spain, the Czech Republic, Slovakia,
Croatia and Slovenia), and even lower in far Eastern Europe, particularly in
Turkey and Russia (7.7% to 8.1%). Are these differences a result of the influence
of national cultures?
Economic causality
There is a strong correlation (r = 0.85) between the environmental activism index
and the European country’s GDP (see detailed variables in Table 4.2). For exam-
ple, Bulgaria, with a GDP per capita of US$15,074 a year, has only 12.4% of
people taking part in at least one environmental action, while Switzerland, with a
GDP per capita of US$51,322 has 43.2%.
However, this correlation is far too general to be relevant, and must also be ver-
ified individually. That’s why we crossed the environmental commitment with the
household income (quartiles) of each country. For each of these, the correlation
is confirmed but remains low: Cramer’s V is only 0.13 for the total (Table 4.2).
Traditionally, two theories explain why environmental activism correlates with
wealth. First of all, the thesis of Maslow (2008), which is close to functionalism,
invokes the hierarchy of human needs: once basic needs are satisfied, the higher
demands gradually replacing them become imperious. Thus, environmental val-
ues would spread among individuals wealthy enough to not to have to worry about
economic or security requirements. But Maslow’s theory was heavily criticized:
if ecologically friendly behaviour gradually supplants others during periods of
abundance, this would not be owing to the satisfaction of mysterious basic needs,
but to the decline in the marginal utility of these latter behaviours (Inglehart &
Flanagan, 1987). This microeconomic reasoning is reinforced at the macrolevel by
Kuznets theory, which predicts that developed countries will improve their envi-
ronmental quality, because people have enough resources to mobilize themselves
and implement environmental policies, which explains the correlation with GDP
seen above (Grossman & Krueger, 1995). Economic causality relies mostly on
rational choice. This theory is also widely used by sociologists to explain political
participation (Olson, 1971), and commitment to the environment, and, therefore,
the policies or practices that result (Diekmann & Franzen, 1999). More broadly,
Table 4.2 ‘Environmental activism index: at least 1 action’ according to socio-demographic variables (ISSP 2010).
GDP by ‘environmental Correlation (Cramer’s V) between ‘environmental activism index: at least one action ‘and . . .
activism index: at least
one action’
= > r = 0.89
GDP** % of . . . quartiles . . . Subjective . . . Post- . . . Years of . . . Social . . . Religious
‘environmental of social class materialism schooling trust belonging
activism index: household placement (4 (3 (Index:
at least 1 income (quartiles) modalities) modalities) q4a + b)
action ’
Austria 41,864 33.0% 0.11 0.09 0.11 0.12 0.14 The data do
Belgium 39,142 27.9% 0.18 0.11 0.13 0.19 0.13 not allow the
Bulgaria 15,074 12.4% 0.16 0.19 0.14 0.23 0.14 calculation of
Croatia 18,969 18.3% 0.12 ns 0.17 0.21 0.09 Cramer’s V
Czech Republic 27,051 18.9% 0.13 0.11 0.14 0.19 0.10
Denmark 41,807 29.5% 0.11 ns 0.18 0.07* 0.08*
Finland 38,297 32.5% ns ns ns 0.15 0.12
France 35,867 34.3% 0.11 0.07 0.22 0.15 0.14
Germany 39,612 28.3% 0.16 0.13 0.21 0.17 0.15
Latvia 17,398 9.8% ns 0.11 ns 0.12 ns
Lithuania 20,071 8.6% 0.09 0.12 0.13 0.20 0.17
Norway 58,772 30.6% ns ns 0.11 0.08 0.12
Russia 20,498 8.1% 0.15 0.11 0.16 0.07* ns
Slovak Republic 24,499 19.9% 0.10 0.16 0.15 0.16 ns
Slovenia 27,589 17.0% 0.21 0.13 0.15 0.22 0.17
Spain 32,351 23.4% 0.24 0.14 0.19 0.25 0.16
(continued)
GDP by ‘environmental Correlation (Cramer’s V) between ‘environmental activism index: at least one action ‘and . . .
activism index: at least
one action’
= > r = 0.89
GDP** % of . . . quartiles . . . Subjective . . . Post- . . . Years of . . . Social . . . Religious
‘environmental of social class materialism schooling trust belonging
activism index: household placement (4 (3 (Index:
at least 1 income (quartiles) modalities) modalities) q4a + b)
action ’
Sweden 41,727 34.1% 0.11 ns 0.15 0.11 0.10
Switzerland 51,322 43.2% 0.14 0.18 0.18 0.22 0.20
Turkey 16,155 7.7% 0.16 0.10 0.13 0.24 ns
United Kingdom 35,855 29.5% 0.12 - 0.10* 0.24 0.11
Total 23.4% 0.13 0.15 0.18 0.19 0.16 0.25
* Significance from 0.05 to 0.10.
** GDP per capita (purchasing-power-parity):Thousands US$ per year in 2010 (source: World Bank)
All other variables are described above in ‘Data and methods’ section.
rational choice is part of Weber’s theory, and corresponds to instrumental rational

actions guided by self-interest.
The explanation by reference to social classes is a variant of the economic
hypothesis, but it adds that the cause of the behaviour is not only material
interest, but also the specific position of the person in the field of economic
production. Indeed, at the European level, we find a correlation similar to that
with household income (Cramer’s V = 0.15) between environmental activism
and class indicator: the subjective social-class placement (Table 4.2). Scholars
explained long ago that it is the middle classes rather than blue-collar workers
who are most involved in green behaviour (Morrison, 1973; Lange & Meier,
2009). This explanation can be considered as a modality of the economic cau-
sality. But overall, the correlations with household income and social class are
fairly low; the logistic regression confirms the weak influence of economic
factors, since the individuals in the highest quartile had only 1.5 more chances
to be involved in an environmental action than those of the lowest quartile
(Table 4.3). Ultimately, the mobilized citizens are the wealthiest, but it is not
really wealth that explains their mobilization.
Theoretically, economic factors do not fall under cultural assumption, in the
sense that rational choice implies a universal and uniform causality, whereas cul-
tures represent the specific choices of each country. However, this argument is not
as definitive as it seems: could the economy exert its influence through culture?
Post-materialism
The answer to that question was provided by Ronald Inglehart (1997). According
to this author, economic abundance has a minimal effect on people’s values in a
direct way, which explains the low correlations relative to economic indicators:
it only has an impact on people who have spent their childhood in a context of
abundance. In short, the economy is a necessity, albeit not a sufficient condition;
in order to determine the values, it must pass through socialization.
The values that lead to environmental activism are already present in indus-
trial societies, but cannot be expressed because of constraints as a result of
economic and security priorities. With the postmodern loosening of these institu-
tional constraints, post-materialist values can come to light; in turn, they become
compelling and, among other consequences, lead citizens to mobilize in order
to promote environmental policies (Inglehart, 1995). Post-materialism consists
of a system of values, concerning mainly the maintainance of order, economic
prosperity, freedom of speech and political participation – all indicators included
in ISSP-2010 (Q3a & b), and homogenous enough to create an index (see ‘Data
and methods’ section).
The overall correlation with post-materialist dimensions is rather strong
(Cramer’s V: 0.18) and found regularly in all countries (Table 4.2). Indeed, only
10% of materialist respondents are engaged in environmental actions, instead of
39% among post-materialist ones. The latter are 2.6 times more likely to mobilize
than the former, as shown by logistic regression (Table 4.3).
The mechanism may be summed up as follows: economic abundance would be

a distant but decisive condition; through the socialization of subsequent genera-
tions, it would trigger the shift of the value system towards post-materialism. This
value system, based primarily on free expression and political participation, is a
form of individualism. Furthermore, research confirms that personal autonomy is
a value closely correlated with the proximity to nature and environmental protec-
tion (Schwartz, 1994).
In the end, economic influence has relatively little direct impact, but it is
exerted through socialization, instilling new values linked to individualism. The
process is theoretically universal: for example, it can just as well be observed in
China as in Europe or elsewhere, and as such, we can challenge the fact that it
falls within culture. However, these values, which are produced by socialization,
are experienced personally as an imperious necessity demanded by institutions.
Consequently, they cannot be reduced to a rational calculation, and, in the end,
they may also be considered as part of the cultural heritage.
Cognitive mobilization
The theory of cognitive mobilization (Inglehart, 1970) is a second explanation
commonly cited for the emergence of both environmentalism and new reper-
toires, such as to petition, demonstrate or become a member of an association.
Since these actions consist of intent behaviours – i.e. behaviours governed by
environmental purposes – activists should be fully aware of them. This cogni-
tive dimension is acquired by socialization through family, school, media and
peers. The main indicators of cognitive mobilization are political interest, talk-
ing about politics, media exposure or the education level of the respondent and
their relatives. ISSP-2010 does not provide all these indicators, only the most
important of them: schooling. On this point, the correlation is strong enough
(Cramer’s V = 0.19) between the duration of schooling and environmental activ-
ism (Table 4.2). Logistic regression confirms this strength: respondents with a
higher education level are 2.4 times more likely to engage in activism than those
with primary-level schooling (Table 4.3).
Cognitive mobilization has to do with rational choice: in real life, human
action is often far from purely rational, because it is limited by access to
information in two ways (Simon, 1955). First, it is limited by the degree of
knowledge of the aims of action: for example, some people are not aware of
ecological issues, which are not restricted to local pollution, but affect the
planet as a whole. Cognitive mobilization expresses this planetary conscious-
ness as a form of contemporary civic culture, opposed to the narrow-minded
vision of the ancient parochial community (Almond & Verba, 1989). Second,
cognition involves not only the goals but also the means, including the skills
for mastering social repertoires. Learning how to create a petition, organize a
demonstration or manage an association presupposes specific social techniques
that activists must acquire (Tilly, 1984).
This global knowledge and these practical skills are learned during repeated
contact with certain institutions, especially university (Andersen, 1990): ISSP-
2010 shows that having attended university is a stronger predictor of environmental
mobilization than a primary or secondary study level. Ultimately, cognitive
mobilization is part of the logic of rational choice according to two theories: that
of access to information ;and that of resources required for social movements
(Oberschall, 1994). In this context, it is a universal phenomenon, which does not
require the concept of culture.

However, access to information and knowledge, within rational choice theory,
is not enough to explain the effectiveness of activism. In fact, most environmen-
tal political actions are oriented towards public good and involve fairly costly
means: collective interest is not sufficient to motivate them. So many people are
involved in free-riding, waiting for other people to act and refusing to engage,
in order to obtain the collective goods for free. In short, the cognitive dimension
per se cannot trigger the action, and cannot account for the fairly strong correla-
tion between education and environmental activism. That’s why there are other
dimensions included in cognitive mobilization, especially the values and norms
specific to citizenship and political participation. These are translated into action
by rational means, but are governed by values, and therefore fall outside pure
rational choice. Moreover, values are acquired through socialization (Inglehart,
1997), which allows us to place them in the world of culture. It is, however,
impossible to measure their exact part, due to the lack of indicators in the ISSP-
2010. Finally, civic culture is already culture, and is highly differentiated among
the European nations.
Social capital
Social capital is frequently invoked to explain environmental commitment. It
completes economic abundance and cognitive mobilization factors that are not
enough to trigger environmental political behaviour on their own.
Social capital is a concept that can be defined as the stock of social rela-
tions based on trust in people or institutions, and is specific to a social group.
It has the effect of reducing transaction costs and facilitating action for the col-
lective good that otherwise would not occur (Coleman, 1988). Scholars have
emphasized the strong inequalities of social capital in the different European
countries (Putnam, Leonardi & Nanetti, 1994), and several of them have found
effective links between social capital and environmentalism (Jones, Malesios &
Botetzagias, 2009).
There are many indicators for measuring social capital, such as participation in
networks or citizenship, but the most common, present in the ISSP-2010, is social
trust (Q4a and b). These two questions were used to construct a cumulative index,
and the results show a significant but modest correlation (Cramer’s V of 0.16) in
most European countries (Table 4.2). Logistic regression confirms this low-level
relationship: the highest quartile of social capital is 1.7 times more likely to take
environmental action than the lowest quartile (Table 4.3). Social capital seems to
be a less decisive factor than cognitive mobilization.
What does social capital have to do with culture? First, this concept was cre-
ated within the framework of rational choice. Based mainly on trust, it introduces
an assurance in the prisoner’s dilemma, furthering reciprocity and thus increasing
the gains from cooperation (Grofman & Pool, 1975); it is more advantageous to
commit oneself to contributing to a public good when you have a guarantee that
others will do so too (Coleman, 1988). Another way of analysing social capital
is to regard it as a resource for collective action (Paxton, 2002). From this point
of view, social capital is thus a way to overcome the obstacles of free-riding,
as it ensures that there are rewards for altruistic actions. All in all, the social
capital theory adds the concept of trust as an element to the rational choice para-
digm. This process remains entirely within the scope of rational choice, since trust
results from personal experiences of reciprocity.
Nonetheless, trust is not only acquired through personal experience, but
also much more by socialization and the norms enshrined in the minds of peo-
ple, as well as the social structure of the society in which one lives (Coleman,
1988). So, it is impossible to explain by experience alone the huge differences
in the levels of trust between different European countries, such as, for exam-
ple Sweden and Bulgaria. Trust is necessarily inculcated through socialization
during early childhood, by the family and during a person’s youth, through
schooling or by interacting with peers. Although social capital is rooted in
rational choice and the logic of interest, it depends largely on the legacy of
national cultures. Unfortunately, the ISSP-2010 does not provide adequate indi-
cators to determine the respective share of rational choice and cultural heritage
in relation to social capital.
Religion
Although religious affiliation and attendance of religious services are less fre-
quently tested variables, they have a strong influence on environmental
activism.
According to some scholars (Hand & Liere, 1984; Stoll, 2013), religion
affects environmentalism by proposing a cluster of values more or less favour-
able to its emergence and development. This influence works by socializing
individuals from an early age, and exerting a social pressure thereafter on the
worshippers. Thus, the religious factor cannot be explained by instrumental
rationality, but pertains to value rationality and to the concept of culture in its
ethnographical sense.
However, the influence of religion is controversial. A longstanding hypoth-
esis suggests that Judeo-Christianity would be the cause of assaults against
nature (White, 1967), and yes, some scholars found that people who attend
religious services are more anthropocentric (Hand & Liere, 1984). But the most
common theories link environmentalism to Protestantism (Stoll, 2013), based
on the idea that nature would be more prevalent in Protestant theology than in
Catholicism. Unfortunately, the results of surveys contradict this: they found
less environmentalism among Protestants, especially among great readers of the
Bible, in contrast to Catholics, who are more distant from the book (Eckberg
& Blocker, 1989; Greeley, 1993; Guth & Green, 1995). Finally, other schol-
ars found a very weak link between religious affiliation and environmentalism
(Boyd, 1999), or no connection at all (Hornsby-Smith & Procter, 1995; Martin
& Bateman, 2013). In reality, the relationship between religious affiliation and
environmentalism changes according to the type of indicator; results differ
depending on whether religion is compared to ecocentrism, concern, and a for-
tiori to various environmental behaviours (Sherkat & Ellison, 2007; Bozonnet,
2014). What about environmental activism in ISSP-2010?
Those who attend religious services at least once a week are clearly less likely
to be mobilized: 7% of them have engaged in environmental action compared to
more than 21% for the others. However, the logistic regression reduces this large
difference showing low or not significant correlations (Table 4.3). This blurred
relationship, which is quite low, reflects in reality very different rates across
Europe: in the Latin countries, religious attendance generates less environmental
commitment, but in countries with a Protestant culture, such as Germany, the UK
and Nordic countries, in contrast, it generates more. So this leads us directly to the
influence of religious affiliation.
Significantly, the differences in environmental activism according to religious
affiliation are the strongest encountered so far (Cramer’s V: 0.25; Table 4.2):
about 30% among Protestants and people without religion, and 25% among
Table 4.3 Percentages and binary logistic regression (exp B): ‘at least one environmental
action’ vs. ‘no action’.
Independent variables % of ‘at least Exp (B)

one action’
Household 1st quartile 14% Ref.

income 2nd quartile 18% ns
(quartiles) 3rd quartile 22% 1.2
4th quartile 28% 1.5
Years of Primary (0 to 8 years) 8% Ref.
schooling Secondary (9 to 13 years) 18% 1.6
Higher education (14+ years and 29% 2.4
still studying)
Age 18–25 years 19% 1.4
26–35 years 24% 1.2*
36–55 years 20% 1.5
56–70 years 14% 1.3
71+ years and 19% Ref.
(continued)
Independent variables % of ‘at least Exp (B)

one action’
Post- Materialist 10% Ref.

materialism Mixed (materialist first) 21% 1.4
Mixed (post-materialist first) 26% 1.8

Post-materialist 39% 2.6
Social trust Low trust 12% Ref.
(Quartile 2nd quartile 16% 1.2
index) 3rd quartile 21% 1.3
High trust 31% 1.7
Religious No religion, agnostic, atheist 28% 3.4
affiliation Protestant (or other Christian) 30% 3.1
Roman Catholic 25% 2.8
Christian Orthodox 7% 2.2
Muslim 8% Ref.
Religious Several time a week or more 7% ns
attendance often
Once a week to once a month 21% 1.4
Once to 3 times a year 20% 1.4
Less frequently or never 21% Ref.
European France-Belgium 33% 5.3
areas Germany-Austria 29% 3.8
Nordic countries (Denmark, 32% 4.0
Norway, Sweden, Finland)
UK 28% 3.9
Switzerland 43% 8.9
Spain 23% 3.9
Central Europe (Slovenia, 17% 2.5
Slovakia, Czech Republic,
Hungary, Bulgaria)
Turkey 8% 3.6
Russia 7% ns
Baltic (Lithuania, Latvia) 9% Ref.
* Significance: 0.05
Note: gender and subjective class are outside the equation.
ns: not significant.
ref: reference value of the variable (= 1).
Catholics, instead of only 7% to 8% among Orthodox Christians and Muslims

(Table 4.3). This correlation is robust because it is confirmed by logistic regression:
for example, Protestants are over three times more likely to be engaged in envi-
ronmental activism than Muslims. These findings suggest that Protestants are the
most likely with regard to environmental activism, but they also show other inter-
esting results that should be discussed.
Discussion
Why is Protestantism related to environmental activism? What are the social
mechanisms behind this correlation? What exactly are the roles of socialization
and hysteresis? And, finally, how could they have shaped national cultures?
For a long time, several sociologists proposed an interpretation of the role of
Protestantism, based on John Calvin’s theology (Badie, 1983). Calvinism strongly
rejects the institutional mediation and obedience preached by the Catholic Church.
This rejection leads, on the one hand, to personal responsibility – the constraining
side of individual freedom, as previously examined here – and, on the other, to the
refusal of artificial iconography in favour of the representation of nature, the imme-
diate image of God (Viard, 2012). Creation is a proof of the existence of God and
therefore fosters the faith of believers; moreover, nature suggests a moral purity
to humans deemed as having been corrupted. This view spread among German
Lutherans, British Puritans and later, among American Transcendentalists such
as Thoreau (Stoll, 2013). As a consequence, this theological reasoning places an
emphasis on the two core values of the Protestant Reformation: personal auton-
omy and sensitivity to nature. Thus, Protestant culture is supposed to legitimize
political activism by encouraging individual action, in keeping with the repertoire
of new social movements. In addition, it would promote admiration and respect
for nature, predisposing its adherents to environmentalist commitments.
It is easy to deduce the formation mechanism of this culture, promoting individ-
ualism and love of nature. The believers, exposed to the influence of the Protestant
Church, acquire these values. Moreover, these are rooted in families, which in turn
transmit them to their children, even if they are not churchgoers, but just claim to
belong to the Protestant religion. This is the classic process of socialization. So,
religious affiliation may be considered largely outside the rational choice, since it
is almost always inherited, converts being a tiny minority everywhere. From this
perspective, the religious factor relates to culture.
However, the concept of socialization is not enough to explain why there is
so little difference between Protestants and Catholics today, and a so large gap
with Orthodox Christianity and Islam concerning environmental activism. The
most relevant explanation here seems Durkheim’s (2013) hypothesis about the
more or less high pervasiveness and coercion of religious institutions. We can
therefore see the high level of activism among Catholics as a result of the Roman
institution weakening, which constrains individuals today less than before, and
emphasizes personal responsibility instead. The waning influence of the Roman
Church would contribute to bringing Catholics values closer to Protestant ones.
This assumption also explains the much lower mobilization among Orthodox
Christians and above all among Muslims: these latter religions are much more
demanding, with highly constraining mandatory rituals, and require obedience
rather than individual initiative, especially with regard to activism. In the end, the
highest rate of the atheists and non-religious people falls under the same logic:
they are largely exempt from the obligations of religious institutions (Bozonnet,
2014). Religious issues help to shed light not only on Protestant environmental
activism, but also on other affiliations’ commitment, even for those who con-
sider themselves outside of religion. Thus, religion is an important dimension of
national culture as a whole, and proves to be a major cultural factor with respect
to environmental activism.
We can now take stock of the different variables fostering environmental activ-
ism, and present, in a synthetic way, how they pertain to national cultures.
Part of them falls primarily under rational choice motivated by interest, includ-
ing direct economic influence, but this impact turns out to be rather low. Other
variables such as post-materialism, cognitive mobilization and social capital are
mixed: they combine the purposes of self-interest and other values, but the ISSP-
2010 does not measure the share of each of them. A last variable, religion, is highly
discriminating and falls quite exclusively within the cultural field because it does
not depend on rational choice, but is mainly received by inheritance. However, all
these variables involve transnational criteria that do not account for the specificity
of each country. They can be regarded as culture in the sense of immaterial activi-
ties, but not in the peculiar choice of a society such as defined above. What about
specific national cultures?
To answer this question, we must return to the differences between countries:
Do they still exist, ceteris paribus? that is to say, by neutralizing all other fac-
tors that may influence the environmental activism rate. Our approach consists of
grouping together these countries for statistical reasons in homogeneous areas (cf.
Data and methods section), and inserting them into the logistic regression (Table
4.3). Although they are far from completely solving the problem of identifying the
national influence, the results are highly suggestive.
Logistic regression shows huge differences of environmental activism between
European areas, regardless of all other factors. This means that there is an irreduc-
ible residue specific to each country or group of countries (Table 4.3). Moreover,
the remaining differences are higher than the simple percentages, since, for exam-
ple the Swiss are 8.9 times more likely to mobilize for the environment than the
inhabitants of the Baltic countries. Do these large discrepancies prove the exist-
ence of national cultures that are conducive, or not, to environmental activism?
A first hypothesis would attribute this residue only to the possible existence of
other variables not taken into account by the ISSP-2010 questionnaire. For exam-
ple, the crucial importance of the relationship to the State, which results from the
country’s political system, is often highlighted to explain the rate of environmental
activism (Dalton, Recchia & Rohrschneider, 2003). In the Baltic countries,
Lithuania and Latvia, and Central Europe (Slovenia, Slovakia, Czech Republic,
Croatia, Bulgaria) this rate is among the lowest, and even less with logistic regres-
sion (Table 4.3). This could be a result of the previous half-century under the rule
of the Soviet Union, with a ban on political participation from civil society.
But there is a second, more general hypothesis. Significantly, the percentages

of activism by country form the same hierarchy as the regression results, with
few exceptions (Table 4.3). This makes it difficult to explain these results by
other hidden variables, and suggests a much more convincing hypothesis: the
hysteresis effect.
Consider the example of religion. Among Christians, going to church has
largely decreased in Europe, so it is no longer the religious variable as such
that would have an impact, but a legacy of values, derived from religion and
transmitted through socialization to the subsequent generations of the country
as a whole (Bréchon, 2004). But this concept of national culture is only identifi-
able through the correlation with countries, a residue of the logistic regression,
which cannot find alternative explanations; this residue has not been thoroughly
investigated and remains the result of a mysterious mechanism (a ‘black box’).
Yet, this phenomenon fits the model of the hysteresis effect well (Bourdieu,
1979). Bourdieu argues that the class habitus, perfectly adapted to a given period,
may become ineffective when society is changing. This often happens for farmers
or the ‘petty bourgeoisie’ such as shopkeepers, who represent declining classes.
For example, the matrimonial strategies of the peasants of Béarn, resulting from
education that had been transmitted for generations, were once effective, but they
no longer work in the present day, and these farmers cannot get married (Bourdieu,
2002). In short, hysteresis comes from the habitus acquired in family education,
which is inadequate today, but still imposes its social norms on the mind. In this
paper, we resume this definition not in a negative but in a positive way: the inculca-
tion of values, thanks to religious socialization, will persist even in the absence of
religious practices, but it will be adapted to other contemporary situations, such as
political participation and, more specifically, environmental activism. So, behav-
iours and collective action repertoires acquired during childhood remain active in
adulthood thereafter, and are passed on to successive generations, even if they are
not churchgoers. Moreover, the large number of these behaviours permeates the
national community as a whole, through the secondary socialization of the media,
schools and other ideological institutions. Two more accurate pieces of evidences
can be advanced in favour of this hypothesis. The first concerns the environmen-
tal activism gap between Protestants and Catholics: in Germany, Belgium and
Switzerland, where the two religions are widely present, their mobilization rates
are very similar, which would reflect to the impregnation of each one by the effect
of hysteresis. The second proof is to be found in Turkey, were we see that the
high multiplier of logistic regression is much larger than one would expect from
the very low percentage (Table 4.3), since the Turks, all things being equal, are as
likely to mobilize for the environment as Western or Northern Europeans. This is
possible because in the case of this country, unlike in others, religious attendance
is high, and its significant impact on activism is taken fully into account by the
variable religion, so that the effect of hysteresis has little consequence.
The same reasoning can be extended to other key variables: environmental
activism does not only take place among the rich, post-materialists and people
with high levels of education and social capital. It is also found among the poor
who have had rich or post-materialist parents, and among people who, though
poorly educated or distrustful, are heirs of parents who are highly educated and
confident in others. In either case, these parents have passed on their capac-
ity for political mobilization. These skills could also be acquired through the
media, schools or other ideological institutions. In short, the hysteresis effect
could be applied to all key variables studied above that are subject to socializa-
tion. This process gives even more to those who already have a lot: Switzerland
is a good example because the Swiss have very high levels of wealth (GDP = US
$ 41,000), post-materialism (= 38%), education (= 38% higher), trust in others
(75% confidence), and Protestant or irreligious culture (58%), leading to by far
the largest multiple in the logistic regression (Table 4.3). Here, we have a typi-
cal example of positive feedback, and this sheds some light on the mechanisms
of hysteresis that work in a systemic way.
Conclusion
The hysteresis effect would be the way in which a specific national culture
becomes favourable, or not, to environmental activism. This process mixes indi-
vidualistic values, political competence and trust in others; all things that are
acquired directly, through inheritance from a family education or through sec-
ondary socialization by institutions such as schools, the media or associations.
Moreover, we have shown that one of the most important variables is probably
the value system derived from a society’s religious heritage. National culture does
not mean, in any way, the existence of an immutable and definitive frame; on
the contrary, it clearly appears as the evolutionary product of the history of each
country. However, in this case, the concept of hysteresis implies the existence of
a form of path dependency.
Finally, these results have the advantage of challenging many prejudices about
the eternal ‘spirit of the people’. The sociologist is a righter of wrongs, who
knocks some European nations off their pedestal and rehabilitates others that are
sometimes poorly judged by public opinion, and tell a story that causes national
cultures to lose some of their mystery, which increases their credibility.
Note
1 To know more about these data, and especially the survey sampling and questionnaire,
go to: http://www.issp.org/
References
Almond, G.A., & Verba, D.S. (1989) The civic culture: political attitudes and democracy
in five nations. New York: Sage.
Andersen, J.G. (1990) ‘Environmentalism’, ‘new politics’ and industrialism: Some theo-
retical perspectives. Scandinavian Political Studies, 13 (2), 101–118.
Badie, B. (1983) Culture et politique. Paris: Economica.

Bourdieu, P. (1979) La distinction. Paris: Ed. de Minuit.
Bourdieu, P. (2002) Le bal des célibataires: crise de la société paysanne en Béarn. Paris:
Ed. du Seuil.
Boyd, H.H. (1999) Christianity and the environment in the American public. Journal for
the Scientific Study of Religion, 38 (1), 36-44.
Bozonnet, J.-P. (2014) L’écocentrisme en Europe : Une mise en récit post-industrielle
et post-religieuse. In Bréchon, P. & Gonthier, F. (eds.). Les valeurs des Européens,

Évolution et clivages. Paris: A. Colin, pp. 89–104.
Bréchon, P. (2004) Influence of religious integration on attitudes : A comparative analysis
of European countries. Revue Française de Sociologie, 45 (5), 26–48.
Coleman, J.S. (1988) Social capital in the creation of human capital. American Sociological
Review, 94 (Supplement), 95–120.
Dalton, R.J., Recchia, S., & Rohrschneider, R. (2003) The environmental movement and
the modes of political action. Comparative Political Studies, 36 (7), 743–771.
Diekmann, A., & Franzen, A. (1999) The wealth of nations and environmental concern.
Environment and Behavior, 31 (4), 540–549.
Dompnier, N. (2014) Valeurs politiques et répertoire d’action des Européens. In: Bréchon, P.,
& Gonthier, F., (eds.) Les valeurs des Européens, Évolution et clivages. Paris : Armand
Colin, pp. 55–72.
Durkheim, E. (2013) Le suicide. Paris: PUF.
Easton, D. (1965) A systems analysis of political life. New York: John Wiley.
Eckberg, D.L., & Blocker, T.J. (1989) Varieties of religious involvement and environmen-
tal concerns: Testing the Lynn White thesis. Journal for the Scientific Study of Religion,
28 (4), 509–517.
Greeley, A. (1993) Religion and attitudes toward the environment. Journal for the Scientific
Study of Religion, 32 (1), 19–28.
Grofman, B., & Pool, J. (1975) Bayesian models for iterated Prisoner’s Dilemma games.
General systems, 20,185–194.
Grossman, G.M., & Krueger, A.B. (1995) Economic growth and the environment. The
Quarterly Journal of Economics, 110 (2), 353–377.
Guth, J.L., & Green, J.C. (1995) Faith and the environment: religious beliefs and attitudes
on environmental policy. American Journal of Political Science, 39 (2), 364-382.
Hand, C.M., & Liere, K.D.V. (1984) Religion, mastery-over-nature, and environmental
concern. Social Forces, 63 (2), 555–570.
Hornsby-Smith, M.P., & Procter, M. (1995) Catholic identity, religious context and envi-
ronmental values in Western Europe: evidence from the European values surveys.
Social Compass, 42 (1), 27–34.
Inglehart, R. (1970) Cognitive mobilization and European identity. Comparative Politics,
3 (1), 45–70.
Inglehart, R. (1989) Culture shift in advanced industrial society. Princeton: Princeton
University Press.
Inglehart, R. (1995) Public support for environmental protection: objective problems and
subjective values in 43 societies. Political Sciences and Politics, 28 (1), 57–71.
Inglehart, R. (1997) Modernization and postmodernization, cultural, economic, and politic
change in 43 societies. New Jersey: Princeton University Press.
Inglehart, R., & Flanagan, S.C. (1987) Value change in industrial societies. The American
Political Science Review, 81 (4), 1289–1319.
ISSP Research Group 2012, International social survey programme: Environment

III – ISSP 2010. Gesis Data Archive, Cologne. ZA5500 Data file version 2.0.0,
doi:10.4232/1.11418
Jones, N., Malesios, C., & Botetzagias, I. (2009) The influence of social capital on willing-
ness to pay for the environment among European citizens. European Societies, 11 (4),
511–530.
Kroeber, A.L. (1952) The nature of culture. Chicago: University of Chicago Press.
Lange, H., & Meier, L. (2009) The new middle classes: globalizing lifestyles, consumerism
and environmental concern. New York: Springer, Chapter 1, pp 1–21.
Lévi-Strauss, C. (2003) Anthropologie structurale. Paris: Pocket.
Martin, W.C., & Bateman, C.R. (2013) Consumer religious commitment’s influence on
ecocentric attitudes and behavior. Journal of Business Research, 67 (2), 5–11.
Maslow, A.H. (2008) Devenir le meilleur de soi-même: Besoins fondamentaux, motivation
et personnalité. Paris: Ed. d’Organisation.
Milfont, T.L. (2012) Cultural differences in environmental engagement. In: Clayton, S.D.,
(ed.) The Oxford handbook of environmental and conservation psychology. Oxford:
Oxford University Press, pp. 181–200.
Morrison, E.E. (1973) The environmental movement: Conflict dynamics. Journal of
Voluntary Action Research, 2 (2), 74–85.
Oberschall, A.R. (1994) Rational choice in collective protests. Rationality and Society, 6
(1), 79–100.
Olson, M. (1971) The logic of collective action: public goods and the theory of groups.
Cambridge (MA): Harvard University Press.
Paxton, P. (2002) Social capital and democracy: An interdependent relationship. American
Sociological Review, 67 (2), 254–277.
Putnam, R.D., Leonardi, R.J., & Nanetti, R.Y. (1994) Making democracy work: civic tradi-
tions in modern Italy. Princeton: Princeton University Press.
Schwartz, S. H. (1994) Are there universal aspects in the structure and contents of human
values? Journal of Social Issues, 50 (4), 19–45.
Sherkat, D.E., & Ellison, C. (2007) Structuring the religion-environment connection:
identifying religious influences on environmental concern and activism. Journal for
the Scientific Study of Religion, 46 (1), 71–85.
Simon, H.A. (1955) A behavioral model of rational choice. Quarterly Journal of Economics,
69 (1), 99–118.
Stern, P.C. (2000) Toward a coherent theory of environmentally significant behavior.
Journal of Social Issues, 56 (3), 407–424.
Stoll, M. (2013) Les influences religieuses sur le mouvement écologiste français. In:
Mathis, C. F. & Mouhot, J.-F. (eds.). Une protection de la nature et de l’environnement
à la française (XIXe–XXe siècles). Seyssel (France): Champ Vallon, pp. 313–326.
Tilly, C. (1984) Les origines du répertoire de l’action collective contemporaine en France
et en Grande Bretagne. Vingtième Siècle, 4 (1), 89–108.
Tylor, E.B. (1871) Primitive culture: researches into the development of mythology,
philosophy, religion, art, and custom. London: J. Murray.
Viard, J. (2012) Penser la Nature. La Tour d’Aigues (France): Ed. de l’Aube.
Weber, M. (1965) Essais sur la théorie de la science. Troisième essai: essai sur quelques
catégories de la sociologie compréhensive. Paris: Plon, et Chicoutimi (Québec): Ed. de
Chicoutimi.
White, L. (1967) The historical roots of our ecologic crisis. Science, 155 (3767), 1203–1207.
Chapter 5
Explaining membership in
pro-environmental groups
A comparative European perspective
Eglė Butkevičienė and Vaidas Morkevičius
Introduction
Environmental activism and other pro-environmental behaviours are emerging
as important forms of civic activity, which engage both experts (social activ-
ists, scientists, politicians) and the public at large. Some research shows that
levels of involvement in pro-environmental activism differ among countries
(Longhofer & Schofer, 2010). However, there is no consensus about factors
which are responsible for this variation. Consequently, this chapter focuses on
studying cross-national variations of membership in pro-environmental groups,
seeking to explain the impact of country-level factors.
The main data source of this chapter is the International Social Survey
Programme (ISSP) module Environment III (ISSP Research Group, 2012).
The sample used in this chapter includes 25,132 respondents from 19 European
countries. The ISSP is one of the best data sources that allows cross-national com-
parison. There are a number of articles that concentrate on analysis of ISSP data,
particularly the thematic module Environment (Franzen, 2003; Hunter, Hatch &
Johnson, 2004; Franzen & Meyer, 2010; Freymeyer & Johnson, 2010; Balzekiene
& Telesiene, 2011; Hadler & Haller, 2011, 2013; Franzen & Vogl, 2013, Reyes,
2013, 2014). Franzen and Meyer (2010) investigated individual as well as cross-
national differences in environmental attitudes, employing Inglehart’s (1989)
theory of post-materialism, Dunlap and Mertig’s (1995) globalization explana-
tion, and the prosperity hypothesis. Hunter, Hatch and Johnson (2004) focused
on cross-national gender variation in environmental behaviours. Freymeyer and
Johnson (2010) investigated contextual and individual factors influencing envi-
ronmental behaviours in 26 countries. Reyes (2014) also took data from the
Environment modules of the ISSP in 1993, 2000 and 2010, and analysed attitudi-
nal dimensions, trends and significant predictors of public and private behaviour
in the Philippines. Balzekiene and Telesiene (2011) have concentrated on the
Lithuanian case and analysed the patterns of environmental behaviour, a statisti-
cal portrait of an environmentally active individual, and what influence the level
of environmental concern and subjectively assessed environmental knowledge
has upon environmental behaviour in Lithuania. Hadler and Haller (2011, 2013)
112 Eglė Butkevičienė and Vaidas Morkevičius
focused their research on two types of pro-environmental behaviour – namely,

public and private environmental behaviour – and analysed the relationship
between an individual’s environmental behaviour and the social context.
This chapter considers quantitative aspects of environmental activism focus-
ing on the trends of participation in environmental groups across 19 European
countries. Such a research approach allows insights into how the lay public is
involved in environmental activism, and how we could explain the differences
among countries regarding this type of behaviour. Thus this research addresses
the following questions that form the rationale of the chapter:
•• What are the patterns of membership in environmental groups across differ-

ent European countries?
•• Which factors explain country differences in levels of public environmen-
tal activism, such as membership in environmental groups across Europe?
The factors influencing environmental activism can be micro- (individual)
or macro- (country) level. In this article our focus is on country-level
determinants.
In the following section, we present theoretical background and discuss poten-

tial explanatory factors of national differences of pro-environmental behaviour
in more detail. The second section presents a fuzzy-set Qualitative Comparative
Analysis (fsQCA) – the method employed for our cross-national comparisons –
and its advantages for our research purposes. The third section discusses data
sources, including a description of the sample of the ISSP module Environment III
(conducted in 2010) and a detailed description of the dependent and independent
variables. The fourth section is devoted to empirical analysis of the data, pre-
senting results of fsQCA. In the last section we provide a discussion of the main
findings and conclusions.
Theoretical background
Environmental activism can be understood as encompassing different dimen-
sions of actions. From the broader philosophical perspective, behaviours that
have a strong, explicitly expressed environmental orientation can be interpreted
as some sort of environmental activism differing in the levels of involvement.
Thus sometimes the conceptual line between environmental activism versus
pro-environmental behaviour is rather blurred. The literature exhibits a variety
of ways to understand and classify pro-environmental behaviours (Thøgersen,
2006; Fielding, McDonald & Louis, 2008; Lenox & Eesley, 2009; Whitmarsh &
O’Neill, 2010; Dono, Webb & Richardson, 2010; Saphores, Ogunseitan &
Shapiro, 2012). One of the ways to define pro-environmental activism is to focus
on its impact, and analyse behaviour that causes the environmental change either
directly or indirectly, by shaping the context in which choices about the environ-
ment are being made (Stern, 2000).
Membership in pro-environmental groups 113
Stern (2000) understands environmental activism as one of the types of

environmentally significant behaviours that encompasses active commitment to
environmental organizations, social movements and active protest forms, such
as demonstrations. On one hand, Stern (2000) relates environmental activism
to environmental citizenship behaviour, but, on the other hand, he distinguishes
it from non-activist behaviours in the public sphere, such as signing petitions
or giving money to environmental issues. In this chapter, the focus is on the
active commitment to environmental organizations, particularly, on the mem-

bership in pro-environmental groups and the determinants that shape this type
of behaviour.
Usually research focuses on microlevel determinants of environmental con-
cern and activism, and mainly incorporates the impact of sociodemographic and
other individual-level factors. Although there are some views that sociodemo-
graphic factors have a very limited explanatory power for pro-environmental
behaviours, still the literature presents some significant correlations. Kvaløy,
Finseraas and Listhaug (2012) found that perception of the seriousness of global
warming, as an environmental issue, correlated positively with high levels of
education. They argue that ‘education is a cognitive resource which facilitates
the understanding of the phenomenon and provides the citizen with capacity
to take part in political action’ (Kvaløy, Finseraas & Listhaug, 2012: p. 12).
Stern (2000) emphasized the impact of personal capabilities (such as availabil-
ity of time to act) on environmentally significant behaviours. He also showed
that environmental activism was negatively associated with age. Considering
gender, Hunter, Hatch and Johnson (2004) found that women tend to engage
in more environmental behaviours than men in many nations, particularly
private behaviours. Freymeyer and Johnson (2010), investigating individ-
ual factors influencing environmental behaviours in 26 countries, found that
pro-environmental behaviours also correlate with gender.
However, environmental activism can be influenced by macrolevel factors as
well. We examine the impact of two groups of factors: (1) pro-environmental
awareness and behaviours of the population; and (2) existing societal institutions
within a country.
Environmentally significant behaviour can be affected by attitudinal factors,
including pro-environmental awareness of the population: beliefs, concern and
knowledge about the environment (Stern, 2000; Givens & Jorgenson, 2013).
Previous studies have revealed that higher environmental concern is associated
with acting more pro-environmentally, although relationships are generally not
strong (Steg & Vlek, 2009). The levels of knowledge people in society have about
the environment could also be a determinant of membership in environmental
groups. Knowledge has been considered to be the main factor that can influence
the perception of risk concerning science and technology (Kim, Choi & Wang,
2013), and one of major determinants of environmentally significant behaviour
(Stern, 2000). A lack of knowledge can be a factor that negatively influences
environmental activism. People join organizations, write petitions and participate
in protest actions when they support or even defend some ideas. That requires
having higher levels of environmental awareness and knowledge. Alternatively, a
lack of awareness and knowledge about threats to the environment should result
in apathy/neutrality to pro-environmental activities.
Another determinant of membership in environmental groups is pro-
environmental behaviour in general. Analysing data from the 1993 ISSP survey,
Hunter et al. (2004) classified pro-environmental behaviour into two types: pri-
vate behaviour and public behaviour. Hadler and Haller (2011) in their study
found that public behaviour is quite similar across countries, but private behav-
iour differs significantly and is more dependent on a national context. They also
predict that ‘the homogeneity of public behaviour across countries can also be
interpreted in another way: similarities across countries can be an outcome of
globalization forces, and the homogeneity of public behaviour can be considered
a global behavioural pattern’ (Hadler & Haller, 2011: p. 328).
These considerations lead to the prediction that countries with a higher level of
concern about environmental issues and more developed private and public pro-
environmental behaviours will be more likely to have more intensive membership
of their people in environmental groups.
Considering economic context, one of the most commonly used macrolevel
determinants is the wealth of a nation. The positive correlation between the wealth
of a nation and the acceptance of science and technology was found by Kim,
Choi and Wang (2013). Franzen and Meyer (2010) argue that, according to their
research, more environmental concern is reported in wealthier countries than in
poorer nations. This also supports Ingelhart’s (1995) view that in modern socie-
ties people from wealthier nations display more pro-environmental attitudes than
those from poorer countries. However, Kvaløy, Finseraas and Listhaug (2012),
examining individual and cross-national variation in the perception of the seri-
ousness of the global-warming issue, found that concern for global warming is
relatively evenly distributed among rich and poor countries and no significant
correlation could be found there. Freymeyer and Johnson (2010) investigated con-
textual factors influencing environmental behaviours for 26 countries and found
that respondents living in wealthier countries, no matter the country’s environ-
mental state, join environmental groups more often, sign petitions concerning the
environment, and give money for environmental causes.
As for social movements and active participation in non-governmental organ-
izations (NGOs), the political context is also very important. The quality of
democracy and social inclusiveness could foster active participation in different
associations and NGOs. As emphasized by Hadler and Haller (2011), ‘democ-
ratization, in general, promotes the participatory aspects of social movements’,
and ‘within repressive systems, movements are usually less successful, less
able to attract members, and might not develop at all’ (Hadler & Haller, 2011:
p. 318). There are many indices that measure the level of democracy in a c ountry:
Freedom House Index, Polity IV Index, World Governance Index by the World
Bank, Democracy Barometer Index etc. All these indices are important as they
indicate the opportunities for various types of civic associations, including

environmental groups, to exist and act in a country. In this chapter we use the
Democracy Barometer index (for more information see www.democracybaro
meter.org), as this index is particularly suited to measure the level of inclusive-
ness of democratic institutions in European democratic countries (other indices
are more suitable for the measurement of the quality of democracy around the
world, and eliminate almost all the variation of the scores among European
democracies). The Democracy Barometer index is constructed from 105 indica-

tors that are divided into subcategories (Merkel et al., 2014): indicators are used
to calculate the values of 53 subcomponents, which again constitute 18 compo-
nents; two for each of the nine functions. In this chapter we use three functions of
this index: namely, Public Sphere (including the components freedom to associ-
ate and freedom of opinion); Participation (including the components equality
of participation and effective participation); and Representation (including the
components substantive representation and descriptive representation).
Religiosity can also influence the concern for environmental issues and pro-
environmental behaviours. Kvaløy, Finseraas and Listhaug (2012) found that
concern for global warming is the stronger the more often people attend religious
services. Compared with Christians, non-religious respondents tend to report
stronger perceived environmental danger, greater willingness to pay or sacri-
fice for the environment, and perform more private environmental behaviours
(Clements, McCright & Xiao, 2014).
Thus, an overview of the literature shows us the possible interrelations between
environmental activism and societal context. This allows us to hypothesize that the
wealth of a country, inclusiveness of its democratic institution and religiousness
of the population, might influence levels of public pro-environmental activism in
the 19 analysed countries.
Methods and data

Since our study focuses on the analysis of diversity in pro-environmental activ-
ism in 19 European countries, the use of traditional statistical techniques is rather
limited. Therefore, we opted for a recently elaborated methodology of Qualitative
Comparative Analysis (QCA), which is especially suited for medium-N situations
in cross-country comparative research (Ragin, 1987, 2000, 2008). The logic of
QCA differs from traditional statistical analyses in that it is based on set theory
and logical analysis (Boolean minimization). Its main assumption is equifinality
and causal complexity since it treats many pathways to the same outcome equally.
QCA also assumes that each pathway to the outcome can contain different com-
binations of explanatory conditions, and the logical analysis is performed in order
to find logically (and/or substantially) minimal combinations of necessary and
sufficient conditions producing the outcome.
On a more technical level, QCA operates with original variables (causal
conditions/explanatory factors) transformed into measures of set membership.
The transformation is usually called calibration (or fuzzification in fuzzy sets

QCA), whereby a researcher transforms the raw data into set membership scores
by applying thresholds (in the best case scenario, derived from theoretical or sub-
stantive knowledge external to the empirical data). An elaborate description of the
procedure of applying QCA to empirical data analysis is beyond the scope of this
article, and is well documented in various sources (see Rihoux & Ragin, 2009;
Schneider & Wagemann, 2012).1 We would just like to point out four important
aspects concerning our application of QCA:
1 We use the fuzzy-set alternative of the QCA as it allows retention of more

original information in the data and, therefore, produces more nuanced results.
2 Before proceeding to the presentation of results we define and discuss fuzzi-
fication thresholds for all of our variables.
3 In the Findings section we employ a most common logic of the QCA –
necessity analysis followed by the truth table construction and analysis of
sufficient conditions (Boolean minimization) – elaborating only results for
positive outcomes, as results for negative outcomes are only complementary
(and only discussed more thoroughly in the most relevant cases).
4 All of our analyses were performed with a free software environment for
statistical computing and graphics R (R Core Team, 2015) and its special
purpose package QCA, version 1.1–4 (Thiem & Dusa, 2013a; Thiem & Dusa,
2013b; Dusa & Thiem, 2014).
Further, we proceed with the presentation of our raw data – dependent (outcome)
and independent (explanatory factors or conditions) variables – and definitions of
fuzzification thresholds. Before doing that we note that the particular 19 European
countries were included in our analysis, because, in 2010, the International
Social Survey Programme (ISSP) conducted interviews on Environment III in
Austria, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Finland, France,
Germany, Latvia, Lithuania, Norway, Russia, Slovak Republic, Slovenia, Spain,
Sweden, Switzerland, the United Kingdom. The ISSP sample included 25,132
respondents from these countries. This survey is the only one that uses elaborate
measures of different aspects of environmentalism and includes questions on pro-
environmental activism.
We used membership in pro-environmental groups as the dependent variable. In
the ISSP Environment III module (2010) membership in pro-environmental groups
was measured by the question: ‘Are you a member of any group whose main aim
is to preserve or protect the environment?’ (answer categories: ‘Yes/No’). Using
answers to this question we constructed raw dependent variables of the percent-
age of respondents in a country reporting that they are members of a group whose
main aim is to preserve or protect the environment (MemberEnvGr). As we can
see in Fig. 5.1, membership in environmental groups is rather low in all European
countries: on average, just 5% (cross-country median) of the respondents in all 19
countries identified that they are members of some environmental group. Generally,
14 13.5
12
10 9.8
8 7.8 7.9
6.4
%
6 5.3 5.4 5.7

4.7 4.8
4 3.6
3.1 3.3
2.3 2.6
2 1.6 1.9
1.2 1.5
0
LVA
BGR
RUS
LTU
SVK
HRV
CZE
ESP
SVN
FIN
NOR
GBR
SWE
DEU
FRA
BEL
AUT
DNK
CHE
Figure 5.1 Membership in environmental groups.
Source: ISSP 2010, Question 21: Are you a member of any group whose main aim is to pre-
serve or protect the environment? (19 countries, N=25,132).
the percentage of the population participating in environmental groups is above

the median in Western Europe (Switzerland, Denmark, Belgium, Austria, France,
Great Britain, Germany and Sweden). On the other side of the continuum is a group
of countries composed of mostly post-communist states: Latvia, Russia, Bulgaria,
Lithuania, Slovakia, Croatia, Czech Republic, Spain and Slovenia. In these Central
and Eastern-European countries the participation in environmental groups ranges
from 1.2% (in Latvia) to 3.6% (Slovenia). Interestingly, Spain is in the cluster of
countries that exhibits lower rates of participation in pro-environmental groups.
Raw continuous variables need to be transformed into fuzzy-set membership
scores in order to be included into fsQCA. For all our variables we used the direct
method of fuzzification (Ragin, 2008), which makes use of a piecewise logistic
function when transforming raw variables. For this type of transformation, we
needed to define three thresholds: full inclusion into the set of countries with
a considerable proportion of the population participating in groups to preserve
the environment; full exclusion; and the crossover point. Accordingly, we choose
10% as a demarcation line for full inclusion in the set of countries, with a con-
siderable proportion of the population participating in pro-environmental groups;
1.5% for full exclusion from this set; and 4% as a crossover point, at which the
country’s position with regard to its membership in the outcome set is the most
ambiguous (see Fig. 5.2).
As our variables representing pro-environmental awareness and behaviour
of populations in analysed countries we included three indicators: salience of
environmental issues; public and private sphere; and environmentally friendly
behaviour. Salience of environmental issues (EnvironmentProblem) was meas-
ured with the ISSP Environment III (2010) question ‘Which of these issues is
Q21|Member of a group to preserve environment

2 4 6 8 10 12
% in country
Mean – 4.86 Median – 4.7 Thresholds – 1.5, 4, 10
Figure 5.2 Thresholds of fuzzification for outcome variable: proportion of population

participating in groups to preserve or protect environment.
Note: ISSP 2010 (Environment III) data, 19 European countries, N=25,132
the most important for [COUNTRY] today?’ (answer options included ‘The
environment’). As we can see in Table 5.1, environmental issues did not get
much attention in all 19 European countries. Rather, issues such as the economy,
poverty, health for respondents’ countries care and education were considered
much more important (figures not reported here). The economy is an issue that is
emphasized by over 30% of populations in both Western countries, such as Spain,
Belgium, Denmark, Finland, and in Central and Eastern-European countries, such
as the Czech Republic, Slovenia, Latvia. In Lithuania and Bulgaria people are the
most concerned about the poverty issue. Northern countries such as Norway and
Sweden concentrate on health care, and Germany, on education. With regard to
the environment, Norway (15.6%) followed by Switzerland (13.5%) exhibit the
highest interest in the environment. The country with the least concerned popula-
tion regarding environmental issues was Lithuania (only 0.9% of the respondents
chose environment as the most important issue for the country). Here, we might
also divide countries into two groups with higher and lower levels of popular
interest in environmental issues: Western countries fall into a group with rela-
tively higher attention to the environment, while the post-communist countries
(like Lithuania, Latvia, Croatia, Bulgaria, Slovak Republic, Slovenia) lag behind
them, with the exception of the Czech Republic (4.8%), which exhibits a higher
score than the United Kingdom and Spain.
The extent of public- and private-sphere environmentally friendly behaviour in
analysed countries was calculated using a technique similar to Hadler and Haller
(2011, 2013). As our indicators, we employed the following eight environmentally
friendly behaviour items included in the ISSP Environment III module (2010):
Table 5.1 Pro-environmental awareness and behaviour of populations and membership

in groups to preserve or protect environment: raw and fuzzified data.
Country Causal conditions Outcome
Environment PublEnv PrivEnvBehaviour MemberEnvGr

Problem Behaviour
% Fuzzified % Fuzzified Mean Fuzzified % Fuzzified
Austria 7.2 0.60 34.3 0.86 1.64 0.92 7.9 0.87

Belgium 8.0 0.66 28.5 0.67 1.61 0.90 7.8 0.87
Bulgaria 2.1 0.09 10.3 0.05 0.88 0.02 1.5 0.05
Croatia 2.0 0.09 17.7 0.19 1.17 0.18 2.6 0.16
Czech R. 4.8 0.33 18.2 0.21 1.27 0.34 3.1 0.26
Denmark 10.6 0.82 28.0 0.64 1.49 0.76 9.8 0.95
Finland 7.2 0.60 32.2 0.80 1.46 0.72 4.7 0.59
France 6.4 0.53 33.4 0.84 1.78 0.97 6.4 0.76
Germany 6.5 0.54 27.4 0.62 1.70 0.95 5.7 0.70
Latvia 1.8 0.08 9.4 0.04 0.98 0.04 1.2 0.04
Lithuania 0.9 0.05 7.8 0.03 1.18 0.19 1.9 0.08
Norway 15.9 0.96 29.9 0.72 1.21 0.24 4.8 0.60
Russia 5.0 0.36 6.9 0.03 0.97 0.04 1.6 0.06
Slovak R. 2.5 0.11 20.8 0.30 1.23 0.27 2.3 0.12
Slovenia 2.9 0.14 15.4 0.13 1.41 0.62 3.6 0.38
Spain 3.1 0.15 22.2 0.37 1.30 0.40 3.3 0.30
Sweden 10.6 0.82 33.5 0.84 1.32 0.44 5.4 0.67
Switzerland 13.8 0.93 42.1 0.97 1.71 0.95 13.5 0.99
United Kingdom 3.3 0.17 27.8 0.63 1.38 0.56 5.3 0.65
•• Three for the public sphere: Q22. In the last five years, have you . . . (answer
scale: 0 = ‘No’, 1 = ‘Yes’)
1 ‘signed a petition about an environmental issue?’
2 ‘given money to an environmental group?’
3 ‘taken part in a protest or demonstration about an environmental issue?’
•• Five for the private sphere: Q20. How often do you . . . (answer scale:
0 = ‘Never’; 1 = ‘Sometimes’; 2 = ‘Often’; and 3 = ‘Always’).
1 ‘make a special effort to sort glass, tins, plastic or newspapers, and so on,
for recycling?’
2 ‘make a special effort to buy fruit and vegetables grown without pesti-
cides or chemicals?’
3 ‘cut back on driving a car for environmental reasons?’
4 ‘reduce the energy or fuel you use at home for environmental reasons?’
5 ‘choose to save or re-use water for environmental reasons?’
We constructed individual-level indices for both types of behaviour and calcu-

lated country means of their values. In the index of public-sphere environmentally
friendly behaviour (PublEnvBehaviour), persons reporting any one type of the
three behaviours were assigned ‘1’ and the rest ‘0’. The index of private-sphere
environmentally friendly behaviour (PrivEnvBehaviour) was constructed as a
mean of the five items included.2 As data in Table 5.1 show, public-sphere envi-
ronmentally friendly behaviour was not a very common type of activity that
populations in the analysed countries were engaged in during the previous five
years. Only Switzerland appeared as a country where a considerable proportion
of inhabitants (42.1%) was involved in pro-environmental public activities; eve-
rywhere else, it was less than a third of the population; and in countries such as
Lithuania, Latvia and Russia, this part constituted less than 10% of the population.
Again we might discern a pattern that Western-European countries exhibit
higher levels of population involvement in public-sphere environmentally friendly
behaviour than their counterparts in Eastern and Central Europe. With regard to
private-sphere environmentally friendly behaviour the data reveal quite similar
patterns. This type of behaviour is most frequently engaged in by populations in
Germany, France and Switzerland, whereas in Bulgaria, Russia and Latvia people
are less frequently involved in private-sphere environmentally friendly behav-
iour. Once again we see a pattern of the West vs. the East in our data distribution
(see Table 5.1).
For the fuzzification of the three variables reflecting pro-environmental aware-
ness and behaviour, included into the fsQCA as explanatory conditions, we defined
the following thresholds:
•• EnvironmentProblem (percentage of people in a country indicating that the

environment is the most important issue for the country today): 15% for full
inclusion in the set of countries with a considerable proportion of the popu-
lation believing that environmental issues are the most salient; 1% for full
exclusion from this set; and 6% for a crossover point, at which the country’s
position with regard to its membership of this set is the most ambiguous.
•• PublEnvBehaviour (percentage of people in a country indicating that they
were involved in any one type of public-sphere environmentally friendly
behaviour): 40% for full inclusion in the set of countries with a consider-
able proportion of the population involved in public-sphere environmentally
friendly behaviour; 10% or full exclusion from this set; and 25% for a cros
sover point, at which the country’s position with regard to its membership of
this set is the most ambiguous.
•• PrivEnvBehaviour (country mean of the index measuring frequency of
engagement in private-sphere environmentally friendly behaviour): 1.70 for
full inclusion in the set of countries with a population engaged relatively often
in private-sphere environmentally friendly behaviour; 1.00 for full exclusion
from this set; and 1.35 for a crossover point, at which the country’s position
with regard to its membership of this set is the most ambiguous.
As variables representing different types of existing societal institutions in

analysed countries we used three indicators: relative wealth of the country;
inclusiveness of democratic institutions; and public detachment from religious
practice. Relative wealth of a country (GDPpc2009) was measured by its annual
Gross Domestic Product (GDP) per capita at current prices (US Dollars) in 2009.3
As we see in Table 5.2, the wealth of the 19 analysed European countries was
substantially different at the end of the last decade. It ranged from US$6,737 in
Bulgaria to US$79,903 in Norway, which reveals that statistically a Norwegian

was, on average, more than 10 times wealthier than a Bulgarian. In general, we see
that Western-European countries are much wealthier than Eastern and Central-
European states.
The inclusiveness of democratic institutions (DBDem2010) in our analy-
sis was measured with the three functions (Public Sphere, Participation and
Representation) from the Democracy Barometer index for 2010.4 Overall,
the full index includes nine democratic functions. However, components
(such as Freedom to associate, Freedom of opinion, Equality of participation,
Table 5.2 Existing societal institutions in country and membership in groups to

preserve or protect environment: raw and fuzzified data.
Country Causal conditions Outcome
GDPpc2009 DBDem2010 ReligNoneESS MemberEnvGr
US$ Fuzzified Mean Fuzzified Mean % Fuzzified % Fuzzified
Austria 47,490 0.93 46.38 0.57 29.60 0.23 7.9 0.87

Belgium 44,723 0.91 52.84 0.82 57.93 0.89 7.8 0.87
Bulgaria 6,737 0.03 46.37 0.57 19.87 0.09 1.5 0.05
Croatia 14,399 0.11 36.83 0.08 19.05 0.08 2.6 0.16
Czech R. 19,619 0.26 37.05 0.09 78.43 0.99 3.1 0.26
Denmark 57,882 0.98 57.22 0.92 41.27 0.54 9.8 0.95
Finland 47,060 0.93 54.34 0.86 43.30 0.60 4.7 0.59
France 41,567 0.88 44.33 0.45 51.53 0.80 6.4 0.76
Germany 41,027 0.87 45.15 0.51 37.97 0.44 5.7 0.70
Latvia 12,383 0.08 45.13 0.51 48.90 0.74 1.2 0.04
Lithuania 12,070 0.07 41.65 0.27 14.67 0.05 1.9 0.08
Norway 79,903 1.00 55.20 0.88 44.67 0.63 4.8 0.60
Russia 8,509 0.04 0.00 0.00 42.03 0.56 1.6 0.06
Slovak R. 16,337 0.15 47.71 0.63 21.93 0.11 2.3 0.12
Slovenia 24,577 0.48 53.54 0.84 44.50 0.63 3.6 0.38
Spain 32,762 0.71 44.98 0.50 30.90 0.26 3.3 0.30
Sweden 46,148 0.92 64.72 0.98 68.63 0.97 5.4 0.67
Switzerland 69,679 0.99 63.50 0.97 32.77 0.30 13.5 0.99
United 37,421 0.81 46.28 0.56 53.07 0.82 5.3 0.65
Kingdom
Effective participation, Substantive representation and Descriptive repre-

sentation) of the Public Sphere, Participation and Representation functions
are the most directly related to the measurement of inclusiveness of demo-
cratic institutions in a country. We calculated a mean of the scores on these
three dimensions for each country. Data in Table 5.2 reveal that there is a
considerable variation in the inclusiveness of democratic institutions among
the analysed countries (roughly, the index varies from 0, the least inclusive,
to 100, the most inclusive democratic institutions). Croatia and the Czech
Republic are the countries with the lowest scores on our index (36.83 and
37.05, respectively), while Switzerland and Sweden exhibit the highest scores
(63.51 and 64.72, respectively). Interestingly, the distribution of the scores
does not follow the previously observed pattern of the West vs. the East.
Therefore, we might infer that inclusiveness of democratic institutions is not
simply a matter of democratic longevity.
As our measure of public detachment from religious practice in a country
(ReligNoneESS), we used the percentage of people in the country who self-
reportedly do not belong to any religion. The data were retrieved from the European
Social Survey (ESS), which has a question measuring religious affiliation in its
core questionnaire (rlgblg| Do you consider yourself as belonging to any particu-
lar religion or denomination; answer scale: 0 = ‘No’, 1 = ‘Yes’). We averaged the
design weighted percentages of people reporting that they do not belong to any
religion or denomination in Rounds 4, 5 and 6.5 With the averaging procedure we
intended to decrease the possible impact of the measurement error. As we see in
Table 5.2 there is vast variation in the religiosity of populations in the analysed
countries, and ranges from as little as 14.67% of a non-religious population in
Lithuania, to as many as 78.43% people in the Czech Republic reporting that they
do not belong to any religion or denomination. As we see from the extreme cases
they are both from Eastern and Central Europe, which indicates that differences of
religious institutions do not vary along the lines of the West vs. the East.
For the fuzzification of the three variables representing existing societal (eco-
nomic, political and cultural) institutions included into the fsQCA as explanatory
conditions we defined the following thresholds:
•• GDPpc2009 (annual GDP per capita at current prices, in US$, in 2009):

US$50,000 for full inclusion in the set of wealthy countries; US$10,000 for
full exclusion from this set; and US$25,000 for a crossover point, at which
the country’s position with regard to its membership of this set is the most
ambiguous.
•• DBDem2010 (country mean of the three functions – Public Sphere, Partici
pation and Representation – in the Democracy Barometer for 2010): 60 for
full inclusion in the set of countries with highly inclusive democratic insti-
tutions; 35 for full exclusion from this set; and 45 for a crossover point, at
which the country’s position with regard to its membership of this set is the
most ambiguous.
•• ReligNoneESS (percentage of people in a country who do not belong to

any religion as measured by ESS4, ESS5 and ESS6): 65% for full inclu-
sion in the set of countries with a highly non-religious population; 15%
for full exclusion from this set; and 40% for a crossover point, at which
the country’s position with regard to its membership of this set is the most
ambiguous.
As previously mentioned, we examined the impact of two groups of factors sepa-

rately: (1) pro-environmental awareness and behaviours of the population; and
(2) existing societal institutions within a country. This decision was based on
the fact that we compared levels of membership in environmental groups among
only 19 countries. If all the six variables were included into the analysis the list of
counter-factual conditions available for the final solution would be too long6 and
the results of the analysis would be less reliant on factual information. Thus, our
findings in the next section are presented separately for each group of factors. It
is a matter of further empirical analysis to see how variables from these groups
interact with one another in producing the outcomes.
Findings and analysis

Analysis of the impact of pro-environmental awareness and behaviours of the popu-
lation on the levels of public involvement in pro-environmental groups revealed
that there is an interesting interplay of these factors. First, the analysis of neces-
sary conditions showed that there is no single factor that might be responsible for
high levels of public involvement in pro-environmental groups (see Table 5.3).
However, there are several groups of alternative factors that satisfy rigorous cri-
teria (inclusion score higher than 0.9; coverage score higher than 0.7) of necessity
relations among the causal conditions and positive outcome PublEnvBehaviour +
PrivEnvBehaviour, EnvironmentProblem + PrivEnvBehaviour and Environment
Problem + PublEnvBehaviour.7 Applying the same criteria for negative outcomes
(low country levels of membership in groups to preserve or protect environment)
we get two separate explanatory factors that satisfy requirements of necessity
relations: ~PublEnvBehaviour (inclusion = 0.911, raw coverage = 0.899) and
~EnvironmentProblem (inclusion = 0.900, raw coverage = 0.813). Therefore, these
Table 5.3 Results of the analysis of necessary conditions for positive outcomes (high
country levels of membership in groups to preserve or protect environment):
impact of pro-environmental awareness and behaviours of population.
Causal conditions Inclusion Raw coverage
PublEnvBehaviour+PrivEnvBehaviour 0.973 0.829

EnvironmentProblem+PrivEnvBehaviour 0.972 0.795
EnvironmentProblem+PublEnvBehaviour 0.919 0.842
two variants of conditions will draw our special attention if and when computing
intermediate solutions for the analysis of sufficient conditions.
Before proceeding to the analysis of sufficient conditions for the positive
outcomes8 the truth table was constructed, which revealed that there are three
groups of countries where a large proportion of the population is involved in pro-
environmental groups (see Table 5.4): 1) Austria, Belgium, Denmark, Finland,
France, Germany, Switzerland; 2) Norway, Sweden; and 3) the United Kingdom.
In the first and largest of them all three explanatory factors have high scores. The
situation is a bit different in the two Nordic countries – Norway and Sweden – and
the United Kingdom. Quite surprisingly, in the first two countries people are not
keen on engaging in private-sphere environmentally friendly behaviour, and in
the United Kingdom salience of environmental issues is quite low. On the other
hand, in the rest of the countries, except Slovenia, all three explanatory factors
have low scores. Slovenia is the only country where people relatively frequently
engage in private-sphere environmentally friendly behaviour.
Further analysis of the truth table involves Boolean minimization which helps
to derive three types of logically minimal combinations of sufficient conditions
for the positive outcomes: complex, parsimonious and intermediate. The com-
plex solution, which does not include counter-factual explanatory conditions (the
three rows in the truth table which do not have an outcome value), produced
the formula PublEnvBehaviour * (EnvironmentProblem + PrivEnvBehaviour)
= > MemberEnvGr.9 This formula tells us essentially the same story about the
relations between pro-environmental awareness and behaviours of the popula-
tion and levels of public involvement in pro-environmental groups, that were
discussed when presenting the truth table constructed from the fuzzified data.
Table 5.4 Pro-environmental awareness and behaviour of populations and membership

in groups to preserve or protect environment: complete truth table for
positive outcomes (inclusion cut-off=0.9).
Environment PublEnv PrivEnv Member Inclusion Cases

Problem Behaviour Behaviour EnvGr
1 1 1 1 0.997 Austria, Belgium, Denmark,

Finland, France, Germany,
Switzerland
1 1 0 1 0.956 Norway, Sweden
0 1 1 1 0.946 United Kingdom
0 0 1 0 0.829 Slovenia
0 0 0 0 0.349 Bulgaria, Croatia, Czech
Republic, Latvia, Lithuania,
Russia, Slovak Republic, Spain
1 0 1 ? 0.975
0 1 0 ? 0.896
1 0 0 ? 0.832
However, it provides more detailed and nuanced information about the causal
paths towards the outcome, and states that there are two ways for a country to
have a considerable proportion of its public participating in pro-environmental
groups. Both of these require that there is a large proportion of the population
involved in public-sphere environmentally friendly behaviour. Still, this condi-
tion is not sufficient. In some countries (especially, Norway and Sweden) it also
requires that people perceived environmental issues as important. In others (such
as the United Kingdom) private-sphere environmentally friendly behaviour has

to be quite prominent.
We might have opted for a more parsimonious solution and included all the
simplifying assumptions into the minimization procedure. In that case, how-
ever, the solution (PublEnvBehaviour = > MemberEnvGr) would include a
hard counter-factual combination of causal conditions: PublEnvBehaviour
* ~EnvironmentProblem * ~PrivEnvBehaviour. It basically states that a large
proportion of the population involved in public-sphere environmentally friendly
behaviour is a sufficient condition for high levels of public involvement in pro-
environmental groups. However, this could hardly be the case, as at least one of
the additional factors seems to need to be present. Additionally, analysis of the
necessity relations showed that the absence of salience of environmental issues
in a country (~EnvironmentProblem) is associated with negative outcomes.
Therefore, we exclude this simplifying assumption from the parsimonious solution.
Additionally, we exclude the remaining two counter-factual combinations of causal
conditions (~PublEnvBehaviour * EnvironmentProblem * ~PrivEnvBehaviour
and ~PublEnvBehaviour * EnvironmentProblem * PrivEnvBehaviour), as they
both include negation of PublEnvBehaviour in the combination of conditions,
which as shown by the analysis of necessary conditions, is again associated with
the negative outcomes. With these restrictions on simplifying assumptions, our
final solution equals the complex solution (more detailed information about the
solution is provided in Figure 5.3). Thus, we conclude that with regard to the
impact of pro-environmental awareness and behaviours of the population on
the levels of public involvement in pro-environmental groups, the main explana-
tory factor seems to be public activism in public-sphere environmentally friendly
behaviour. Results show that it is a necessary factor. However, it is not sufficient.
Additionally, a country’s population has to either exhibit high concern regard-
ing environmental issues or frequently perform private-sphere environmentally
friendly behaviours in order for the country to have a large proportion of the
population involved in pro-environmental groups.
Further, we proceed with the analysis of the impact of existing societal insti-
tutions on levels of public involvement in pro-environmental groups in the
analysed countries. Here, the analysis of necessary conditions (analysis crite-
ria: inclusion score higher than 0.9, coverage score higher than 0.6) showed that
there is a separate factor (see Table 5.5) – wealth of a country (GDPpc2009) –
that must be present in a majority of cases (raw coverage = 0.805), when a coun-
try exhibits high levels of public involvement in pro-environmental groups.
Solution for positive outcomes (membership scores):

PUBLENVBEHAVIOUR* (ENVIRONMENTPROBLEM+PRIVENVBEHAVIOUR) => MEMBERENVGR
1.0 CHE
DNK
BEL AUT
0.8 FRA
Outcome: MEMBERENVGR
DEU
GBR SWE
0.6 NOR
FIN
0.4 SVN
ESP
CZE
0.2
HRV
LTU
SVK
RUS BGR
0.0 LVA
0.0 0.2 0.4 0.6 0.8 1.0

Conf iguration of causal conditions: PUBLENVBEHAVIOUR*
(ENVIRONMENTPROBLEM+PRIVENVBHAVIOUR)
Figure 5.3 Impact of pro-environmental awareness and behaviours of country populations

on levels of their public involvement in pro-environmental groups: results of
fsQCA.
Notes: Included simplifying assumptions (SA): none; Inclusion=0.917; Raw coverage=0.879
However, influence of the other two factors – inclusiveness of democratic institu-

tions or religiousness of the population (DBDem2010 + ReligNoneESS) – might
also be necessary in some cases (raw coverage = 0.633). Applying similar criteria
for negative outcomes (inclusion score higher than 0.9; coverage score higher
than 0.7), we get further confirmation that explanatory factor GDPpc2009 is
a very important condition. However, here only its interplay with religiousness
of the population – ~GDPpc2009 * ~ReligNoneESS (inclusion = 0.934, raw
coverage = 0.807) – guarantees the negative outcome. Therefore, careful consider-
ation of a factor GDPpc2009 will be performed if and when including simplifying
assumptions in the analysis of sufficient conditions.
After analysis of necessity relations, we proceed with the construction
and examination of the truth table for the positive outcomes.10 Again, we find
three groups of countries where the population is relatively highly involved in
Table 5.5 Results of the analysis of necessary conditions for positive outcomes

(high country levels of membership in groups to preserve or protect
environment): impact of societal institutions.
Causal conditions Inclusion Raw coverage
GDPpc2009 0.989 0.805

DBDem2010+ReligNoneESS 0.932 0.633
pro-environmental groups: 1) France; 2) Austria, Germany, Switzerland; and 3)

Belgium, Denmark, Finland, Norway, Sweden, United Kingdom (see Table 5.6).
In the last and largest group all three explanatory factors have high scores. The
situation is a bit different in German-speaking countries – Austria, Germany,
Switzerland – and France. In the former group, religion is a rather important aspect
of societal fabric (they are the Western-European countries with the largest share
of religious population) and in France, inclusiveness of the democratic institutions
is relatively low (lowest among the analysed Western-European democracies).
On the other hand, negative outcomes are produced by quite different com-
binations of explanatory factors. However, in all of these cases wealth of the
country is relatively low, with the exception of Spain. In this country the other two
explanatory conditions – inclusiveness of democratic institutions and detachment
from religion – exhibit low scores. Quite the opposite situation occurs in Latvia
and Slovenia, where the population is very non-religious and has inclusive demo-
cratic institutions, however, the countries are relatively poor. This combination of
explanatory factors still produces the negative outcome (low level of membership
in groups to preserve or protect environment). In the rest of the cases, only one
of the explanatory conditions is present or none are present at all (as in Croatia
Table 5.6 Existing societal institutions and membership in groups to preserve or

protect environment: complete truth table for positive outcomes (inclusion
cut-off = 0.91).
GDP DBD Relig Member Inclusion Cases

pc2009 em2010 NoneESS EnvGr
1 0 1 1 0.987 France
1 1 0 1 0.944 Austria, Germany, Switzerland
1 1 1 1 0.933 Belgium, Denmark, Finland,
Norway, Sweden,
United Kingdom
1 0 0 0 0.907 Spain
0 1 1 0 0.739 Latvia, Slovenia
0 1 0 0 0.548 Bulgaria, Slovak Republic
0 0 1 0 0.543 Czech Republic, Russia
0 0 0 0 0.388 Croatia, Lithuania
and Lithuania). As expected, this type of existing societal institution in a country

produces a negative outcome.
Lastly, we present results of the analysis of the sufficient conditions for posi-
tive outcomes, where only a complex solution is possible as we do not have
counter-factual combinations of causal conditions in the truth table (more
detailed information about the solution is provided in Fig. 5.4). A Boolean
minimization procedure produced the formula GDPpc2009 * (DBDem2010 +
ReligNoneESS) = > MemberEnvGr.11 Again this solution states very simi-

lar things about the impact of existing societal institutions on levels of public
involvement in pro-environmental groups that were discussed when presenting
the constructed truth table. It shows that there are two ways for a country to
have a considerable proportion of the public participating in pro-environmental
groups, both of which require that a country is quite wealthy. However, this
condition is not sufficient. Some countries (especially, Austria, Germany and
Switzerland) also need to have inclusive democratic institutions, and in others
(such as France) the population has to be quite non-religious. Therefore, we
Solution for positive outcomes (membership scores):

GDPPC2009*(DBDEM2010+RELIGNONEESS) => MEMBERENVGR
1.0 CHE
DNK
AUT
BEL
0.8
DEU FRA
SWE
MEMBERENVGR
GBR
0.6 NOR
FIN
0.4 SVN
ESP
CZE
0.2
HRV
SVK
RUS LTU
BGR
0.0 LVA
0.0 0.2 0.4 0.6 0.8 1.0

GDPPC2009*(DBDEM2010+RELIGNONEESS)
Figure 5.4 Impact of existing societal institutions on levels of their public involvement in

pro-environmental groups: results of the fsQCA.
Notes: Included simplifying assumptions (SA): none; Inclusion=0.857; Raw coverage=0.927
conclude that with regard to the impact of existing societal institutions on the
levels of public involvement in pro-environmental groups the main explanatory
factor seems to be the wealth of a country. It is a necessary factor although it is
not sufficient by itself. Additionally, a country either has to have a rather non-
religious population or inclusive democratic institutions in order for it to have a
large proportion of the population involved in pro-environmental groups.
Discussion and conclusions

Environmental activism and other expressions of environment friendly behaviour
have become a valuable part of our contemporary societies. Usually the research
in this field focuses on analysis of individual-level factors such as sociodemo-
graphic characteristics, social and political trust and other personal attitudes and
behaviours. However, a number of questions about environmental activism in a
cross-country comparative perspective are still understudied. In this chapter we
presented an analysis of cross-national variation of public involvement in pro-envi-
ronmental groups in 19 European countries. Contrary to some observations (Hadler
& Haller, 2011, 2013), we found that membership in groups to preserve or protect
the environment varies quite considerably in European countries (in some as low
as 1.2%; in others as high as 13.5%). Additionally, we see that the dividing line
between countries exhibiting high and low levels of pro-environmental activism
might be drawn along the line dividing Europe into its Western and Eastern parts.
The only exception here is Spain and it still needs to be seen in further empirical
studies (as only Spain was included in the analysis from the Southern-European
countries) if we can conclude that Southern Europe is more similar to Eastern (or
Western) Europe in terms of its patterns of pro-environmental activism.
We examined the impact of two groups of country-level factors on pro-
environmental activism separately: (1) pro-environmental awareness and
behaviours; and (2) existing societal institutions. In both cases, the results
of fsQCA showed that differences in levels of public involvement in pro-
environmental groups in analysed countries could be explained by several
combinations of explanatory factors. The most important factors are the com-
parative wealth of a country (measured as GDP per capita) and high levels
of public environmentally friendly behaviour (e.g. signing a petition about
an environmental issue, giving money to an environmental group, or taking
part in a protest or demonstration about an environmental issue) in a country.
However, these factors are not sufficient to produce the positive outcomes
(relatively high involvement of the population in groups to preserve or pro-
tect the environment) by themselves. High levels of public environmentally
friendly behaviours need to be complemented by either high salience of envi-
ronmental issues in a country or frequent engagement of its population in
private-sphere environmentally friendly behaviours. Similarly, the relative
wealth of a country has to be amplified by the existence of inclusive demo-
cratic institutions or the presence of a rather non-religious population.
These combinations of explanatory conditions are alternative ways of producing

high levels of pro-environmental activism. All of them are equally legitimate and
pronounced in a considerable number of analysed countries. They might be con-
sidered by pro-environmental groups in trying to get support and attract attention
to their causes. Some – such as existing societal institutions – are hard to influence.
However, others – such as salience of environmental issues or private-sphere envi-
ronmentally friendly behaviours – seem to be more amenable to change.
Acknowledgements
The article is based on findings from three interrelated research projects, financed
by the Research Council of Lithuania: (1) ‘Monitoring of social problems: imple-
mentation of the International Social Survey Program (ISSP)’, 2010–2011;
(2) ‘International Social Survey Programme: Monitoring of Lithuanian social
problems (ISSP-LT)’, 2012–2013; and (3) ‘International Social Survey Pro
gramme: citizenship, work orientations and social welfare in Lithuania (ISSP
LT-CIWO)’, 2014–2016. The research of V. Morkevičius was also carried out
as part of the activities of the Lithuanian membership of the European Research
Infrastructure Consortium ‘European Social Survey’ funded by the Research
Council of Lithuania (Contract No. MTI-01/14).
Notes
1 Of course, these sources contain explications of the potential pitfalls of applying QCA
in different research situations. However, they also provide ways of avoiding them.
2 Cronbach’s alpha for the included items equals 0.766.
3 As our data source we used the annually updated National Accounts Main Aggregates
database of the Economic Statistics Branch of the United Nations Statistics Division
(see data.un.org/Data.aspx?q=gdp&d=SNAAMA&f=grID%3a101%3bcurrID%3aUS
D%3bpcFlag%3a1; last update: 15 April, 2015).
4 For an elaborate description of conceptualization, operationalization and deriva-
tion of these functions, see www.democracybarometer.org/documentation_en.html.
Russia was assigned a score of 0, as it is not included in the dataset because it is not a
democratic country.
5 For details of measurement, see ESS4 (2008), edition 4.3; ESS5 (2010), edition 3.2;
and ESS6 (2012), edition 2.1 at www.europeansocialsurvey.org/data.
6 Six explanatory variables produce 64 combinations of causal conditions in the truth
table. Since we have only 19 cases to populate it, in the most favourable scenario we
would still have more than two thirds of empty rows in the truth table.
7 We use notation with tilde for writing Boolean expressions introduced into QCA
by Ragin (2000): name of the condition or outcome variable with tilde means its
negation.
8 Truth table and sufficiency analysis for the negative outcomes is not reported in detail
as the results are only complementary to the ones retrieved in the case of positive
outcomes.
9 Inclusion score of the solution: 0.917; coverage of the solution: 0.879.
10 Again, truth table and sufficiency analysis for the negative outcomes is not reported in
detail as the results are only complementary to the ones retrieved in the case of positive
outcomes.
11 Inclusion score of the solution: 0.857; coverage of the solution: 0.927.
References
Balžekienė, A & Telešienė, A 2011, ‘Explaining private and public sphere personal envi-
ronmental behaviour’, Socialiniai mokslai, vol. 74, no. 4, pp. 7–19.
Clements, JM, McCright, AM & Xiao, C 2014, ‘Green Christians? An empirical exami-
nation of environmental concern within the US general public’, Organization &
Environment, vol. 27, no. 1, pp. 85–102.
Dono, J, Webb, J & Richardson, B 2010, ‘The relationship between environmental activ-
ism, pro-environmental behaviour and social identity’, Journal of Environmental
Psychology, vol. 30, no. 2, pp. 178–186.
Dunlap, RE & Mertig, AG 1995, ‘Global concern for the environment: is affluence a pre-
requisite?’ Journal of Social Issues, vol. 51, no. 4, pp. 121–137
Dusa, A & Thiem, A 2014, Qualitative Comparative Analysis. R Package Version 1.1–4.
Available from: <http://cran.r-project.org/package = QCA>. [accessed 9 November 2015].
Fielding, KS, McDonald, R & Louis, WR 2008, ‘Theory of planned behaviour, identity and
intentions to engage in environmental activism’, Journal of Environmental Psychology,
vol. 28, no. 4, pp. 318–326.
Franzen, A 2003, ‘Environmental Attitudes in international comparison: an analysis of the
ISSP surveys 1993 and 2000’, Social Science Quarterly, vol. 84, no. 2, pp. 297–309.
Franzen, A & Meyer, R 2010, ‘Environmental attitudes in cross-national perspective:
a multilevel analysis of the ISSP 1993 and 2000’, European Sociological Review,
vol. 26, no. 2, pp. 219–234.
Franzen, A & Vogl, D 2013, ‘Acquiescence and the willingness to pay for environmen-
tal protection: a comparison of the ISSP, WVS, and EVS’, Social Science Quarterly,
vol. 94, no. 3, pp. 637–660.
Freymeyer, RH & Johnson, BE 2010, ‘A cross-cultural investigation of factors influencing
environmental actions’, Sociological Spectrum, vol. 30, no. 2, pp. 184–195.
GDP per capita, World Development Indicators. Available from: <http://data.worldbank.
org/indicator/NY.GDP.PCAP.CD>. [accessed 17 November 2014].
Givens, JE & Jorgenson, AK 2013, ‘Individual environmental concern in the world polity:
a multilevel analysis’, Social Science Research, vol. 42, no. 2, pp. 418–431.
Hadler, M & Haller, M 2011, ‘Global activism and nationally driven recycling: the influ-
ence of world society and national contexts on public and private environmental
behaviour’, International Sociology, vol. 26, no. 3, pp. 315–345.
Hadler, M & Haller, M 2013, ‘A shift from public to private environmental behavior:
Findings from revisited and extended’, International Sociology, vol. 28, pp. 484–489.
Hunter, LM, Hatch, A & Johnson, A 2004, ‘Cross-national gender variation in environ-
mental behaviors’, Social Science Quarterly, vol. 85, no. 3, pp. 677–694.
Inglehart, R. 1989, Culture Shift in Advanced Industrial Society. Princeton: Princeton
University Press.
Inglehart, R 1995, ‘Public support for environmental protection: objective problems and
subjective values in 43 societies’, Political Science and Politics, vol. 28, pp. 57–72.
ISSP Research Group 2012, International social survey programme: Environment III – ISSP
2010. Gesis Data Archive, Cologne. ZA5500 Data file version 2.0.0, doi:10.4232/1.11418.
Kim, S, Choi, SO & Wang, J 2013, ‘Individual perception vs. structural context: searching
for multilevel determinants of social acceptance of new science and technology across
34 countries’, Science and Public Policy, vol. 41, pp. 44–57.
Kvaløy, B, Finseraas, H & Listhaug, O 2012, ‘The publics’ concern for global warming: A
cross-national study of 47 countries’, Journal of Peace Research, vol. 49, no. 1, pp. 11–22.
Lenox, MJ & Eesley, CE 2009, ‘Private environmental activism and the selection and
response of firm targets’, Journal of Economics & Management Strategy, vol. 18,
no. 1, pp. 45–73.
Longhofer W & Schofer E 2010, ‘National and global origins of environmental associa-
tion’, American Sociological Review, vol. 75, no. 4, pp. 505–533.
Merkel, W, Bochsler, D, Bousbah, K, Bühlmann, M, Giebler, H, Hänni, M, Heyne, L,
Müller, L, Ruth, S & Wessels, B 2014, Democracy Barometer. Methodology. Version
4. Aarau: Zentrum für Demokratie.
Ragin, CC 1987, The Comparative Method: Moving Beyond Qualitative and Quantitative
Strategies. Berkeley: University of California Press.
Ragin, CC 2000, Fuzzy-Set Social Science. Chicago: University Chicago Press.
Ragin, CC 2008, Redesigning Social Inquiry: Fuzzy Sets and Beyond. Chicago: University
of Chicago Press.
R Core Team 2015, R: A Language and Environment for Statistical Computing.
R Foundation for Statistical Computing, Vienna, Austria. Available from: <www.R-
project.org>. [accessed 17 November 2014].
Reyes, JAL 2013, ‘Cross-section analyses of attitudes towards science and nature from
the International Social Survey Programme 1993, 2000, and 2010 surveys’, Public
Understanding of Science, vol. 0, no. 0, pp. 1–20.
Reyes, JAL 2014, ‘Environmental attitudes and behaviors in the Philippines’, Journal of
Educational and Social Research, vol. 4, no. 6, pp. 87–102.
Rihoux, B & Ragin, CC (eds.) 2009, Configurational Comparative Methods: Qualitative
Comparative Analysis (QCA) and Related Techniques. Thousand Oaks: Sage.
Saphores, JDM, Ogunseitan, OA & Shapiro, AA 2012, ‘Willingness to engage in a pro-
environmental behavior: an analysis of e-waste recycling based on a national survey of
U.S. households’, Resources, Conservation and Recycling, vol. 60, pp. 49–63.
Schneider, C & Wagemann, C 2012, Set-Theoretic Methods for the Social Sciences: A
Guide for Qualitative Comparative Analysis. Cambridge: Cambridge University Press.
Steg, L & Vlek, C 2009, ‘Encouraging pro-environmental behaviour: an integrative
review and research agenda’, Journal of Environmental Psychology, vol. 29, no. 3,
pp. 309–317.
Stern, PC 2000, ‘New environmental theories: toward a coherent theory of environmen-
tally significant behavior’, Journal of Social Issues, vol. 56, no. 3, pp. 407–424.
Thiem, A & Dusa, A 2013a, ‘QCA: A package for qualitative comparative analysis’, The
R Journal, vol. 5, no. 1, p. 87–97.
Thiem, A & Dusa, A 2013b, Qualitative Comparative Analysis with R: A User’s Guide.
New York: Springer.
Thøgersen, J 2006, ‘Norms for environmentally responsible behaviour: An extended tax-
onomy’, Journal of Environmental Psychology, vol. 26, no. 4, pp. 247–261.
Whitmarsh, L & O’Neill, S 2010, ‘Green identity, green living? The role of pro-environmental
self-identity in determining consistency across diverse pro-environmental behaviours’,
Journal of Environmental Psychology, vol. 30, no. 3, pp. 305–314.
Chapter 6
Petitioner, contributor, protester

The profile of Europeans performing
different ‘public-sphere’
pro-environmental behaviours
Iosif Botetzagias, Nikoleta Jones and

Chrysovalantis Malesios
Introduction
According to Stern (2000) the vast array of pro-environmental behaviours may be
summarized into four major types. The first, labelled ‘environmental activism’,
relates to active involvement in environmental organizations and protests. The
second, ‘non-activist public behaviours’, includes contributing money to envi-
ronmental organizations, signing petitions in favour of environmental protection,
or willingness to pay higher environmental taxes. The next cluster includes
‘private-sphere environmentalism’ relating to ‘the purchase, use, and disposal
of personal and household products with an environmental impact’ (Stern, 2000,
p.409). Finally, the ‘other environmentally significant behaviours’ include the
pro-environmental decisions and actions individuals take in the context of their
professional lives.
Past research has demonstrated that these types (e.g. Dietz, Stern and
Guagnano, 1998; Stern et al., 1999), and to a certain extent even the individual
behaviours within each type (e.g. Urban and Scazny, 2012; Botetzagias, Malesios
and Poulou, 2014; for energy-saving behaviours), depend on a different com-
bination of predictors or factors. Overall, four broad factor categories can be
highlighted (cf. Stern, 2000):
1 ‘attitudinal factors’, including norms, values and beliefs; in short, the socio-
psychological predictors;
2 ‘contextual factors’, such as interpersonal and institutional influences, the
existence of a supportive framework, both physically and politically, the
generic socio-economic context etc.;
3 ‘personal capabilities’ for performing the behaviour, which are related to an
individual’s socio-demographic characteristics to a considerable extent;
4 ‘habit or routine’.
In this chapter, we will establish the differentiating characteristics of those per-

forming each of three ‘public-sphere’ pro-environmental behaviours: signing
134 losif Botetzagias et al.
a petition on the environment, giving money to an environmental organiza-

tion and taking part in a protest or demonstration about an environmental
cause. The focus will be on these three behaviours, out of a host of possible
others, because we agree with Stern, who notes that, ‘Although these behav-
iors affect the environment only indirectly, by influencing public policies,
the effects may be large, because public policies can change the behaviors
of many people and organizations at once’ (Stern, 2000, p.409). While most
of the available research has grouped these behaviours (in some cases along
with many others) under single, umbrella factors – e.g. ‘intermediate group
environmental collective action’ (Duit, 2010); ‘activism’ (McFarlane and
Hunt, 2006; Marquart-Pyatt, 2011); ‘individual political action’ (Blake, 2001)
– they represent different types of civic engagement for protecting the envi-
ronment, thus they are likely to depend on different predictors. To state the
obvious, the well-meaning individual, who writes concerned letters to their
local MP, or signs a petition on environmental issues, should be a rather differ-
ent beast from the committed activist who takes to the streets, investing their
time or even risking confrontation with the police. Actually, in the few cases
where these behaviours have been examined separately, they were found to
depend on different predictors (e.g. Stern et al., 1999; Wakefield et al., 2006).
Accordingly, this line of research will be taken one step further by studying
each of these behaviours in their exclusivity, focusing on those individuals
who had performed only one of these behaviours. Each of these behaviours
represents a different level of commitment, requires different resources and
entails different costs/benefits for the individual. For this reason, grouping, for
example, person X (who only signed a petition) with person Y (who signed a
petition and contributed to an environmental non-governmental organization;
ENGO) and studying them as two representatives of the same ‘sign a petition’
type of behaviour runs the risk of reaching erroneous results.
Thus, in this chapter we aim to offer further evidence concerning the dis-
tinctiveness of ‘public-sphere’ environmental behaviours. Although it follows
existing research, and in particular, that using ISSP data (e.g. Franzen, 2003;
Freymeyer and Johnson, 2010; Marquart-Pyatt, 2011; Franzen and Vogl, 2013),
our analysis differs from existing research in certain crucial ways. First, we will
study each behaviour in its exclusivity. Second, we will study and compare dis-
tinct behaviours and not a composite factor of ‘environmental concern’, similar
to Freymeyer and Johnson (2010), and unlike Franzen and Vogl (2013), Franzen
(2003) and Marquart-Pyatt (2011). Third, in this comparison we will employ
both micro-(individual; as Freymeyer and Johnson, 2010) and macro-(country)-
level predictors. Fourth, we make use of two ISSP waves – ISSP (2000) and
ISSP (2010) – in order to check for differences over time in the performance
of civic pro-environmental behaviours. Since past research demonstrated that
the influence of the different predictors on ‘environmental concern’ differs
between years (e.g. Dunlap and York, 2008; Botetzagias and Malesios, 2012), it
is important to check whether this is also the case for environmental behaviours
Petitioner, contributor, protester 135
in order to identify those characteristics which identify each of these behaviours

consistently over time.
Literature review
In this section we will discuss the available research concerning the relevance of
a number of individual and contextual level variables on the three different types
of public-sphere environmentalism: 1) signing of petition about a green issue;

2) giving money to environmental groups; and 3) taking part in a green protest or
demonstration.
Individual-level variables
While it seems tautological that people who worry more about the condition of
the natural environment, and of the possible impacts of its degradation, will be
more active in protecting it, available research has returned varied results. Thus,
Wakefield et al.’s (2006) study of Hamilton, Canada, found that environmental
concern had a positive effect on protesting for the environment yet not on donat-
ing money to environmental NGOs. Greenspan et al. (2012) reported a positive
influence of students’ ‘biospheric’ values on donating money, while Stern et al.
(1999) reported that concern for the environment, expressed in terms of accept-
ance of the New Ecological Paradigm – i.e. the view that humans have important,
adverse effects on a fragile natural environment (Dunlap and Van Liere, 1978;
Dunlap et al., 2000) – has no statistically significant effect on demonstrating
for the environment. However, Botetzagias and Van Schuur (2012) found that
‘post-materialism’ – i.e. the value-disposition which prioritizes quality of life and
civic participation issues over ‘materialistic’ concerns of material, financial and
personal security (cf. Inglehart, 1977) – is not a statistically significant predictor
of Green Party members’ activism (i.e. participating in demonstrations, organ-
izing meetings etc.). Thus the available research suggests that those performing
different public-sphere environmental behaviours will also have different levels
(or intensity) of environmental values/concern.
Regarding the role of environmental awareness and knowledge on environ-
mental activism, conflicting claims are presented in the available literature. Some
scholars have supported the fact that citizens who are aware of and are better
informed about environmental problems and threats tend to be more active (cf.
Stern, 2000; Lubell, 2002). However, higher levels of environmental knowl-
edge do not necessarily imply more engaged citizens (Kempton et al., 1996).
An important parameter in this context is the level of acceptance of scientific
knowledge which seems to determine, to a significant extent, the level of envi-
ronmental knowledge of individuals, and should also be taken into consideration
when exploring environmental activism.
A final note relates to the possible influence of other pro-environmental behav-
iours. Dubbed the ‘behavioural spill-over effect’, the basic idea is that individuals
who already perform some pro-environmental behaviours will be more likely to

engage in others, and/or more demanding ones. While there exists ample evidence
that pro-environmental behaviours are correlated, it is still unclear when, how,
and if at all one green behaviour leads to another (see Truelove et al. (2014) for a
recent review). Furthermore, since the correlation between different environmen-
tal behaviours increases with the (perceived) similarities between them (cf. Bratt,
1999; Thøgersen, 2004), it may well be the case that low intensity, ‘personal or
private-sphere’ and pro-environmental behaviours (such as recycling) will impact

on similar, less demanding ‘low-commitment active citizenship’ behaviours
(such as signing a petition or giving money to an environmental group), but not
on ‘committed public activism’ behaviours (such as demonstrating and protesting
about the environment; cf. Thøgersen & Crompton, 2009, p.144)
Another factor which may prompt an individual to participate in an activity
for the protection of the environment is the sense of efficacy and, in particular,
personal efficacy (Lubell, 2002). The sense that someone will make a differ-
ence by participating in collective actions, such as demonstrations and the
signing of petitions, can influence the decision to participate in such activities
(Finkel, Muller & Opp, 1989; Lubell, 2002).
Social trust has also been linked to environmental behaviour and activism.
Social trust refers mainly to the level of trust which exists among members of a
community or social group (Putnam, 2000), and it is considered that citizens who
tend to trust their fellow citizens are also likely to believe that other members of
the community will act for the protection of the common good (Pretty, 2003). As
a result, these citizens are expected to be more willing to contribute money for the
environment (Jones et al., 2009; Polyzou et al., 2010) and they are also more will-
ing to participate in actions for the protection of the environment (Jones, 2010)
such as demonstrations and petitions.
Perceptions of the national government’s efficacy in relation to environ-
mental issues is another parameter that may influence environmental activism.
Perceptions of governmental efficacy are connected to feelings of political trust,
and available research shows that citizens who tend to trust the government also
consider it more efficient in managing natural resources and as a result these cit-
izens may be more willing to contribute money for the environment (Polyzou
et al., 2010). Similarly, citizens who are confident about government efficacy are
also expected to be more environmentally active, arguably because they feel that
their voices will be heard through environmental activities, and thus their partici-
pation in specific actions will make a difference (Lubell, 2002).
Social networks are also considered an important parameter for environmen-
tal activism (Wakefield et al., 2006; Jones et al., 2010). However, an important
challenge is to define social networks and, most importantly, to measure them accu-
rately. One way of doing this is by examining memberships of formal organizations
within a community. Recent evidence has shown that being a member of environ-
mental non-governmental organizations (ENGOs) influences how much citizens
are willing to pay for the environment (Polyzou et al., 2009; Jones et al., 2010) and
also their level of environmental activism (McFarlane and Hunt, 2006; Wakefield
et al., 2006; Fielding et al., 2008). This can be attributed to the flow of informa-
tion relevant to environmental issues which is higher when someone is member
of an NGO (Jones and Clark, 2014). It can also be related to the fact that several
environmental groups are linked to specific environmental activities, in which their
members are expected to participate (Fielding et al., 2008): this refers both to activ-
ities, such as demonstrations and petitions, but also to the payment of money which
often comes as part of membership fees (ibid.).

Finally, an individual’s demographic characteristics have also been linked to
environmental activism. Starting with educational attainment, it is considered a
significant determinant of environmental activism (Oloffson and Ohman, 2006;
Freymeyer & Johnson, 2010), with recent studies showing that higher levels of
education positively influence the signing of a petition (Freymeyer and Johnson,
2010) or giving money for the environment (ibid.; Greenspan, Handy & Katz-
Gerro, 2012). Concerning an individual’s gender, there is no clear tendency of
how it relates to environmental activism (Tindall, Davies & Mauboules, 2003). It
has been suggested that women are less likely to be active than men but are more
environmentally responsible and more concerned about environmental issues
(Mohai, 1992). Wakefield et al. (2006) found that gender has no influence on
protesting about the environment or on donating money to ENGOs. The latter
finding is also corroborated by Greenspan et al. (2012) as well as Freymeyer and
Johnson (2010) – the latter further reporting that females are less likely to sign an
environmental petition.
Regarding income level, individuals who are financially more stable are more
likely to participate in actions regarding environmental problems (Freymeyer
and Johnson, 2010). Furthermore, in the environmental economics literature, it is
widely accepted that individuals with a higher income level are more willing to
contribute money for the environment (Mitchell and Carson, 1989). Yet one should
note that this was not the case for younger people (see Greenspan et al., 2012).
Available research shows that age has no (or, at best, a minimal) relation with
all three of the public-sphere environmental behaviours examined in this chapter
(cf. Wakefield et al., 2006; Freymeyer and Johnson, 2010). Regarding place of res-
idence, inhabitants of rural areas are less likely to sign a petition or donate money
(Freymeyer & Johnson, 2010); a finding congruent with the literature suggesting
that urbanites will be more active with regard to environmental issues because
the reasons for environmental grievances (poor environmental conditions and the
existence of polluters) are more visible in urban centres (Gillham, 2008), and/or
because living in an urban area offers the possibility for similarly minded indi-
viduals to come together and mobilize (ibid.; Dalton, 2005; Sonderskov, 2008).
Influence of national-level variables

Most of the research concerning the possible influences of contextual/national-
level variables on an individual’s pro-environmental concern and behaviours,
takes its cue from Inglehart’s (1995) ‘objective problems, subjective values’
hypothesis. In his article, Inglehart claimed that ‘environmental concern’, meas-
ured as an individual’s willingness to incur personal financial costs in order to
protect the environment, would be influenced through two distinct pathways:
on the one hand, a negative one, where poor environmental conditions would
lead to higher environmental concern; on the other, a positive pathway, where
post-material values (typical of affluent nations) are responsible for higher
environmental concern.
Subsequent research, which again focused predominantly on the willingness
to pay (WTP) for environmental protection, returned contradictory results con-
cerning the influence of environmental conditions and national affluence. At the
national level, Dunlap and Mertig (1995) found that a country’s affluence was
statistically significant and positively correlated with the national-level WTP; a
relation reaffirmed by Diekmann and Franzen (1999) and Kemmelmeier, Krol,
and Kim (2002) for a number of behaviours entailing personal financial sac-
rifices. Nevertheless, later research returned non-statistically significant results
(Franzen, 2003; Dunlap and York, 2008) while, more alarmingly, it pointed
towards a negative correlation between national affluence and the public’s
willingness to accept fiscal costs for the benefit of the environment (Dunlap
and York, 2008). These differences re-emerged when researchers focused on
an individual’s WTP/willingness to incur costs for the environment. While
Kemmelmeier, Krol, and Kim (2002) report a positive relation between national
affluence and an individual’s willingness to sacrifice part of their income for the
environment, Gelissen (2007) found a statistically significant negative correla-
tion, and Mostafa (2011) reports a non-statistically significant relation. As far as
national environmental conditions are concerned, Gelissen (2007) reports mixed
results for different types of environmental pollution, while Nawrotzki (2012)
found a negative, albeit not statistically significant, relation with WTP for the
environment. However, and more relevant to our own research, Freymeyer and
Johnson (2010) found that citizens of wealthier nations, and with better environ-
mental conditions, are more likely to ‘give money to an environmental NGO’.
Concerning the two other ‘public-sphere’ pro-environmental behaviours stud-
ied in this chapter (‘signing a petition’ and ‘demonstrating/protesting’ about an
environmental cause) the available research concerning the influence of macro
variables is quite limited. Freymeyer and Johnson (2010) found that national
affluence and poorer environmental conditions have a positive effect on ‘sign-
ing a petition’, while Botetzagias and Van Schuur (2012), studying Green Party
members, report that better environmental conditions actually increase the odds of
an individual protesting in favour of the environment. It is logical to assume that
national affluence would also have a positive impact on ordinary citizens demon-
strating, since wealthier nations are more likely to possess those socio-economic
characteristics which facilitate environmental political protest (cf. Dalton, 2005;
Dalton, Van Sickle, & Weldon, 2010). A similar finding should be anticipated
as far as political freedoms are concerned (Dalton, 2005; Dalton et al., 2010):
the more open and democratic the political structures in a country, the easier and
more likely it would be for its citizens to become involved in all kinds of ‘public-
sphere’ political actions, such as signing petitions and demonstrating.
Data
For our analysis, the data for European countries was used, originating from the
ISSP surveys from 2000 and 2010; all the individual-level variables originate
from the ISSP questionnaires.
Dependent variable: Our dependent variable originates from the following
ISSP question: ‘In the last 5 years have you: signed a petition about green issues,
given money to a green group, or taken part in a green protest or demonstration?’
for which the respondent could answer in a simple yes/no format. These behav-
iours will be studied in their ‘pure’ form. Thus, we are interested in individuals
who have, for example, only signed a petition and have not engaged in any of the
other two public-sphere behaviours. Therefore, respondents who have engaged in
more than one of these behaviours are not included in our analyses. The percent-
ages for each country and time period are shown in Table 6.1.
Explanatory variables
For our analyses, we use both single, observed variables, as well as latent items
(factors), which are constructed by combining the single variables readily avail-
able in the ISSP questionnaires. The statements used in the construction of the
latent items are measured, unless stated otherwise, on a five-point Likert scale
ranging from ‘1-Strongly Agree/ very willing/extremely dangerous’ to ‘5-Strongly
disagree/ very unwilling/not at all dangerous’. The option ‘Can’t choose’ was also
offered in the questionnaires, yet these answers were not included in the calcula-
tion of the latent variables.
Individual-level predictors
Apart from the standard demographic variables (age, gender, educational attain-
ment, place of residence and personal income; note that data for ‘income’ are not
available for year 2010), the following predictors were used in our analyses:
‘Environmental attitudes’: 4-item scale (reversed) (Cronbach’s α [year 2000 =
0.700; year 2010 = 0.656]; (To which extent do you agree?: We worry too much
about the future of the environment and not enough about prices and jobs today;
People worry too much about human progress harming the environment; There
are more important things to do in life than protect the environment; Many of the
claims about environmental threats are exaggerated).
‘Environmental risks’: 7-item scale (Cronbach’s α [2000 = 0.710; 2010 = 0.747]);
(How dangerous do you think the following are for the environment?: Air pollu-
tion caused by cars; Air pollution caused by industry; Pesticides and chemicals
Table 6.1 European citizens’ exclusive engagement in one out of three public-sphere pro-environmental behaviours over the past 5 years
(ISSP 2000; ISSP 2010).
Year 2000 Year 2010
Country/ N of Only signed Only given Only N of Only signed Only given Only
Behaviour respondents petition money protested respondents petition money protested
Pooled 18,950 8.1% 8.4% 0.7% 25,340 8.2% 5.8% 1.1%

Austria — — — — 1,019 11.9% 10.6% 0.8%
Belgium — — — — 1,142 12.7% 5.1% 0.9%
Germany 1,825 13.4% 4.1% 0.3% 1,623 8.9% 4.2% 0.2%
France — — — — 2,253 14.4% 3.6% 2.1%
UK 972 10.7% 5.6% 0.3% 928 10.3% 5.9% 0.1%
Ireland 1,232 10.6% 6.3% 0.6% — — — —
Netherlands 1,609 4.5% 27.3% 0.3% — — — —
Norway 1,452 5.7% 18.5% 0.5% 1,382 6.2% 12.9% 0.5%
Sweden 1,067 12.9% 12.1% 0.7% 1,181 9.5% 12.7% 0.8%
Czech R. 1,244 9.8% 5.1% 0.9% 1,428 7.6% 3.2% 1.3%
Slovenia 1,077 8.4% 7.3% 2% 1,082 7.2% 3.5% 0.7%
Slovakia — — — — 1,159 8.9% 5.1% 0.3%
Bulgaria 1,013 2.7% 1.2% 1.5% 1,003 5.4% 0.5% 1.4%
Russia 1,705 3.2% 0.9% 0.4% 1,619 2.5% 1.4% 0.6%

Spain 958 7.6% 1.8% 2.2% 2,560 7.6% 2.1% 3.2%
Latvia 905 7.4% 0.7% 1.4% 1,000 2.6% 3.1% 1.4%
Lithuania — — — — 1,023 4.3% 0.3% 1.2%
Portugal 959 2.6% 1% 0.5% — — — —
Denmark 1,069 8% 12.6% 0.7% 1,305 5.6% 9% 0.5%
Switzerland 735 13.6% 12.3% 1% 1,212 7.8% 14.1% 0.4%
Finland 1,128 12.1% 13.9% 0.2% 1,211 5.3% 16.7% 0.3%
Croatia — — — — 1,210 12.4% 1.5% 0.6%
Question: ‘In the last 5 years have you: signed a petition about an environmental issue, given money to an environmental group or taken part in a green protest
or demonstration?’
Source: ISSP (2000); ISSP (2010)
used in farming; Pollution of [COUNTRY’s] rivers, lakes and streams; A rise in

the world’s temperature caused by climate change; Genetic modification of certain
crops; Nuclear power stations).
‘Green lifestyle’: 2-item scale (Cronbach’s α [2000 = 0.67; 2010 = 0.53]);
(How often do you make a special effort to?: Sort glass, tins, plastic, or newspa-
pers, and so o,n for recycling; . . . Cut back on driving a car for environmental
reasons. Possible answers ranging from ‘1: Always’ to ‘4: Never’).
‘Post-materialism’: Dichotomous (Dummy) variable taking the values 0/1;

(Which two should be the country’s top priorities?: ‘Give people more say in gov-
ernment decisions’ and ‘Protect freedom of speech’ [post-materialist position];
‘Maintain order in the nation’ and ‘Fight rising prices’ [materialist position]; any
other combination of answers [mixed position]).
‘Personal efficacy’: 1-item variable; (To which extent do you agree?: It is just
too difficult for someone like me to do much about the environment). For our
analyses, the variable was re-coded into a 3-point Likert scale: (Strongly) Agree;
neither . . . nor; (Strongly) Disagree.
‘Social trust’: 2-item scale; (Cronbach’s α [2000: not asked; 2010 = 0.71]);
(items measured on 5-point Likert scales ranging from: ‘most people can be
trusted’ [to] ‘you cannot be too careful’; ‘most people are fair’ [to] ‘they try to
take advantage of you’).
‘Governmental environmental efficacy’: 1-item scale; (Some countries are
doing more to protect the world environment than other countries are. In general,
do you think that [COUNTRY] is doing: ‘more than enough’; ‘about the right
amount’; ‘too little’).
‘Membership of ENGO’: 1-item dichotomous variable (1: yes)
Macrolevel predictors
All macrolevel predictors refer to individual countries.
‘EPI-Environmental Conditions’: For measuring a country’s objective environ-
mental conditions for years 2000 and 2010, data originating from the Environmental
Performance Index (EPI) were used (YCLP, 2012) and, in particular, the
‘Ecosystems Vitality’ (EV) component. EV assesses a country’s natural environ-
ment conditions and the policies affecting it. It measures the national performance
on 7 environment-related domains (air pollution; water; biodiversity and habitat;
agriculture; forestry; fisheries; and climate), through 18 indicators ranging from
pollutants’ emissions and natural resources’ overexploitation to pesticide regulation
and renewable electricity generation within a given country.
‘Political Freedom’: An index score calculated for years 2000 and 2010, being
the average score of ten ‘political rights’ and fifteen ‘civil liberties’ indicators,
with each indicator scored between 0 (smallest degree of freedom) and 4 (greatest
degree of freedom) (Freedom in the World, 2014).
‘PPP GDP’: The country’s Gross Domestic Product (GDP) per capita for years
2000 and 2010 converted to international dollars, using purchasing power parity
(PPP) rates. An international dollar has the same purchasing power over GDP as
the US dollar has in the United States (World Bank, 2013).
Methods
A multinomial logistic regression analysis (Agresti, 2002) was conducted in order
to examine which of the scales (exploratory variables) has the greatest power
in discriminating between the respondents’ past pro-environmental behaviours,

as measured by means of the three mutually exclusive questions (demonstrat-
ing, signing a petition, contributing money to an environmental group). As a
reference category in the dependent variable we have chosen the arguably most
time-demanding type of the three public-sphere environmental behaviours –
‘demonstrating’ – in order to be able to easily compare the ensuing differences
with the two other less demanding types.
Thus, two logistic regression models were fitted to the data, one for each of the
corresponding years (2000 and 2010) with the available data. For the adaptation of
the final model and the estimation of the beta coefficients, the Maximum Likelihood
method was utilized. For selecting the final explanatory variables to include in the
final optimal models, backward selection was used as a suitable method, starting
with fitting a model with all the variables of interest and then dropping the least
significant variable, so long as it is not significant at our chosen critical level (5%).
Thus, in Tables 6.1 and 6.2 only those predictor variables which have achieved
statistical significance in distinguishing between the behaviours are presented. The
reader is reminded that, owing to a lack of data for the particular year in the respec-
tive ISSP datasets, the predictor ‘social trust’ was not included in the analysis for
the year 2000, while ‘personal income’ was not included in the analysis for 2010,
respectively. We acknowledge this as a shortcoming of our analysis.
Results
Tables 6.2 and 6.3 show the results of the analysis for the years 2000 and 2010,
respectively, including parameter estimates, along with their corresponding
significances (p-values). For increased clarity, in the tables we only report the
parameter estimates of those variables which are statistically significant in dis-
tinguishing between any two of the three public-sphere behaviours. Also, in the
last column of Tables 2 and 3, we present separately the odds ratios (OR) of the
models for each one of the statistically significant predictor variables (i.e. the ratio
of probabilities of the ‘signing petition’ and ‘contributing money to an environ-
mental group’, using as a reference group the ‘demonstrating’ category). An OR
value greater than 1 means that the probability of having performed behaviour X
is greater than having performed the reference behaviour; i.e. ‘Demonstrating’.
Thus, in Table 2 we see that, for the respondents in the year 2000, those who have
‘environmental attitudes’ are 1.339 times more likely to have ‘signed a petition’
than ‘demonstrated’ (and 1.17 times more likely to ‘have signed a petition’ than to
have ‘given money to an environmental group’). In contrast, an OR value smaller

than 1 indicates that the probability of having performed behaviour X is smaller
than having performed the reference behaviour of ‘demonstrating’. Accordingly,
in Table 3 we find, for example, that for respondents in year 2010 those who
have a high perception of the ‘environmental risks’ surrounding a number of
activities are 0.831 times less likely to have ‘signed a petition’ rather than to
‘have demonstrated’ about the environment. On the other hand, the respondents’
perceptions of ‘environmental risks’ do not differentiate (i.e. the differences are

statistically non-significant [n.s.]) between those who ‘demonstrated’ and those
who ‘contributed money to an environmental group’.
The results and insights of both tables are also reported in Table 4 in a com-
parative format.
Table 6.2 Discriminant characteristics of European citizens who have engaged

exclusively in one out of three public-sphere environmental behaviours over
the period 1996–2000.
Logistic regression results, Reference group (/behaviour): ‘Demonstrated
for environmental cause’)
X Beta coefficient p-value Odds Ratio

(b) (OR)
Signed petition for Intercept -1.989 0.018**

environmental
cause Membership of ENGO (ref.: No)
YES -0.473 n.s. 0.623
Post materialism (ref.: Mixed)

Materialism 0.033 n.s. 1.033
Post -0.95 0.001*** 0.387
materialism
Governmental Environmental Efficacy (ref.: too little)

More than -0.645 n.s. 0.525
enough
About the 0.405 n.s. 1.499
right
amount
Envir. Risks 0.133 n.s. 1.143
Envir. 0.292 0.012** 1.339
Attitudes
Place of residence (ref.: Rural)

Urban 0.631 0.007*** 1.879
Income -0.000001 0.008*** 1
EPI 0.059 0.002*** 1.061
PPP GDP 0.077 <0.001*** 1.08
X Beta coefficient p-value OR
Given money to Intercept -3.609 <0.001***

environmental
group Membership of ENGO (ref.: No)
YES 0.746 0.049** 2.108
Post materialism (ref.: Mixed)

Materialism 0.185 n.s. 1.203
Post -0.734 0.005*** 0.408
materialism
Governmental Environmental Efficacy (ref.: too little)
More than -0.05 n.s. 0.951
enough
About the 0.676 0.017** 1.966
right
amount
Envir. Risks 0.32 0.008*** 1.377
Envir. 0.157 n.s. 1.17
Attitudes
Place of residence (ref.: Rural)
Urban 0.541 0.023** 1.717
Income 0.000004 n.s. 1
EPI 0.076 <0.001*** 1.079
PPP GDP 0.092 <0.001*** 1.096
-2 Log likelihood 4134.91
Chi-square test 394.508 <0.001***
Nagelkerke’s R Square 0.183
Cox & Snell R Square 0.149
*Coefficient is significant at a 10%significance level; **Coefficient is significant at a 5%significance

level; ***Coefficient is significant at a 1% significance level; n.s.: Coefficient is non-significant.
Source: ISSP (2000); European countries, sample size 3,260 respondents
Only three predictors seem able to differentiate over time between the behav-
iours under examination. The first is rather straightforward: being a member of
an ENGO makes a person more likely to have contributed money to an environ-
mental group than having signed a petition or demonstrated. The other two are
national-level variables: living in a more affluent country and living in a country
with better environmental conditions both make it more likely that an individual
has signed a petition or donated money to the environment instead of demonstrat-
ing. Moreover, this holds true over time. We will return to the discussion of these
findings in the concluding section of this chapter.
Table 6.3 Discriminant characteristics of European citizens who have engaged
exclusively in one out of three public-sphere environmental behaviours over
the period 2006–2010.
Logistic regression results, Reference group (/behaviour): ‘Demonstrated
for environmental cause’)
X Beta coefficient p-value Odds Ratio (OR)

(b)
Signed Intercept -1.352 0.041**

petition for
environmental Membership of ENGO (ref.: No)
cause YES -0.511 0.033** 0.6
Envir. Risks -0.186 0.02 **
0.831
Envir. 0.216 0.004*** 1.241
Attitudes
Education -0.011 0.001*** 0.989
Political 0.351 0.037 **
1.42
Freedom
EPI 0.032 <0.001*** 1.033
PPP GDP 0.055 <0.001 ***
1.056
X Beta coefficient p-value OR
Given money to Intercept -4.143 <0.001***

environmental
group Membership of ENGO (ref.: No)
YES 0.749 0.001*** 2.114
Envir. Risks -0.08 n.s. 0.992
Enirv. 0.374 <0.001*** 1.453
Attitudes
Education -0.011 0.002*** 0.989
Political 0.771 <0.001 ***
2.163
Freedom
EPI 0.032 0.001*** 1.032
PPP GDP 0.09 <0.001 ***
1.094
-2 Log likelihood 6123.23
Chi-square test 441.6 <0.001***
Nagelkerke’s R Square 0.142
Cox & Snell R Square 0.118
*Coefficient is significant at a 10%significance level; **Coefficient is significant at a 5%significance

level; ***Coefficient is significant at a 1% significance level; n.s.: Coefficient is non-significant.
Source: ISSP (2010). European countries, sample size 3,826 respondents
Table 6.4 Comparison of 2000 and 2010 profiles of Europeans who have ‘donated money to an environmental group’ vs. ‘signed a petition’
vs. ‘demonstrated/protested for the environment’ over past 5 years.
Variable Respondent’s The 2000 European has . . . The 2010 European has . . .
characteristics
Signed petition Given money to Signed petition Signed petition Given money to Signed petition
instead of group instead of instead of instead of group instead of instead of giving
demonstrating demonstrating giving money to demonstrating demonstrating money to group
group
Membership of Is ENGO No difference Much more Much less Much less Much more Much less
ENGO member likely likely likely likely likely
Post materialism Is post- Much less Much less No difference No difference No difference No difference
materialist likely likely
Governmental Thinks govt. is No difference Much more Much less No difference No difference No difference
Environmental doing ‘about likely likely
Efficacy the right
amount’ in
protecting
global
environment
Environmental Considers a No difference Much more Much less Much less No difference Much less
Risks number of likely likely likely likely
activities
as ‘very
dangerous’
for the
environment
Environmental Recycles & cuts Much more No difference Much more Much more Much more No difference
Attitudes back on car likely likely likely likely
use
(continued)
Variable Respondent’s The 2000 European has . . . The 2010 European has . . .
characteristics
Signed petition Given money to Signed petition Signed petition Given money to Signed petition
instead of group instead of instead of instead of group instead of instead of giving
demonstrating demonstrating giving money to demonstrating demonstrating money to group
group
Place of Lives in urban Much more Much more No difference No difference No difference No difference
residence area likely likely
Income Has higher Much less No difference Much less Variable is missing
income likely likely
Political Lives in No difference No difference No difference Much more Much more No difference
Freedom country, likely likely
. . . with higher
civil liberties/
rights
EPI . . . with Much more Much more No difference Much more Much more No difference
better envir. likely likely likely likely
conditions/
policies
PPP GDP . . . with higher Much more Much more No difference Much more Much more No difference
GDP likely likely likely likely
Question: In last 5 years have you: signed a petition about an environmental issue, given money to an environmental group or taken part in a green protest or
demonstration?
Discussion and conclusion

This chapter sets out to examine the profiles of Europeans who have engaged
in three public-sphere environmental behaviours: signing a petition; contribut-
ing money to an environmental NGO; and participating in a pro-environmental
demonstration. Taking our cue from existing research, we anticipated that
each behaviour would relate to a different individual profile. Thus, in order
to identify who has been a ‘petitioner’, a ‘contributor’ or a ‘demonstrator’

for the environment, these behaviours are examined exclusively; i.e. we are
interested in individuals who have performed one of these behaviours and not
the other two. Furthermore, in order to assess whether the profiles remain the
same over time, the results are compared across two IPSS surveys, for the
years 2000 and 2010.
Our analyses show that each distinct behaviour is indeed related to different
individual and national characteristics, and there exists considerable temporal
variation. Thus, a number of variables turned out to be statistically insignificant
in distinguishing between the behaviours: leading a green lifestyle, one’s per-
ceived personal efficacy in protecting the environment and the level of social
trust all failed to differentiate between the three political pro-environmental
behaviours. Of the predictors which did prove statistically significant, some
returned contradictory results across the years, such as post-materialism, the
perceived efficacy of the national government’s environmental efforts, the
perception of environmental risks, having environmental attitudes, years of
schooling and place of residence.
Only a few variables returned the same results across time and between behav-
iours. At the microlevel, being a member of an ENGO means, rather unsurprisingly,
that a person is more likely to have contributed money to an environmental
group, while having stronger environmental attitudes makes an individual less
likely to have demonstrated instead of signing petitions or giving money. At the
macrolevel, living in a more affluent country which has better environmental
conditions, makes a person less likely to have demonstrated instead of signing
petitions or giving money. Why is this the case? Two plausible explanations come
to mind. On the one hand, citizens in ‘developed’ countries – economically, politi-
cally and environmentally (policy-wise) – have, overall, the objective political
and financial resources necessary to engage in public-sphere environment behav-
iours rather than demonstrating. Thus, a more advanced and democratic country
would arguably allow for the proliferation of civil-society groups, such as envi-
ronmental groups. Therefore, citizens in these countries have both the opportunity
and the financial means to express their pro-environmental concerns by support-
ing such environmental groups (however, Sonderskov (2008), who shows that this
nexus is not necessarily direct). Furthermore, a more democratic country offers its
citizens more pathways for promoting their demands, such as joining/supporting
issue-specific groups or signing petitions, thus demonstrating is just one of many
available options. On the other hand, the availability of these alternative avenues
of public-sphere environmentalism – and arguably their enhanced relevance in a

more democratic country – may give citizens the subjective feeling that they do
not have to incur the personal costs of active participation, but can protect the
environment through other middle-course ways (i.e. through signing a petition
or supporting an ENGO). Nevertheless, these are tentative arguments and further
research is needed to clarify the point.
So, who is the European environmental ‘petitioner’, ‘contributor’ and ‘demon
strator’? To start with, the performance of one public-sphere environmental

behaviour over the other two does not relate to a person’s age or gender, whether
they trust other people, or if they lead a green lifestyle. Having pro-environmental
attitudes, living in an affluent country which has better environmental conditions/
policies makes it more likely that they are a ‘petitioner’ than a ‘demonstrator’.
Being an ENGO member, having pro-environmental attitudes, living in an afflu-
ent country which has better environmental conditions/policies makes it more
likely that they are a ‘contributor’ than a ‘demonstrator’. And finally, being an
ENGO member and considering a number of human/industrial activities as risky
makes it more likely that they are a ‘contributor’ than a ‘petitioner’.
response response
Node 0 Node 0
Region Region
Adj. P-value = 0.000, Adj. P-value = 0.000,
Chi-square = 100.407, Chi-square = 22.302,
df = 2 df = 2
European non-European non-European European
Node 1 Node 2 Node 1 Node 2
Signed petition Signed petition

Contributed money Contributed money
Demonstrated Demonstrated
Figure 6.1 Comparison of differences between European and rest-of-the-world countries

concerning performance of three, exclusive, public-sphere environmental
behaviours. Results of decision-tree analysis for 2000 and 2010 data.
Source: ISSP (2000): European countries (N = 3,260 respondents); rest-of-the-world countries
(N = 1,586)
ISSP (2010): European countries (N = 3,826); rest-of-the-world countries (N = 2,251)
As a final note, we would like to compare the European findings to other coun-
tries around the world. Using the decision-tree procedure of the SPSS© software,
we compared the two blocks of countries (European vs. rest of the world), con-
cerning the performance of the three public-sphere environmental behaviours for
the years 2000 and 2010. The results, shown in Figure 6.1, show no statistically
significant differences between the two blocks (p-value < 0.001) thus contributing
to the growing evidence that we are witnessing a ‘globalization’ of environmental
concern and activism (cf. Dunlap & York 2008).
Acknowledgement
The authors would like to thank Philippa Harris for proofreading this chapter.
References
Agresti, A 2002. Categorical data analysis. London: Wiley.
Barkan, S E 2004, ‘Explaining public support for the environmental movement: a civic
voluntarism model’, Social Science Quarterly, vol. 85, no. 4, pp. 913–937.
Blake, D E 2001, ‘Contextual effects on environmental attitudes and behavior’, Environment
and Behavior, vol. 33 no. 5, 708–725.
Botetzagias, I and Malesios, C 2012, ‘The influence of economic affluence and environ-
mental conditions on an individual’s concern for the environment: a Greek case study
(2005–2007)’, Local Environment, vol. 1, pp. 93–113.
Botetzagias, I, Malesios, C and Poulou, D 2014, ‘Electricity curtailment behaviors in
Greek households: different behaviors, different predictors’, Energy Policy, vol. 69,
pp. 415–424.
Botetzagias, I and Van Schuur, W 2012, ‘Active Greens: an analysis of the determinants
of Green party members’ activism in environmental movements’, Environment and
Behavior, vol. 44, no. 4, pp. 509–544.
Bratt, C 1999, ‘Consumers’ environmental behavior: generalized, sector-based, or compen
satory?’, Environment and Behavior, vol. 31, no. 1, pp. 28–44.
Dalton, R J 2005, ‘The greening of the globe? cross-national levels of environmental group
membership’, Environmental Politics, vol. 14, no. 4, pp. 441–459.
Dalton, R, Van Sickle, A and Weldon, S 2010, ‘The individual-institutional nexus of
protest behaviour’, British Journal of Political Science, vol. 40, no. 1, pp. 51–73.
Diekmann, A and Franzen, A 1999, ‘The wealth of nations and environmental concern’,
Environment and Behavior, vol. 31, no. 4, pp. 540–549.
Dietz, T, Stern, P C and Guagnano, G A 1998, ‘Social structural and social psychological
bases of environmental concern’, Environment and Behavior, vol. 30, no. 4, pp. 450–471.
Duit, A 2010, ‘Patterns of environmental collective action: some cross-national findings’,
Political Studies, vol. 59, no. 4, pp. 900–920.
Dunlap, R E and Mertig, A G 1995, ‘Global concern for the environment: is affluence a
prerequisite?’, Journal of Social Issues, vol. 51, pp. 121–138.
Dunlap, R E and Van Liere, K D 1978, ‘The “new environmental paradigm”: a proposed
measuring instrument and preliminary results’, Journal of Environmental Education,
vol. 9, no. 4, pp. 10–19.
Dunlap, R E, Van Liere, K D, Mertig, A G and Jones, R E 2000, ‘Measuring endorsement

of the new ecological paradigm: a revised NEP scale’, Journal of Social Issues, vol. 56,
no. 3, pp. 425–442.
Dunlap, R E and York, R 2008, ‘The globalization of environmental concern and the limits
of the postmaterialist values explanation: evidence from four multinational surveys’,
Sociological Quarterly, vol. 49, no. 3, pp. 529–563.
Fielding K S, McDonald R, Louis W R 2008, ‘Theory of planned behaviour, identity and
intentions to engage in environmental activism’, Journal of Environmental Psychology,

vol. 28, pp. 318–326.
Finkel, S E, Muller, E N, & Opp, K D 1989, ‘Personal influence, collective rationality, and
mass political action’, American Political Science Review, vol. 83, pp. 885–903.
Franzen, A 2003, ‘Environmental attitudes in international comparison: an analysis of the
ISSP surveys 1993 and 2000’, Social Science Quarterly, vol. 84, no. 2, pp. 297–308.
Franzen, A and Vogl, D 2013, ‘Two decades of measuring environmental attitudes: a
comparative analysis of 33 countries’, Global Environmental Change, vol. 23, no. 5,
pp. 1001–1008.
Freedom in the World (2014), Available at: http://www.freedomhouse.org/report-types/
freedom-world#.VEDirmd_uWM, Accessed 17 October 2014.
Freymeyer, R H and Johnson, B E 2010, ‘A cross-cultural investigation of factors influenc-
ing environmental actions’, Sociological Spectrum, vol. 30, no. 2, pp. 184–195.
Gelissen, J 2007, ‘Explaining Popular support for environmental protection: a multilevel
analysis of 50 nations’, Environment and Behavior, vol. 39, no. 3, pp. 392–415.
Gillham, P F 2008, ‘Participation in the environmental movement: analysis of the European
Union’, International Sociology, vol. 23, no. 1, pp. 67–93.
Greenspan, I, Handy, F and Katz-Gerro, T 2012, ‘Environmental philanthropy: is it simi-
lar to other types of environmental behavior?’, Organization & Environment, vol. 25,
no. 2, pp. 111–130.
Harring, N 2013, ‘Understanding the effects of corruption and political trust on will-
ingness to make economic sacrifices for environmental protection in a cross-national
perspective’, Social Science Quarterly, vol. 94, no. 3, pp. 660–671.
Inglehart, R 1977. The Silent Revolution: Changing Values and Political Styles among
Western Publics. Princeton University Press: Princeton.
Inglehart, R 1995, ‘Public support for environmental protection: objective problems and sub-
jective values in 43 societies’, PS: Political Science and Politics, vol. 28, no. 1, pp. 57–72.
ISSP (International Social Survey Program) 2000, ‘“Environment II” – ZA No.3440’.
Available at: http://www.gesis.org/issp/modules/issp-modules-by-topic/environment/
2000/, Accessed 7 October 2015.
ISSP (International Social Survey Program) 2010, ‘“Environment III” – ZA No.5500’,
Available at: http://www.gesis.org/issp/modules/issp-modules-by-topic/environment/
2010/, Accessed 7 October 2015.
Jones, N 2010, ‘Environmental activation of citizens in the context of policy agenda
formation and the influence of social capital’, The Social Science Journal, vol. 47,
pp. 121–136.
Jones, N and Clark, J R A 2014, ‘Social capital and the public acceptability of climate
change adaptation policies: A case study in Romney Marsh, UK’, Climatic Change,
vol. 123, no. 2, pp.133–145.
Jones, N, Evangelinos, I K, Iosifides, T, Halvadakis, P C and Sophoulis, C M 2010, ‘Social
factors influencing perceptions and willingness to pay for a market-based policy aiming
on solid waste management’, Resources, Conservation and Recycling, vol. 54, no. 9,
pp. 533–540.
Kemmelmeier, M, Krol, G and Kim, Y H 2002, ‘Values, economics, and proenvironmental
attitudes in 22 societies’, Cross-Cultural Research, vol. 36, no. 3, pp. 256–285.
Kempton, W, Boster, J S and Hartley, J A 1996. Environmental Values in American
Culture, Cambridge, MA: MIT Press.
Lubell, M 2002, ‘Environmental activism as collective action’, Environment and Behavior,
vol. 34, pp. 431–454.

Marquart-Pyatt, S T 2011, ‘Explaining environmental activism across countries’, Society &
Natural Resources, vol. 25, no. 7, pp. 683–699.
McFarlane, B L and Hunt, L M 2006, ‘Environmental activism in the forest sector: social
psychological, social-cultural, and contextual effects’, Environment and Behavior,
vol. 38, no. 2, pp. 266–285.
Mitchell R.C., Carson, R.T., 1989. Using Surveys to Value Public Goods: The Contingent
Valuation Method. Washington DC: Resources For the Future.
Mohai, P. 1992. Men, women, and the environment: an examination of the gender gap
in environmental concern and activism. Society and Natural Resources, vol. 5, no. 1,
pp. 1–19.
Mondejar-Jimenez, J, Mondejar-Jimenez, J A, Vargas-Vargas, M and Gazquez-Abad,
J C 2012, ‘Personal attitudes in environmental protection’, International Journal of
Environmental Research, vol. 6, no. 4, pp. 1039–1044.
Mostafa, M M 2011, ‘Wealth, Post-materialism and consumers’ pro-environmental
intentions: a multilevel analysis across 25 nations’ Sustainable Development, vol. 21,
no. 6, pp. 385–399.
Nawrotzki, R J 2012, ‘The politics of environmental concern: a cross-national analysis’,
Organization & Environment, vol. 25, no. 3, pp. 286–307.
Olli, E, Grendstad, G and Wollebaek, D 2001, ‘Correlates of environmental behav-
iors: bringing back social context’, Environment and Behavior, vol. 33, no. 3,
pp. 181–208.
Olofsson, A, and Öhman, S 2006, ‘General beliefs and environmental concern: transatlan-
tic comparisons’, Environment & Behavior, vol. 38, no. 6, pp. 768–790.
Oreg, S and Katz-Gerro, T 2006, ‘Predicting proenvironmental behavior cross-nationally:
values, the theory of planned behavior, and value-belief-norm theory’, Environment
and Behavior, vol. 38, no. 4, pp. 462–483.
Polyzou, E, Jones, N, Evangelinos, I K, and Halvadakis C P 2010, ‘Willingness to pay
for drinking water quality improvement and the influence of social capital’, Journal of
Socio-Economics, vol. 40, pp. 74–80.
Pretty J 2003, ‘Social capital and the collective management of resources’, Science,
vol. 302, pp. 1912–1914.
Putnam R 2000. Bowling Alone: The Collapse and Revival of American Community, New
York: Simon & Schuster.
Reyes, J A 2013, ‘Cross-section analyses of attitudes towards science and nature from
the International Social Survey Programme 1993, 2000, and 2010 surveys’, Public
Understanding of Science, vol. 24, no. 3, pp. 338–357.
Sonderskov, K M 2008, ‘Environmental group membership, collective action and general-
ised trust’, Environmental Politics, vol. 17, no. 1, pp. 78–94.
Stern, P C 2000, ‘New environmental theories: toward a coherent theory of environmen-
tally significant behavior’, Journal of Social Issues, vol. 56, no. 3, pp. 407–424.
Stern, P C, Dietz, T, Abel, T, Guagnano, G A and Kalof, L 1999, ‘A value-belief-norm

theory of support for social movements: the case of environmentalism’, Human Ecology
Review, vol. 6, no. 2, pp. 81–97.
Thøgersen, J 2004, ‘A cognitive dissonance interpretation of consistencies and inconsisten-
cies in environmentally responsible behavior’, Journal of Environmental Psychology,
vol. 24, no. 1, pp. 93–103.
Thøgersen, J and Crompton, T 2009, ‘Simple and painless? The limitations of spillover in
environmental campaigning’, Journal of Consumer Policy, vol. 32, no. 2, pp. 141–163.

Tindall, D B, Davies, S, and Mauboules, C 2003, ‘Activism and conservation behaviour in
an environmental movement: the contradictory effects of gender’, Society and Natural
Resources, vol. 16, no. 10, pp. 909–932.
Truelove, H B, Carrico, A R, Weber, E U, Raimi, K T and Vandenbergh, M P 2014,
‘Positive and negative spillover of pro-environmental behavior: an integrative review
and theoretical framework’, Global Environmental Change, vol. 29, pp. 127–138.
Urban, J and Scazny, M 2012, ‘Exploring domestic energy-saving: the role of environmen-
tal concern and background variables’, Energy Policy, vol. 47, pp. 69–80.
Wakefield, S, Elliott, S, Eyles, J and Cole, D 2006, ‘Taking environmental action: the role
of local composition, context, and collective’, Environmental Management, vol. 37,
no.1, pp. 40–53.
World Bank 2013. GDP per capita, PPP (current international $), Available at: http://data.
worldbank.org/indicator/NY.GDP.PCAP.PP.CD, Accessed 17 October 2014.
YCLP 2012. Yale Center for Environmental Law and Policy, Yale University; Center for
International Earth Science Information Network (CIESIN), Columbia University; World
Economic Forum (WEF); and Joint Research Centre (JRC), European Commission. 2012.
2012 Environmental Performance Index and Pilot Trend Environmental Performance
Index. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC).
Available at: http://dx.doi.org/10.7927/H48913SG. Accessed 17 October 2014.
Part III
mindsets and behaviours

Exploring European green

Chapter 7
How far for the money?

Affluence and democratic
degrowth potential in Europe
Mladen Domazet and Branko Ančić
Introduction
The dominant socio-economic model of developed countries which is based
on the imperative of economic growth is leading societies to an unsustain-
able path marked by environmental, social and economic detrimental effects
within this century. A shift to the democratically based degrowth concept,
connected life practices and cultural patterns is a transformational scenario to
avoid a whole-scale collapse. It is important to stress at the outset that the term
‘degrowth’ is a provocative, performative label and not a literal description
of a political-economic strategy focused exclusively on the continued reduc-
tion of the GDP index. It is a call for the liberation of the public debate from
the narrow path of economism, in lieu of creating social structures that will
use fewer natural resources and enable radically different lives (cf. D’Alisa,
Demaria and Kallis 2015).
The main goal of this paper is to explore the extent of environmentally
motivated degrowth potential in Europe at the level of attitudes and behaviour.
Environmental motivations are not the only drivers of degrowth thinking, but
present an important subsection, especially in Europe (Demaria et al. 2013). In
addition, it explores the extent to which this potential is connected to the gen-
eral affluence of selected European countries. We use the ISSP research module
Environment survey data from 2010, comparing findings for 19 European ‘old’
and ‘new’ democracies against the remaining 13 countries participating in the
Environment module (ISSP 2012). Our findings indicate that there is a common
awareness of environmental limits to growth across our sample, but that there are
differences in evaluations of the potential to address these limits between ‘richer’
(core) and ‘poorer’ (semi-peripheral) states.
Degrowth works as [an environmentally beneficial strategy], but as an idea is

not a priority of governments and is not supported by public opinion.
(François Gemenne, lecture on climate change, Zagreb,
29th September 2014; part of ‘promotion’ for the COP21
UN Climate Change Conference, Paris, December 2015)
158 Mladen Domazet and Branko Ančić
A debate on the environmental consequences of continuous global economic

growth and related material consumption cannot be avoided in the context of
increasingly obvious extreme weather events around the world (Schneider 2008,
2010), and the fallout of the landmark UN Framework Convention on Climate
Change’s (FCCC) Conference of the Parties (COP) meeting held in Paris in
December 2015. The global strategy response can either call for a rapid and
radical transition to a ‘green economy’ that is significantly decoupled from
material use and carbon emissions in energy production (UNEP 2011a, 2011b;
Creech et al. 2014), or completely move the focus away from economic growth,
which has been advocated as the key societal instrument of life improvement
by many political subjects (Jackson 2009; Kallis 2011; Kallis, Kerschner and
Martínez-Alier 2012; Dietz and O’Neill 2013). Whilst the former strategy
also requires a social transition standardly accompanying a major technologi-
cal change (in energy production and distribution, for example; Polanyi 1968;
Pomeranz 2009; Domazet et al. 2014b), it is currently being hampered by a
lack of wide-scale technological replacement for existing material and energy
sources (Tienhaara 2009; Steinberger et al. 2013). But should it ever material-
ize, it promises to require less upheaval to the social structures through which
embodied solar energy is distributed than the latter. For much of the ‘developed
world’s’ population the ‘green economy’ option, decoupled from material use
and carbon emissions, is more akin to a business-as-usual scenario, where the
associated societal transformation is incremental and almost imperceptible. For
much of the ‘developing world’ it is a new name for the aspirational project of
development and technological catch-up, largely constructed out of an increase
in material pressure on the environment, of the type the ‘developed world’ now
takes for granted (Schaffartzik et al. 2014).
The strategy of turning away from economic growth as the necessary instrument
of life improvement – the approach broadly labelled as ‘degrowth’ – involves a
technological change as well, for the current energy-transformation technologies
and material-consumption volumes of individual societies are clearly unsustain-
able (Schaffartzik et al. 2014). A global population would have to dramatically
reduce the use of existing fossil-fuel energy, increase the efficiency and replace
it with renewable energy sources wherever possible (Jackson 2009). However,
such a transition is associated with a much more radical social transformation
than is the case of ‘greening’ the economy, for it conceptually decouples social
and personal development and improvement aspirations from an increase in eco-
nomic production and consumption (material or ‘dematerialized’) within a given
population. In simpler terms, it asks the population to be prepared to do with less
in terms of energy and material products.
Given that current material extraction and resource consumption are already
showing signs of reaching global environmental limits, and the ‘business-
as-usual’ scenario intends to take them to twice their current value within a
couple of decades even without population growth (Krausmann et al. 2009;
Schaffartzik et al. 2014), most present-day societies need a noticeable change
Affluence and democratic degrowth 159
in material throughput and associated expectations, aspirations, behaviours and

attitudes. Owing to vast inequalities inherent in global material production and
consumption (and energy embodied therein) this is truer of societies that are
the physical net importers. As we show below, most countries included in the
ISSP surveys fall into that group. In order to equalize access to energy and
resources for all present-day societies, for the wealthiest countries this includes
a sacrifice of existing material resources shaping individual and social well-
being, and a reduction in the associated extraction of resources and dumping

of waste (Barnett, Matthew and O’Brian 2010; Griggs et al. 2013). But the
current development model, even in societies with a very high level of human
development, bases improvements to wellbeing on growth of material con-
sumption and utilization of energy (Barry 2012), making the required sacrifice
supposedly impossible through democratic support (Meyer 2010). This is why
its associated social transformation is considered to be more radical than the
one accompanying projected adjustments to new low-carbon and low-material-
throughput technologies.
The term ‘degrowth’ (originally décroissance in French) solidified at the turn
of the century as a notion that societies can aim to shrink production and con-
sumption in order to approach social and ecological sustainability. It entails, but
is not reducible to, GDP decline, and expresses a general desire or direction for
a host of societal developmental metrics without focusing on a single variable
measure of success (Kallis et al. 2012). However, it has shied away from attempt-
ing to foster a global political, and as the experience of ‘sustainable development’
shows, false consensus (Demaria et al. 2013). Importantly, it is not an exclusively
economic concept. Conceptually it is an attempt to repoliticize the development
debate concerning the appropriate and socially acceptable mechanisms of socio-
ecological transformation. It affirms a critique of dominant world-development
models and explores locally contextualized alternatives. As mentioned, it is a cri-
tique of the current development model hegemony (Rist 2003; Sugihara 2003),
but does not on its own offer a replacement conceptual and scientific paradigm
that provides model problems and solutions for the scientific community. By
doing so it does not provide straightforward universal research instruments in the
social sciences, to serve as a measurement tool for the material and self-acclaimed
potential of countries for adopting degrowth strategies. A piecemeal mapping of
country potentials from the selected data sources is therefore as justified as any
other, opening a debate within the research community as to what the appropriate
mechanisms of the socio-ecological transformation might be in a given setting.
Conceptualizing degrowth for empirical research

As previous analyses show (Fournier 2008; Martínez-Alier et al. 2010; Demaria
et al. 2013) degrowth is a multifaceted concept comprising several philosophical,
research and activist currents within ecological and social thought. Nonetheless,
Demaria et al.’s (2013) overview identifies its ecological and democratic streams.
The ecological or environmental stream stresses the current competition and

future strategic trade-off between ecosystems and the industrial production and
consumption systems. Here, degrowth is seen as a possible path to preserve the
vitality of ecosystems in the living environment, by reducing the impact that
societal reproductive and expansive activities have on ecosystems, as well as chal-
lenging the idea that the decoupling of impacts and economic growth is possible.
The democratic stream, on the other hand, addresses the lack of democratic debate
on the directions of economic development, growth, progress and technological

intervention into social metabolisms. It is primarily on these two streams that we
base the notion of degrowth addressed in this article. With respect to the current
understanding of degrowth as a multidimensional concept, incorporating activism
and research in justice, critiques of development and utilitarianism, well-being
and meaning of life, reformism and radical alternative practices, and range of
scales from local to global, the notion we operationalize in this text might be more
precisely termed ‘environmentally motivated democratic degrowth’ (Romano
2012). However, for the purposes of our analysis we use the term ‘degrowth’.
In developed countries popular motivations for degrowth-supportive attitudes
involve pro-environmental, frugal, altruistic and equitable orientations, and are
shown empirically to frequently overlap (De Young 1993; Schultz 2001; Hulme
2009; Corral Verdugo 2012; Hedlund-deWitt 2012). As is illustrated by Dietz
and O’Neill (2013), the argument for a degrowth orientation is, conceptually,
framed most readily in terms of an awareness of the breach of environmental
limits, and the need for maintaining the stability of the carrying capacity of the
regional and global environment, at least in societies with a high level of human
development (Kallis et al. 2012). In fact, without some awareness of global envi-
ronmental limits, arguments for degrowth motivated by a decrease in inequality,
institutional restructuring or cultural conservatism would lose urgency and, in
some cases, their rational foundations (Schneider 2010; Rundgren 2013). If
there is endless bounty in the extraction and consumption of physical resources
and subsequent waste disposal embedded in ecosystems, then material sacrifice
seems irrational for all those who can justify their needs as aspirational necessi-
ties rather than excesses (Kasser 2002).
The environmental module of ISSP is a rich source of data for the analyses
of attitudes mentioned above. Previous comparative analyses of differences of
mean scores of attitude indices within the set of 181 European countries (Dolenec,
Domazet and Ančić 2014; Domazet, Ančić and Brajdić Vuković 2014a) tested
primarily the so-called affluence thesis,2 as presented in Franzen and Meyer
(2010). The affluence thesis (Diekmann and Franzen 1999; this text uses the term
‘prosperity thesis’) refers to a strong correlation between national wealth and
environmental concern, showing that national populations with higher national
average wealth (as measured by GDP) also express higher levels of environmental
concern (Franzen and Meyer 2010). Furthermore, whilst Franzen and Vogl (2013)
find that environmental concern also depends on individuals’ sociodemographic
characteristics such as age, gender, education, and income, as well as general
trust in others, party affiliation and post-materialistic values, the connection with
national GDP remains prominent across countries and ISSP findings of 1993,
2000 and 2010. This remains a highly problematic finding for the social trans-
formation required for degrowth and the associated prospect of conserving the
carrying capacity of the planet, as Franzen and Vogl (2013) themselves warn. As
Steinberger et al. conclude ‘[the] idea to grow first and to deal with environmental
issues later has been proven false empirically’ (Steinberger et al. 2013, p. 9).
Domazet, Ančić and Brajdić Vuković (2014a) and Dolenec, Domazet and
Ančić (2014) disaggregated the complex environmental concern index, used
in justification of the affluence thesis, into several components partially over-
lapping with Franzen and Meyer’s (2010) component measures. Even when
including inequality adjustments in contextual indicators replacing GDP we
found that the affluence thesis was upheld on indices of individual concern and
activation among the European countries analysed (cf. Domazet et al. 2014a).
When describing both their individual awareness of the scope of environmen-
tal problems and efficacy in implementing appropriate lifestyle and behaviour
changes, the European countries’ sample averages correlate strongly with a
measure of national inequality-adjusted income. The correlation is even stronger
between the said income measure and expressed willingness among populations
to commit to material sacrifices and support policies that demand greater mate-
rial investments from them (Domazet et al. 2014a). However, when indicators
focused more on commitments along general societal development trends, rather
than when individuals’ personal values and behaviour are considered against the
same measures of wealth, a deviation away from the linear correlation of afflu-
ence thesis is recorded. It is to these latter indicators we turn to for a more careful
analysis in this text. It is precisely a deeper understanding of differences of these
non-market, political dimensions of commitment to pro-environmental strategies
among European populations that is required for an assessment of democratic
degrowth potential (cf. Meyer and Liebe 2010; Whitmarsh and O’Neill 2010).
The effect of the affluence thesis is undeniably strong in the European sample,
and presents a sizeable problem for a social transformation to stay within environ-
mental limits, and thus the carrying capacity of the planet. But, as stated above, the
degrowth orientation does not consist solely of the immediate willingness to sacri-
fice, regardless of material and social context and collective instruments available
in the overall population. After all, further analyses on the same dataset showed
that inequality is a major obstacle to degrowth transformation within and between
societies (Dolenec, Domazet and Ančić 2014), and that pro-environmental atti-
tudes are present among less prosperous European populations, and are dampened
by a social paradigm combining economic growth and techno-optimism regarding
the transcendence of environmental limits (Brajdić Vuković 2014). Finally, on
the variables deemed as indicative of a potential degrowth orientation among the
sample we compare the national averages of European ISSP samples with those of
the remaining countries in the ISSP. This helps situate the respective populations
into the global context as much as possible within the available dataset.
Research design
In this chapter we aim to identify the possible sociodemographic predictors
of the degrowth orientation among the European sample surveyed in the ISSP
Environment module, and compare the findings for European ISSP countries
with the findings for the remaining ISSP countries. Whilst individual nation
states are still key political units for enacting transitions to a more sustainable
existence, from a global perspective, regional and continental grouping allows

for a more realistic positioning on material and resource consumption, energy
and food security, and so on.
Almost all the countries presented in this comparative analysis are countries of
high and very high level of human development (UNDP 2011). Of 19 European
countries,3 only Bulgaria and Russia fall into the group of high rather than very
high human development, whilst of the remaining 13 Rest-of-the-world (RoW)
countries4 only the Philippines and South African Republic share a medium level
of human development (HDI; UNDP 2011). As development models, dominant
social paradigms and prevalent cultural trends differ among global national popu-
lations we aim to situate the sample of the European countries’ populations within
the current global context. We expect that developmental and cultural differences
are smaller among neighbouring European populations than they are among other
countries participating in this ISSP module. Russia is included in the European
group of countries in line with the other analyses in this volume, and is classified
as an Eastern-European country by the Organization for Economic Cooperation
and Development (OECD). Despite the fact that Russian sovereign territory cov-
ers a large segment of the Asian part of the Eurasian landmass, its population is
densest in European Russia, and culturally that is the region most strongly repre-
sented in the ISSP survey used here. Furthermore, the European sample is divided
into 12 core5 and 7 semi-peripheral6 countries (cf. ‘Results’ below), using the
inequality-adjusted affluence measure as the proxy instrument of core-periphery
distinction (Arrighi and Drangel 1986).
‘Core’ and ‘semi-periphery’ are not labels of normative value of economic
performance or level of development of regions or countries, and are certainly not
meant to promote or demote, or even offend certain national populations. They
are also not fixed incontestable labels for a given individual country, but are used
here as proxy indicators of certain nature-society interactions in the longer devel-
opment trend and world system. They are functional labels based on Wallerstein’s
world-system theory (Wallerstein 1979). Like many concepts employed in the
‘straddling’ of established academic disciplines, world-system categories are
much maligned and, at times, ambiguous. They are nonetheless useful as onto-
logical manifolds in the narratives addressing ‘wicked’ socio-environmental
challenges (Domazet and Marinović Jerolimov 2014). World systems are inter-
societal networks in which material flow and cultural interactions (economy,
conflict, migration, culture etc.) are important for the reproduction of the internal
structures of the components of which the system consists. What is even more
important is that these interactions affect the changes that occur within the local
structures, and put forth a possible explanation for the differences in processes
of nature-society interactions observed in different varieties of these components
(Chase-Dunn and Hall 1997).
From that perspective it is indubitable that there are regions of core and periph-
ery in contemporary European geography and associated societal metabolisms.
A core is a region of the world system that dominates the system and consists
of the most complex social groups in the system. Although the contemporary
human population is mostly united in a globalized world system, for historical
and metabolic reasons outlined above, we can focus on Europe as a sufficiently
emergent world system to illustrate the relationships between its components.
Periphery is a region dominated by the core and semi-periphery, and has the
least complex social groups in the system, as well as providing the greater
flow of natural resources to the core-semi-periphery segment than vice versa.
Semi-periphery is an intermediate location in an intersocietal core-periphery
structure, and was added as a further category to Wallerstein’s original world-
system structure by Chase-Dunn and Hall (1997). The core-periphery structure
is the fundamental form of any modern world system and can last longer in time
regardless of national development. It can also be altered by radical develop-
mental turns and that allows for regrouping of societies under categorical labels,
but mostly regions ‘run hard in order to stay in the same place’ within the world-
system structure through the process of historical development (Chase-Dunn
and Hall 1997).
Whilst Chase-Dunn and Hall (1997, p. 37) describe at length the controversy
around the definition of semi-periphery in the modern world system, for our pur-
poses it is significant that a description of societies mixes both core and peripheral
forms of organization and nature-society interaction that can be located between
two or more competing core regions, and that have transformative potential to
influence the core regions in the way that peripheral regions alone do not. This
makes the national populations, labelled as semi-peripheral in this text, poten-
tially transformative for the overall European world system (without normative
determination of the direction of that transformation), as well as mediating
between European core and peripheries in a historical and geographical sense
(e.g. in recent conflict and migration). They transmit material flows between core
and peripheral regions in Europe, and exhibit social and cultural institutional fea-
tures intermediate in form between adjacent core and peripheries (Domazet and
Marinović Jerolimov 2014). For the purposes of this text it is a suggestive label,
not a predictive determinant, at least until further research into a combination
of cultural patterns (e.g. attitudes expressed here) and material flow accounts is
conducted (cf. Schaffartzik et al. 2014).
For the purpose of our analysis we use two composite measures constructed
on the basis of the ISSP dataset, named the Attitudes of Concern and Activation
(ACA) index and the Attitudes of Material Sacrifice (AMS) index, following
Domazet et al. (2014a), as well as a set of four single variable indicators
of agreement/disagreement with general statements concerning economic
growth-environment trade-off.
The composite measure Attitudes of Concern and Activation (ACA) includes
measures of disagreement with statements (a) through (e)7 in Table 7.1. Higher
performances on this composite indicator suggest a sample with a higher propor-
tion of individuals who say that they know what effect their lifestyles have on
the environment, as well as how disquieting those effects can be. They also see
themselves as efficacious in implementing appropriate behavioural and lifestyle
changes based on their intrinsic motivation, and are thus willing to activate on
behalf of environmental protection within their society. This activation, nonethe-
less, focuses on individual actions (‘it is not the case that doing something is too
difficult, and it is not the case that there is no point in me doing something if others
won’t’; cf. Table 7.1), nor necessarily on changes in social structures and instru-
ments of material extraction and social metabolism.
Table 7.1 Instruments: Attitudes of Concern and Activation (ACA) index, Attitudes of

Material Sacrifice (AMS) index, active degrowth indicators, passive degrowth
indicators.
Indexes Items
‘How much do you agree or disagree with each

of these statements?’a
Attitudes of concern and a ‘there are more important things to do in
activation index (ACA) life than protect the environment’
reliability coefficient (Cronbach’s b ‘many of the claims about environmental
alpha) for the scale for 18 threats are exaggerated’
countries varied from 0.59–0.77
0.35–0.54 inter-item correlation c ‘it is just too difficult for someone like me
to do much about the environment’
d ‘there is no point in doing what I can for
the environment unless others do the
same’
e ‘I find it hard to know whether the
way I live is helpful or harmful for the
environment’
Attitudes of material sacrifice ‘How willing would you be to:’ b
index (AMS)
reliability coefficient (Cronbach’s a ‘pay much higher prices in order to
alpha) for the scale for 18 protect the environment’
countries varied from 0.73–0.89
0.56–0.80 inter-item correlation b ‘pay much higher taxes in order to protect
the environment’
c ‘to accept cuts in your standard of living in
order to protect the environment’
Single-Item Indicators Items
Active Degrowth Q10a ‘And how much do you agree or disagree with
indicators each of these statements?’a
‘We worry too much about the future of the
environment and not enough about prices
and jobs today’

Q11a ‘How much do you agree or disagree with:’a
‘Our country needs economic growth in
order to protect the environment’
Q11b ‘How much do you agree or disagree with:’a
Passive Degrowth
indicators ‘Economic growth always harms the
environment’
Q19c ‘How much do you agree or disagree with:’a
‘Economic progress in [COUNTRY] will
slow down unless we look after the
environment better’
a Measured on Likert scale: 1 = ‘strongly agree’; 2 = ‘agree’; 3 = ‘neither agree nor disagree’; 4 = ‘disagree’;
5 = ‘strongly disagree’
b Measured on Likert scale: 1 = ‘very willing’; 2 = ‘fairly willing’; 3 = ‘neither willing nor unwilling’;
4 = ‘fairly unwilling’; 5 = ‘strongly unwilling’
The composite measure Attitudes of Material Sacrifice (AMS) records the

willingness to: a) pay much higher prices; b) pay much higher taxes; and c) accept
cuts in one’s own standard of living,8 as presented in Table 7.1. It is thus a direct
expression of the diffusion within a particular state’s population of individuals
who are willing to make sacrifices and support policies that demand greater mate-
rial investments from individuals, whilst providing smaller material benefits for
them from resource extraction. The latter is what is required for the reduction of
the ecological footprint of individual societies and the overall global population
in the end, in order to bring them in line with the projected sustainable state.
On top of composite measures of individual commitment to environmental
protection and willingness for material sacrifice, we also use single variable indi-
cators of agreement/disagreement with general statements concerning economic
growth-environment trade-off. These report on the prevalence within individual
states of individuals who agree or disagree with statements whose affirmation or
denial is a lemma of ecological and democratic strands of the degrowth concep-
tualization. We divide these statements into those that express awareness of the
immediate trade-off between growth as we know it and environmental protection,
and those that express awareness of general ecological limits of growth. These are
called Active Degrowth and Passive Degrowth, respectively.
Active Degrowth is represented by two single-variable (attitude) indicators
(ISSP 2010 Q10a and Q11a). The first, concurrent with the Environment-economy
trade-off indicator in Domazet, Ančić and Brajdić Vuković (2014), exhibits the
prevalence of individuals within states’ populations, who disagree with the state-
ment that concern for the future of the environment is exaggerated and unjustifiably
prioritized over economic activities and progress. This is linked to considerations
of efficacy at the individual and collective levels, perceptions of which are vital to
one’s willingness to make degrowth commitments concerning economic activity
(Meyer 2010; Ostrom 2010). The second single-variable (attitude) indicator (ISSP
2010 Q11a) measures the normative side of the economy-environment trade-off.9
It captures the prevalence within states’ populations of individuals who disagree

with a single-item statement that their country needs economic growth in order
to protect the environment. Whilst these variables do not in and of themselves
indicate a prevalence of support for full-blown economic and social degrowth
transformations (Kallis 2011; Demeria et al. 2013), they do indicate general
conceptual support for a shift away from the economic growth developmental
paradigm, according to which economic growth leads to remedying individual
scarcity and national environmental problems.
Passive Degrowth is represented by two single-variable (attitude) indicators
(ISSP 2010 Q11b and Q19c). The first captures the prevalence within national
populations of individuals who agree with a single-item statement that economic
growth always harms the environment.10 Such agreement does not signal that
in the growth-environment trade-off the environment should be favoured, but
expresses an awareness that such a trade-off exists and thus might indicate the
abandonment of hope in ‘green economies’ that would successfully and rapidly
decouple economic growth from environmental impact. Whilst less important to
efficacy of action, this awareness nonetheless presents a fertile ethical foundation
for considering future development strategies that would aim to balance ecologi-
cal concerns with democratic material aspirations.
The second is even more explicit in the expression of limits to growth aware-
ness; it reports on the percentage of individuals within national populations
expressing agreement with a statement that economic progress in their coun-
try will slow down unless better environmental protection is enacted. This is
an explicit expression of awareness that some environmental limits of growth
have been reached, and that further economic progress (though the term ‘growth’
is not used explicitly in this variable) depends on the protection rather than the
destruction of environmental conditions. This awareness, again, can be perceived
as a warning but not an expression of taking sides in the environment-growth
trade-off, as different environmental protection strategies may still be seen to be
compatible with or resultant from a focus on achieving economic growth.
Our study analyses the indicators described above against national values
of the Human Development Index (HDI), Inequality-adjusted Income Index
(III) and Ecological Footprint (EF). HDI is a composite measure of life expec-
tancy, literacy, education and standards of living in countries worldwide. It
is part of the spectrum of indices that provide a background of distinctions
between development trends and benefits in various countries (UNDP 2011).
The Inequality-adjusted Income Index (III) is an indicator of the respective
p opulations’ inequality-adjusted average national income (GNI, which is slightly

different from GDP), thus representing a given state’s level of affluence in our
analyses (UNDP 2011). Ecological footprint (EF) is conceptualized as a meas-
ure of the land and water required to sustain the material standard of living of a
given population, given the reliance on the dominant technology. It is a measure
of the human demand for extraction from the earth’s ecosystems; that is, of
the human utilization of natural capital contrasted with the planet’s ecological
capacity to regenerate (Chambers, Simmons and Wackernagel 2000). Globally,

the situation is not bright as the average global citizen has an eco-footprint of
about 2.7gha (global hectares per person), whilst the top sustainable limit is
1.8gha. However, the differences between individual citizens from global North
to global South are on a scale of almost one order of magnitude (i.e. 1 to10).
Results
All countries included in the ISSP Environment module belong to a group of
countries with very high and high HDI (0.711–0.955) according to the 2011 val-
ues, except for the Philippines and South Africa whose HDI is in the medium
HDI group (0.536–0.710) (UNDP 2011). That makes most (30 out of 32) of the
countries representatives of the group where further increases in GDP tend to
bring comparatively less benefits in human development, and thus countries for
which some form of degrowth is advisable from the perspective of joint action
to protect the global environmental commons (cf. Shue 1999). By the United
Nations Development Programme’s (UNDP) Inequality-adjusted Income Index
(III) values and estimates, the European countries in the sample occupy the top
half of spread of III values for the overall ISSP sample; likewise with values of
ecological footprint (EF; Global Footprint Network 2015). On the whole, viewed
from the comparative position on these indices, European countries are better
suited to a strategic orientation to degrowth than most of those in the group’s
remaining 13 countries.
Individual concern and sacrifice

Whilst this volume and this chapter are primarily concerned with European
results, it is interesting to comment here for the purposes of context, that the gen-
eral affluence thesis trend is not observed in the Rest-of-the-World (RoW) sample
of countries. There are no correlations comparable to those analysed in Domazet,
Ančić and Brajdić Vuković (2014), between national affluence (as measured by III
index), and indices of Attitudes of Concern and Activation (ACA) and Attitudes
of Material Sacrifice (AMS). Overall, on the ACA index the European sample is
more skewed to higher values and exhibits a higher mean value (16.01) than is
the case for the RoW sample (15.07). The difference is statistically significant
(Mann-Whitney test – U = 183707858.500; p<0.001). This is an index depict-
ing a proportion of individuals who say they know the effect their lifestyle has
on the environment and can implement lifestyle changes on the basis of intrinsic
motivation. Being, on average, a more prosperous group (mean III value of 0.700
vs. mean III value of 0.604) it could be expected that their average performance
on the AMS index is also higher. But somewhat surprisingly, especially from
the perspective of the affluence thesis, it is not. The mean value of the European
sample AMS index is 7.67, and for the RoW sample it is 8.12. The difference is
statistically significant (Mann-Whitney test – U = 255411851.500; p<0.001).
Given that the AMS index measures the diffusion within a population of indi-
viduals self-reporting their willingness to make material sacrifices, and support
polices that demand greater material investments from individuals, this finding
challenges the potential European championing of proactive degrowth globally.
A similar ranking is repeated on items taken to indicate general degrowth attitudes
unrelated to individual sacrifice; our measures of Passive and Active Degrowth.
On the whole RoW group scores higher than or as high as the European group.
The only exception is the denial of direct benefit of growth (ISSP 2010 Q11a) for
successful national environmental protection activities. In that case we see the
trend of ecological modernization in the perceived relationship between national
development attainment and environmental protection within national borders. In
the more developmentally and culturally diverse RoW group countries there is
less explicit disagreement with the statement that further economic growth will
contribute to better environmental protection within the country. Table 7.2 pre-
sents values for Europe and RoW samples.
At first glance this suggests a paradoxical position for the degrowth-attuned
and highly developed European populations, suggesting that even Americans
and South Africans are more degrowth-oriented according to available ISSP
measures. In particular, with respect to Passive Degrowth (Q11b and Q19c)
indicators, Europeans come across as techno-optimists, who do not see a histori-
cal connection between growth and environmental deterioration nor a breach of
environmental limits by unimpeded economic growth, compared to the diverse
sample of the RoW countries. We need to take a closer look at the composi-
tion of the European average values, and note again the great diversity of the
small RoW sample to ascertain what expectations and attitudes might drive this
seeming growth-apologetic trend in Europe. Most importantly, we need to sepa-
rate the growth-optimistic and growth-sceptical national populations in Europe
by a spectrum of degrowth measures developed in this analysis. We know from
related analyses that the European degrowth-compliant population is neither neg-
ligent nor irrelevant in size (Ančić and Domazet 2015), but differs in more and
less affluent European societies on the readiness to lead on degrowth proactive
policies. The distinction between ‘richer’ and ‘poorer’ Europeans in aspirational
choice, between technologically induced ‘green economy’ and socially induced
degrowth, might be driving down the overall European support for degrowth strat-
egies in this dataset.
Based on the support for the affluence thesis trend, established in Domazet
et al. (2014a) on ‘individual’ measures and its weakening on the further
Table 7.2 Mean values of indices and indicators for European and RoW groups of
countries participating in ISSP 2010 Environment module.
Active Degrowth Passive Degrowth

indicators indicators
(% disagree) (% agree)
III EF ACA AMS
about environment over jobs

Q11a Environment: protect

2011* 2011* index index
Q19c Economic progress
environmental protection
Q11b Economic growth:
harms the environment
(M) (M)
Q10a Worry too much
will slow down without

by economic growth
European 0.700 4.5 16.01 7.67 50.3% 32.3% 39.2% 63.8%
mean
RoW 0.604** 3.9*** 15.07 8.12 50.07% 24.7% 52.2% 77.3%
mean
*Mean values for independent variables (III and EF) are calculated as arithmetic mean of individual
country values, and not as mean value for the aggregate subsamples consisting of all the contribut-
ing countries’ populations, incomes, material consumption and bio-capacities.
**Values of III for Taiwan, New Zealand and South Africa estimated from the values of respective
variables of countries of similar rank in HDI and GNI ranking in UNDP (2011), as UNDP does not
provide III values for these ISSP countries.
***There is no EF value for Taiwan, and it is not included in this mean.
degrowth measures, and the dominant size of the European samples within the
overall set of countries participating in ISSP 2010/2011, it is interesting to note
a separation of the European group into a set of countries above the mean III
score (0.700) and those below it. The analyses in Dolenec, Domazet and Ančić
(2014) and Brajdić Vuković (2014) allow for slightly different separations
into subsamples of European countries with GDP above or below US$20,000
per capita; or those with the Low DSP-Low KSI11 combination and the rest.
These different separations to do not yield overall significantly different results
from those according to a simpler separation by above-below mean III score.
We shall thus point to differences in performance between a subsample of 12
European core countries and 7 European semi-peripheral countries, on the basis
of affluence as a proxy instrument of core-periphery hierarchy (Arrighi and
Drangel 1986).
The semi-peripheral European countries’ subsample has a marginally higher
mean III value (0.616) than the RoW sample (0.604), and the same average EF
(3.9gha). Their mean score on the ACA index (15.22) is similar to that of the
mean for RoW sample (15.07), and much lower than that of the core European
Table 7.3 Mean values of indices and indicators for European subsamples.
Active Passive
Degrowth Degrowth
indicators indicators
(% disagree) (% agree)
III EF ACA AMS
about environment over jobs

Q11a Environment: protect

2011* 2011* index index
Q19c Economic progress
environmental protection
Q11b Economic growth:
harms the environment
Q10a Worry too much
will slow down without

by economic growth
Core European 0.749 4.8 16.47 8.16 53.4% 39.1% 38.6% 60.9%
mean
Semi-peripheral 0.616 3.9 15.22 6.86 44.9% 21.6% 40.2% 68.2%
European
mean
*Mean values for independent variables (III and EF) are calculated as arithmetic mean of individual
country values, and not as mean value for aggregate subsamples consisting of all the contributing
countries’ populations, incomes, material consumption and bio-capacities.
subsample (16.47). More importantly, their mean score on the AMS index is much
lower than the mean for the core European subsample (8.16) and the RoW sample
(8.12). In this we might be seeing the affluence thesis trends at work in Europe.
The group of semi-peripheral countries seems to be dragging the European aver-
ages down on indicators of individual activation and willingness for material
sacrifice, despite their comparative affluence from a global perspective.
Whilst the RoW sample may be both too small and too diverse in the inequality-
adjusted national income (III) – not to mention developmental history and cultural
patterns – to provide context for the affluence thesis trends established in the
larger European sample, within the latter even this separation into two subgroups
suggests that money talks when it comes to individual degrowth concerns and
commitments, but with a possible additional proviso that ‘money’ is a relative
term depending on the historical and cultural surroundings to which populations
compare themselves.
Attitudes to environment–development trade-offs

On the indicators of degrowth-supportive attitudes the affluence thesis weakens
or reverses even in the group of 18 European countries (Domazet et al. 2014a).
With Russia included in that group in this text (elaborated above), the trend
remains unchanged. These are measures not of individuals’ personal values and
behaviour, but their comment on general societal development trends in the trade-
offs between environmental conservation and economic activity. A weak trend
following the affluence thesis expectation can still be seen to exist in support for
an Active Degrowth orientation (Kendall´s tau_b = 0.542, sig .002, Kendall´s
tau_b = 0.549, sig .001; see Figures 7.1 and 7.2), and semi-peripheral European
countries’ scores differ significantly from those of the core European subsam-
ple (Table 7.3). The correlations with national III are still much weaker than in
the indices mapping individual concern and commitment, but are positive. On
the Passive Degrowth orientation measures the trend is reversed, correlation is
weak and negative, and the semi-peripheral countries’ subsample exhibits as high
or higher means than is the case with the core European countries’ subsample
(Figures 7.3 and 7.4).
When measuring the proportion of national populations that disagrees with
the statement that their country needs economic growth in order to successfully
protect the environment (Active Degrowth: ISSP 2010 Q11a), the subsample of
semi-peripheral European countries records a mean proportion of 22.9%, whilst
the mean for the core countries’ subsample is 39.1%. The scatter of individual
country sample proportions against the measure of national affluence is given in
Figure 7.1. The population supportive of degrowth proactive policies within a
given country might also be expected to disagree with the statement that presently
we worry too much about the future of the environment and not enough about
current prices and jobs (Q10a). The mean proportion of population expressing this
55.0%
Environment: protect by economic growth
FI
50.0% CH
45.0%
40.0%
NO
(% disagree)
35.0%
AT
30.0% FR
ES DK
DE
25.0% BE
GB
20.0% SI SE
HR
15.0% LV CZ
LT
RU SK
10.0% BG
5.0%
0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850
Inequality-adjusted Income Index
Figure 7.1 III 2011, Q11a (Active Degrowth: Environment: protect by economic growth)
indicators’ scatter of values for individual European country samples
disagreement is 45.0% in the semi-peripheral subsample. In the core countries’

subsample the mean proportion is 53.4%, with the scatter of individual country
sample proportions given in Figure 7.2.
The Passive Degrowth orientation uses measures of awareness of negative envi-
ronmental impacts of growth and development, without explicit trade-off with the
current economic institutions and development model. It measures the proportion of
the national populations expressing agreement with: (1) the statement that economic
growth is always harmful to the environment; and (2) agreement with the state-
ment that economic progress in a given country will slow down if better care is not
taken of the environment. People agreeing with such statements might see them as
long-term general truths, but might not apply them to support immediate economic
policies. The reasoning might be summed up as: ‘Growth on the whole is harmful
and will eventually come up against environmental limits, but not right now’.
On the Passive Degrowth orientation greater national affluence in Europe
is, overall, inversely correlated to the proportion of the population supportive
of its measures. A similar trend is observed in the RoW sample of countries
too (Kendall´s tau_b = -0.437, sig .033; Kendall´s tau_b = -0.412, sig .043).
But these correlations are weaker and less significant than those established for
the affluence thesis trends on ACA and AMS indices, and analysed in Domazet
et al. (2014) and Dolenec et al. (2014a). The stronger measure of the Passive
Degrowth orientation, suggesting that progress in a given country will slow down
if the environment is not accorded better protection, again draws large propor-
tions of European populations to express support; the mean of core countries’
subsample stands at 60.9%, and that of the semi-peripheral countries’ subsample
60.0%
Worry too much about environment over jobs
55.0% DK
CH
50.0% NO
DE SE
(% disagree)
45.0% AT
40.0% BG
ES
FI
GB
35.0% CZ
LT BE
HR RU SK FR
30.0% LV SI
25.0%
0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850
Figure 7.2 III 2011, Q10a (Active Degrowth: Worry too much about environment over
jobs) indicators’ scatter of values for individual European country samples.
at 67.6% (Table 7.3). The range of national sample proportions is presented as

scatter against III national values in Figure 7.3.
Proportions of the degrowth-orientation supportive population on the meas-
ure of growth’s necessarily detrimental impact on the environment mean values
for the subsamples of both core and semi-peripheral countries stand at around
40% (Table 7.3). The less affluent European countries’ populations might be more
aware of environmentally harmful development within their living memory, but
also see more of a direct link between the environment and prospects for further
progress. Nonetheless, the affluence thesis trend breaks down on this measure
also, and as the scatter of values in Figure 7.4 indicates, affluence as measured
by the rising III value is not a predictor of a greater proportion of population
agreeing with the statement that economic growth is necessarily harmful to the
environment. Most importantly, overall the whole European group of countries
lags behind the RoW mean (Table 7.2) and maximal national scores, especially in
the developing economies of the global South.
We have also tested the correlations between national mean scores on the ACA
and AMS indices and individual indicators of Active and Passive Degrowth, to
examine the echoes of the affluence thesis established between national III scores
and ACA and AMS index mean scores (Table 7.4).
The weakness of correlation coefficients of ACA and AMS indexes with the
selected ISSP variables (Q10a, Q11a, Q11b and Q19c) shows that European
countries with a high average mean score of individual concern for the environ-
ment, or willingness to sacrifice in order to protect the environment, are not
65.0%
Economic progress will slow down without
BG
environmental protection (% agree)
55.0% ES SI
DE
DK
LT RU
45.0% CH
HR LV SE
SK
FI AT
35.0% CZ FR NO
GB
25.0% BE
15.0%
5.0%
0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850
Figure 7.3 III 2011, Q19c (Passive Degrowth: Economic progress will slow down without
environmental protection) indicators’ scatter of values for individual European
country samples.
45.0%
Economic growth: harms the environment
40.0% CH
BG FI
SI AT
35.0%
LT SK FR
30.0% RU
DE
(% agree)
CZ
LV ES
25.0%
DK
BE
20.0% HR
15.0% GB
SE NO
10.0%
5.0%
0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850
Figure 7.4 III 2011, Q11b (Passive Degrowth: Economic growth: harms the environment)
indicators’ scatter of values for individual European country samples.
necessarily those with a higher proportion of degrowth-oriented population.

Having in mind the already presented connection between the affluence of a
society and the ACA and AMS indices, as well as the weak correlation of these
indices with degrowth indicators, the limits of the affluence thesis are once again
accentuated. Not only are semi-peripheral Europeans more obstinate to lead on
Table 7.4 Correlations between national individual concern (ACA) and sacrifice (AMS)
indices and active and passive degrowth indicators (ISSP 2010: Q10a, Q11a,
Q11b, Q19c).
Kendall’s tau_b; Active Active Degrowth Passive Passive Degrowth

sig. Degrowth Q11a: Degrowth Q19c: Economic
Q10a: Worry Environment: Q11b: progress will slow
too much about protect by Economic down without
environment economic growth: harms environmental
over jobs growth the environment protection
ACA: individual 0.321 0.220 -0.130 -0.028

concern and 0.000 0.000 0.000 0.000
activation
index
AMS: individual 0.187 0.090 0.029 0.111
willingness 0.000 0.000 0.000 0.000
for material
sacrifice index
the commitment to individual sacrifice (cf. Ančić and Domazet 2015), but even
the people from wealthier nations may hope to stay within environmental lim-
its without substantial change to economic institutions and development trends
dominant in the second half of twentieth century (cf. Guiterrez Karp 1996;
Brajdić Vuković 2014).
Discussion and conclusions

Our analysis detects undeniable traces of affluence environmentalism and the con-
sequent degrowth trend compliant with the affluence thesis within the European
sample of ISSP participating countries (Diekman and Franzen 1999). Whilst the
equivalent trend is not detected on the individual commitment ACA and AMS
indices in the Rest of the World (RoW) sample of ISSP participating countries, the
dominance of the European subsample in the total ISSP Environment 2010–2011
field drives the affluence thesis trend described in the Introduction. The RoW
countries are, on the whole, slightly less prosperous and have a lower environ-
mental footprint, despite including such developed giants as Canada, the USA,
Japan and South Korea.
Within Europe, further comparison of core and semi-peripheral countries’ sub-
samples (Chase-Dunn and Hall 1997) indicates that the semi-peripheral countries’
affluence aspirations are shaping the overall affluence thesis trend. The semi-
peripheral countries are of below-average affluence, but also have a comparatively
lower environmental impact (cf. Table 7.3) and lower mean performance on ACA
and AMS indices. These indices mark the self-reported individual commitment to
acting knowledgeably on behalf of environmental protection, and a readiness to
make individual material sacrifices. These can be expected to be parts of demo-
cratic degrowth participation which enjoy comparatively lower support among
the populations of less prosperous European countries. It appears that the people
of the semi-peripheral European countries are still wedded to the promise of eco-
nomic growth ideology (Barry 2012), which would deliver them greater affluence
akin to that of the core European populations, even if that means paying a higher
environmental price and raising their global environmental impact closer to the
overall European mean.
This is in line with the findings of previous analyses (Dolenec, Domazet and
Ančić 2014), indicating that relative poverty decreases respective populations’
concern for the environment and their readiness to change behaviour or incur
material sacrifice to this end. In fact, overall income inequality within population
groups and subgroups has already been shown to be a major obstacle to reori-
entation towards more sustainable social metabolisms, both within and between
nation states. As the less prosperous and generally ‘environmentally lighter’
European states are also characterized by greater internal income inequali-
ties (Haanpää 2006; Eurostat 2010; UNDP 2011; Dolenec, Domazet and Ančić
2014), their overall populations’ degrowth orientation is severely diminished. On
one hand, this may be an example of infrastructural insufficiencies; in this case,
socio-economic rather than material, as in the case of differences between indi-

vidual ‘willingness to pay’ and ‘ability to pay’. Even on normative grounds, it can
be considered unfair to expect those with fewer resources in the contemporary
globalized economy to lead the contribution to the common endeavour, or as Shue
says ‘if the wealthy have no general obligation to help the poor, the poor certainly
have no general obligation to help the wealthy’ (Shue, 1999, p. 543).
However, previous research has indicated that despite the individual sac-
rifice aspect of a degrowth orientation being comparatively dampened in the

semi-peripheral European societies, their environmental concern and support for
collective degrowth strategies are not similarly impeded. In fact, less prosperous
European populations exhibit heightened awareness of the risks posed by envi-
ronmental degradation, and have stronger preferences for the redistribution of
existing accumulated material wealth (Bieri and Stoilova 2014; Brajdić Vuković
2014; Dolenec, Domazet and Ančić 2014). In this analysis we tested this further
with explicit measures of proportions of national populations supportive of Active
and Passive Degrowth measures, extracted from the ISSP questionnaire. It is on
Active Degrowth measures, pitching a contemporary trade-off between economic
growth and environmental protection in favour of the latter, that semi-peripheral
European countries lag behind the core European countries.
An affluence thesis trend can still be detected in correlations between indi-
vidual national affluence and Active Degrowth measures (cf. Figures 7.1 and 7.2),
though it is weaker and less significant than is the case with individual concern
and commitment indices. Scatter figures in particular indicate that the core–semi-
periphery boundary is extremely permeable on Active Degrowth measures, as the
proportions of supportive populations in many of the core countries are identical
to those placed into the semi-periphery group based on the measure of national
affluence. In both cases of Active Degrowth indicators almost half of the countries
included in the core subsample exhibit proportions of Active Degrowth support
equivalent to the more coherent semi-peripheral subsample. It would be interest-
ing to investigate further correlations with contextual variables such as educational
attainment, national informational campaigns for environmental protection, and
dominant social paradigms concerning techno-optimistic transcendence of envi-
ronmental limits of development (cf. Brajdić Vuković 2014).
However, when general awareness of environmental limits is tested, semi-
peripheral countries actually score as high or outperform the core countries, and
the affluence thesis trend among European countries reverses (cf. Figures 7.3 and 7.4).
Thus, in terms of a proportion of the population acknowledging a permanent
trade-off between economic growth and environmental stability semi-peripheral
and core European countries stand neck to neck, and well below the RoW mean.
The RoW group includes countries with levels of affluence and overall human
development a whole category lower than any of the European countries. The
respective ecological footprints of some of them are also much closer to globally
sustainable levels than is the case with even the lowest footprint of a European
country in the sample. But more importantly, on the awareness of the pressure
of environmental limits against further economic progress, the semi-peripheral

countries (with a lower reliance on a techno-optimistic knowledge society; cf.
Brajdić Vuković 2014) exhibit higher proportions of people supporting Passive
Degrowth, which completely reverses the affluence thesis trend. This puts the
semi-peripheral European societies in a specific historical bind of aspiring to
attain greater affluence even at the cost of immediate environmental stability
(based on Active Degrowth), whilst nonetheless exhibiting heightened aware-
ness of the environmental limits of such a development model (based on Passive

Degrowth). It could also indicate their popular attachment to ‘ecological mod-
ernization’ strategies (Cifrić et al. 1998). Such strategies present a hang-up from
the developmental orientations hooked on the growth ideology in both core and
semi-peripheral European societies throughout the twentieth century, combined
with techno-optimism about the reversibility of environmental damage.
A logical follow-up to these indicative findings would be to test the rela-
tionship between contextual sociodemographic characteristics of degrowth
supporters across Europe, regardless of citizenship, and viable and positive
degrowth-oriented mid-term policies and long-term development strategies. It
would also be interesting to see whether there is an overlap in these characteristics
and support between European and global populations, especially those already
expressing a high proportion of support for some of our degrowth-orientation
measures. Although correlations do not prove causality, an exploration of drivers
of development strategies deviating from the socio-metabolically paradoxi-
cal affluence thesis is a much needed research task in the face of an inevitable
transition to a novel humanity-nature regime in this century (Fischer-Kowalski,
Krausmann and Pallua 2014).
To what extent is degrowth potential present in Europe, given the noted disso-
nance between the discursive expressions of environmental leadership and global
concern, and lagging behind the RoW subsample on our measures of individual
sacrifice and collective awareness of the clash between economic growth and
environmental limits? Whilst the European degrowth-compliant population is not
of negligent size and standing, it seems to be divided between more and less
affluent European nations by development aspirations. To strengthen the overall
degrowth potential Europe needs to break down the boundaries between core
and periphery in allowing values of redistribution, community cohesion and
metabolic localization rooted in social trust to diffuse throughout (Guiterez Karp
1996; Latouche 2009).
Notes
1 The same set of European countries as is presented in this text, excluding Russia. As
later analyses in this text show, Russia does not deviate from these trends.
2 Franzen and Meyer (2010) and Domazet et al. (2014a) use the term ‘prosperity
thesis’, but to align with other texts in this volume ‘affluence thesis’ will be used
throughout without affecting the general sense of the argument based on those and
related references.
3 Austria, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Finland, France,

Germany, Latvia, Lithuania, Norway, Russia, Slovak Republic, Slovenia, Spain,
Sweden, Switzerland, United Kingdom.
4 Argentina, Canada, Chile, Israel, Japan, Mexico, New Zealand, Philippines, South
Africa, South Korea, Taiwan, Turkey, United States.
5 In order of decreasing value of Inequality-adjusted Income Index – III (UNDP 2011):
Norway, Sweden, Denmark, Austria, Switzerland, Finland, Belgium. Germany, France,
Slovenia, Czech Republic, United Kingdom.

6 In order of decreasing value of Inequality-adjusted Income Index – III (UNDP 2011):
Slovakia, Spain, Russia, Lithuania, Bulgaria, Latvia, Croatia.
7 Reliability analysis on these five items for all of the ISSP countries indicates an accept-
able internal consistency (Cronbach’s α is 0.68). Therefore, we have constructed an
additive index ranging from 5 to 25.
8 Reliability analysis on these three items for all of the ISSP countries indicates a good
internal consistency (Cronbach’s α is 0.84). Therefore, we have constructed an additive
index ranging from 3 to 15.
9 In Domazet et al. (2014a) this was used as a Non-growth indicator 1.
10 In Domazet et al. (2014a) this was used as a Non-growth indicator 2.
11 DSP: dominant social paradigm; KSI: knowledge society indicator (Brajdić Vuković,
2014).
References
Ančić, B., and Domazet, M. 2015. Potential for degrowth: attitudes and behaviours across
18 European countries. Teorija in Praksa, 52(3), 456–475.
Arrighi, G. and Drangel, J. 1986. Stratification of the world-economy: an explanation of
the semiperipheral zone. Review 10: 9–74.
Barnett, J., Matthew, R. A. and O’Brian, K. L. 2010. Introduction. In R. A. Matthew,
J. Barnett, B. McDonald and K. L. O’Brien (eds.) Global Environmental Change and
Human Security, Cambridge, MA: MIT Press, pp. 3–32.
Barry, J. 2012. The Politics of Actually Existing Unsustainability. Oxford: Oxford
University Press.
Bieri, F. and Stoilova, R. 2014. Environmental concern in Bulgaria: The role of social
trust. In M. Domazet & D. Marinović Jerolimov (eds.), Sustainability Perspectives from
the European Semi-periphery. Zagreb: Institut za društvena istraživanja u Zagrebu;
Heinrich Böll Stiftung-Hrvatska, pp. 225–242.
Brajdić Vuković, M. 2014. The sustainability potential of the knowledge society: empiri-
cal study. In M. Domazet & D. Marinović Jerolimov (eds.), Sustainability Perspectives
from the European Semi-periphery. Zagreb: Institut za društvena istraživanja u Zagrebu;
Heinrich Böll Stiftung-Hrvatska, pp. 195–222.
Chambers, N., Simmons, C. and Wackernagel, M. 2000. Sharing Nature’s Interest:
Ecological Footprints as an Indicator of Sustainability. London: Earthscan.
Chase-Dunn, C. and Hall, T. D. 1997. Rise and demise: comparing world-systems. Boulder,
CO: Westview Press.
Chatteron, P. and Pickerell, J. 2010. Everyday activism and transitions towards post-
capitalist worlds. Transactions of the Institute of British Geographers 35: 475–490.
Cifrić, I., Čaldarović, O., Kalanj, R. and Kufrin, K. 1998. Društveni razvoj i ekološka mod-
ernizacija - prilozi sociologiji tranzicije. Zagreb: Hrvatsko sociološko društvo.
Corral Verdugo, V. 2012. The positive psychology of sustainability. Environment,

Development and Sustainability, 14: 651–666.
Creech, H., Paas, L., Huppé, G., Voora, V., Hybsier, C. and Marquard, H. 2014. Small-scale
social-environmental enterprises in the green economy: supporting grassroots innova-
tion, Development in Practice 24:3, 366–378, DOI: 10.1080/09614524.2014.89956.
D’Alisa, G., Demaria, F., and Kallis, G. 2015. Degrowth: A vocabulary for a new era.
Abingdon, UK: Routledge.
Demaria, F., Schneider, F., Sekulova F. and Martínez-Alier, J. 2013. What is Degrowth?
From an Activist Slogan to a Social Movement. Environmental Values 22: 191–215.
De Young, R. 1993. Changing behavior and making it stick: the conceptualization and
management of conservation behavior. Environment and Behavior, 25: 485–505.
Diekmann, A., and Franzen, A. 1999. The wealth of nations and environmental concern.
Environment and Behaviour, 31: 540–549.
Dietz, R., and O’Neill, D. 2013. Enough is enough: Building a sustainable economy in a
world of finite resources. San Francisco: Berrett-Koehler.
Dolenec, D., Domazet, M., and Ančić, B. 2014. Why power is not a peripheral concern:
exploring the relationship between inequality and sustainability. In M. Domazet &
D. Marinović Jerolimov (eds.), Sustainability Perspectives from the European
Semi-periphery. Zagreb: Institut za društvena istraživanja u Zagrebu; Heinrich Böll
Stiftung-Hrvatska, pp. 173–194.
Domazet, M., Ančić, B., and Brajdić Vuković, M. 2014a. Prosperity and environmental sac-
rifice in Europe: Importance of income for sustainability-orientation. In M. Domazet &
D. Marinović Jerolimov (eds.), Sustainability Perspectives from the European
Semi-periphery. Zagreb: Institut za društvena istraživanja u Zagrebu; Heinrich Bőll
Stiftung-Hrvatska, pp. 145–172.
Domazet, M., Dolenec, D., Cvijanović, V., Tomašević, T., Walton, J. F., Doolan, K.,
and Žitko, M. 2014b. We need to change: ideas of growth and development in a
time of crisis of fossil energy and capitalism. Etnološka Tribina, 44 (37), 37–71.
doi:10.15378/1848-9540.2014.37.01.
Domazet, M., and Marinović Jerolimov, D. 2014. Sustainability on the semi-periphery:
an impossible topic in a non-existent place? In M. Domazet & D. Marinović
Jerolimov (Eds.), Sustainability Perspectives from the European Semi-periphery.
Zagreb: Institut za društvena istraživanja u Zagrebu; Heinrich Bőll Stiftung-Hrvatska,
pp. 19–49.
Eurostat. (2010). Income Poverty and Material Deprivation in European Countries.
Eurostat Methodologies and Working Papers. Luxembourg.
Fischer-Kowalski, M., Krausmann, F. and Pallua, I. 2014. A sociometabolic reading of the
Anthropocene: modes of subsistence, population size and human impact on Earth. The
Anthropocene Review, 1(1): 8–33.
Fournier, V. 2008. Escaping from the economy: the politics of degrowth. International
Journal of Sociology and Social Policy 28: 528–545.
Franzen, A., and Meyer, R. 2010. Environmental attitudes in cross-national perspective:
a multilevel analysis of the ISSP 1993 and 2000. European Sociological Review, 26:
219–234.
Franzen, A. and Vogl, D. 2013. Two decades of measuring environmental attitudes: a com-
parative analysis of 33 countries. Global Environmental Change, 23(5): 1001–1008.
doi:10.1016/j.gloenvcha.2013.03.009.
Global Footprint Network. 2015. Available at: www.footprintnetwork.org
Griggs, D., Stafford-Smith, M., Gaffney, O., Rockström, J., Öhman, M. C., Shyamsundar,
P. W., Steffen, Glaser, G., Kanie, N. and Noble, I. 2013. Sustainable development goals
for people and planet. Nature, 495: 305–307.
Gutierrez Karp, D. 1996. Values and their effect on pro-environmental behavior.
Environment and Behavior, 28(1): 111–133.
Haanpää, L. (2006). Environmental concern worldwide: an examination of two-way the-
sis of environmentalism and national variation. In T. A. Wilska & L. Haanpää (Eds.),
Lifestyles and Social Change. Essays in Economic Sociology. Tampere: Esa Print,
pp. 179–204.
Hedlund-de Witt, A. 2012. Exploring worldviews and their relationships to sustain-
able lifestyles: towards a new conceptual and methodological approach. Ecological
Economics, 84: 74–83.
Hulme, M. 2009. Why we disagree about Climate Change: Understanding Controversy,
Inaction and Opportunity. Cambridge: Cambridge University Press.
ISSP Research Group 2012, International social survey programme: Environment III – ISSP
2010. Gesis Data Archive, Cologne. ZA5500 Data file version 2.0.0, doi:10.4232/1.11418
Jackson, T. 2009. Prosperity without Growth: Economics for a Finite Planet. London:
Earthscan.
Kallis, G. 2011. In defense of degrowth. Ecological Economics, 70: 873–880.
Kallis, G.; Kerschner, C. and Martínez-Alier, J. 2012. The economics of degrowth.
Ecological Economics 84: 172–180.
Kasser, T. 2002. The High Price of Materialism. Cambridge, MA: MIT Press.
Krausmann, F., Gingrich, S., Eisenmenger, N., Erb, K.-H., Haberl, H., and Fischer-
Kowalski, M., 2009. Growth in global materials use. GDP and population during the
20th century. Ecological Economics, 68 (10): 2696–2705.
Latouche, S. (2009). Farewell to Growth. Cambridge: Polity Press.
Martínez-Alier, J., Pascual, U., Vivien, F., and Zaccai, E. 2010. Sustainable de-growth:
Mapping the context, criticisms and future prospects of an emergent paradigm.
Ecological Economics 69: 1741–1747.
Meyer, J. M. 2010. A democratic politics of sacrifice? In M. Maniates, and J. M. Meyer
(eds.) The Environmental Politics of Sacrifice, Cambridge, MA: MIT Press, pp. 13–32.
Meyer, R. and Liebe, U. 2010. Are the affluent prepared to pay for the planet? Explaining
willingness to pay for public and quasi-private environmental goods in Switzerland.
Population and Environment 32(1): 42–65.
Meyer, J. M. and Maniates, M. 2010. Must we sacrifice? Confronting the politics of sacrifice
in an ecologically full world. In M. Maniates & J. M. Meyer (Eds.), The Environmental
Politics of Sacrifice. Cambridge MA: MIT Press, pp. 1–12.
Ostrom, E. 2010. Polycentric systems for coping with collective action and global environ-
mental change. Global Environmental Change, 20: 550–557.
Polanyi, K. 1968. Primitive, Archaic, and Modern Economies: Essays of Karl Polanyi.
Garden City, NY: Anchor Books.
Pomeranz, K. 2009. Introduction: world history and environmental history. In E. I. Burke &
K. Pomeranz (Eds.), Environment and World History. London: University of California
Press, pp. 3–32.
Rist, G. 2003. The History of Development: From Western Origins to Global Faith.
London: Zed Books.
Romano, O. 2012. How to rebuild democracy, re-thinking degrowth. Futures 44: 582–589.
Rundgren, G. 2013. Garden Earth. From Hunter and Gatherer to Global Capitalism and
Thereafter. Uppsala: Garden Earth.
Schaffartzik, A., Mayer, A., Gingrich, S., Eisenmenger, N., Loy, C., and Krausmann, F.
2014. The global metabolic transition: regional patterns and trends of global mate-
rial flows, 1950–2010. Global Environmental Change, 26, 87–97. doi:10.1016/j.
gloenvcha.2014.03.013.
Schneider, F. 2008. Macroscopic rebound effects as argument for economic degrowth.
Proceedings of degrowth conference. Paris 18–19 April 2008. Barcelona: Research

and Degrowth.
Schneider, F. 2010. Degrowth of production and consumption capacities for social justice,
well-being and ecological sustainability. Proceedings of the second conference on eco-
nomic degrowth for ecological sustainability and social equity. Barcelona: University
of Barcelona.
Schultz, P. W. 2001. The structure of environmental concern: concern for self, other peo-
ple, and the biosphere. Journal of Environmental Psychology, 21: 327–339.
Shue, H. 1999. Global environment and international inequality. International Affairs 75:
531–545.
Steinberger, J. K., Krausmann, F., Getzner, M., Schandl, H., and West, J. 2013. Develop
ment and Dematerialization: An International Study. PLoS ‘ONE’, 8(10), e70385.
doi:10.1371/journal.pone.0070385.
Sugihara, K. 2003. The East Asian path of economic development. In G. Arrighi,
T. Hamashita and M. Selden (eds.) The Resurgence of East Asia: 500, 150 and 50 year
perspectives. London: Routledge-Curzon, pp. 78–123.
Tienhaara, K. 2009. A Tale of Two Crises. What the Global Financial Crisis Means for the
Global Environmental Crisis. Amsterdam: Global Governance Working Paper.
UNDP. 2011. Human Development Report 2011: Sustainability and Equity: A Better
Future for All. New York: UNDP.
UNEP. 2011a. Towards a Green Economy: Pathways to Sustainable Development and
Poverty Eradication. www.unep.org/greeneconomy (accessed 8 March 2016).
UNEP. 2011b. Green Economy: Why a Green Economy Matters for the Least Developed
Countries. www.unep.org/greeneconomy (accessed 8 March, 2016).
Wallerstein, I. (1979). The Capitalist World-Economy. Cambridge: Cambridge University
Press.
Whitmarsh, L., and O’Neill, S. (2010). Green identity, green living? The role of pro-
environmental self-identity in determining consistency across diverse pro-environmental
behaviours. Journal of Environmental Psychology, 30(3): 305–314. http://doi.org/10.
1016/j.jenvp.2010.01.003.
Chapter 8
Environmental concern in
cross-national comparison
Methodological threats and
measurement equivalence
Jochen Mayerl
Introduction
During the last few decades there has been an increasing and rapidly growing
interest in cross-national differences in environment-related perceptions of peo-
ple around the world. Do perceived environmental problems, values, attitudes
and perceptions of environmental concern, in general, differ across nations,
countries, cultures or contexts?1 Investigational interests in this field differ in
their focus on country-specific causes of environmental concern, correlates of
environmental concern, effects of environmental concern and the extent of quan-
tifiable country-specific levels of environmental concern; i.e. mean values and
ranking of countries.
Another strand of research deals with methodological threats to the validity
of cross-national studies, especially when conducting cross-national surveys.
Results of this research raise doubt over whether or not empirical results on every
kind of latent construct, thus including environmental concern, are even compa-
rable between countries in a valid way. In the context of the Confirmatory Factor
Analysis (CFA) approach, multiple group analyses of different nations, cultures
and social contexts are conducted to investigate measurement equivalence of
latent constructs. Do we measure the same latent construct when specifying a spe-
cific measurement model of environmental concern with the same indicators and
factorial structure all across the world? In a non-technical sense: is the meaning of
these constructs the same for people coming from different countries with more
or fewer differences in cultural background?
When linking these two research frameworks, environmental sociology must
ask whether the shared theoretical and empirical knowledge on cross-national dif-
ferences, related to environmental concern, are biased, misleading or even wrong
when it turns out that the measurement scales of environmental concern do not
share the same meaning for people across the world.
The next section discusses some empirical results in the literature on cross-
national differences related to environmental concern. The following section
then deals with methodological problems of cross-national surveys with a special
focus on measurement equivalence of environmental concern. In the subsequent
Cross-national comparison 183
part, empirical testing is conducted using Multiple Group Confirmatory Factor

Analysis (MGCFA), with ISSP 2010 data including 32 countries. The analyses
focus on the question of whether measurement equivalence may be reached across
European countries, as well as in a worldwide perspective. This chapter does not
aim to propose a ‘new’ measurement model of environmental concern, but rather
to demonstrate the empirical problems that comparative researches must face and
deal with when comparing environmental issues across nations.
Environmental concern in Europe and worldwide

Environmental concern can be defined as ‘the degree to which people are aware of
problems regarding the environment and support efforts to solve them and/or indi-
cate a willingness to contribute personally to their solution’ (Dunlap and Jones,
2002, p. 485). Thus, it is a rather broad concept of generalized attitudes towards
the environment and environmental problems. From a cross-national comparative
perspective, it is of the utmost importance to describe and explain differences or
similarities in the extent of environmental concern worldwide.
According to the theoretical framework of the globalization hypothesis
(e.g. Brechin and Kempton, 1994; Dunlap and Mertig, 1995), environmen-
tal problems are global phenomena, thus environmental concern should be
high globally, independent of state wealth or dominant values. In contrast,
the affluence hypothesis assumes that national wealth is positively correlated
with environmental concern (e.g. Franzen, 2003). In this sense, environmen-
tal issues are luxury concerns to which citizens can give attention when basic
needs are satisfied. Analyses with ISSP concluded that national wealth is
positively connected to environmental concern (e.g. Diekmann and Franzen,
1999; Franzen and Meyer, 2010; Franzen and Vogl, 2013). However, analy-
ses of Dunlap and Mertig (1995) with the Health of the Plant Survey (HOP),
Fairbrother (2013) with World Value Survey Data, and Marquart-Pyatt (2012)
with ISSP data, show no evidence in favour of the affluence hypothesis at the
level of national wealth. However, at the level of individual socio-economic
status (e.g. income), most studies show a significant positive effect on envi-
ronmental attitudes (Fairbrother, 2013; Pampel, 2013). Another theoretical
framework argues that environmental concern is positively related to the raise
of post-materialist values in a country (e.g. Inglehart, 1995). Further, Inglehart
(1995) argues that post-materialism and national wealth interact in the sense
that national wealth acts as a moderator variable. Following this line, post-
materialism is strongly related to environmental concern in wealthy countries
only (see Inglehart, 1995; Mayerl and Best, 2014). In contrast to these find-
ings, Brechin (1999) did not find such an interaction effect. A relatively stable
result is that post-materialism is positively associated with environmental con-
cern (e.g. Fairbrother, 2013; Franzen and Meyer, 2010; Mayerl and Best, 2014;
Pampel, 2013). In addition, on an individual level, there is quite consistent
evidence that more highly educated as well as younger people, show higher
184 Jochen Mayerl
levels of environmental concern all over the world (Franzen and Meyer, 2004;
Pampel, 2013; Mayerl and Best, 2014).
All in all, these cross-national studies show that empirical results on the causes
of global environmental concern are rather inconsistent, depending on the year of
conducting the survey, the concepts and operationalizations, the level of analy-
sis (macro- vs. microlevel) and the statistical methods adopted. Thus, empirical
results vary strongly across studies depending on single- vs. multiple-indicator
approaches, content of the used indicators and characteristics of the specified

latent construct of environmentalism (e.g. global vs. local; specific vs. general-
ized; values vs. primitive environmental beliefs vs. environmental behaviour;
economic vs. social aspects, etc.). According to Best and Mayerl (2013), scales of
environmentalism can be ordered in a hierarchical causal chain model. Thus, post-
materialist values act as antecedents of general primitive beliefs concerning the
environment, which in turn lead to general environmental attitudes. Such general-
ized attitudes then determine specific environmental attitudes, which then lead to
specific environmental behaviour. Thus, different correlates and predictive power
may be found depending on which scale is used. These points bring attention to
the necessity of considering the broader context of environmental cognitions. In
this sense, comparability of environmental studies is questionable when studies
differ in their concepts and measurements of environmental concern.
Aside from the aforementioned, problems can evolve from a lack of measure-
ment equivalence of environmental concern. Even when asking the same questions
and attempting to operationalize the same construct, did these studies really meas-
ure the same phenomenon across the world; i.e. do the respondents share the same
meaning of the indicators and the construct, independent of their national and
cultural background? We focus on this issue in the following sections.
Another popular way of analysing environmental concern through a cross-
national perspective is to generate rankings of nations according to means or
percentage values (e.g. Brechin and Kempton, 1994; Diekmann and Franzen,
1999; Dunlap, 1994; Franzen, 2003; Inglehart, 1995; Schultz, 2001). These
rankings show a high variation, among others, owing to the selection of coun-
tries under investigation, as well as due to different applications of concepts and
measurements of environmental concern (see ‘Overview of threats to validity of
cross-national surveys’). In tendency, one could cautiously conclude that mean
environmental concern is higher in Western countries (from the north) in contrast
to Eastern countries (from the south). On the other hand, Guha and Martinez-
Alier (1997) distinguished between problem-driven environmentalism of the poor
in southern countries and value-driven environmentalism of the rich in northern
countries. Thus, high variations in the extent of environmental concern may be
found worldwide, but for different reasons.
Then again, such rankings may be seriously biased owing to the lack of meas-
urement equivalence. As we will see in the following sections, cross-national
comparison of the means of constructs is even more problematic, and ridden with
even more prerequisites than the comparison of relationships across countries.
Methodological problems of cross-national

comparisons with surveys
Overview of threats to validity of cross-national surveys
Empirical results on environmental concerns in cross-national studies are undoubt-
edly manifold, and environmental sociology is advancing at an ever-increasing rate.
Cross-national survey programmes like the World Value Survey, the International
Social Survey Program, the European Value Study, the European Social Survey,
Eurobarometer etc. offer a great data basis to evolve theoretical, substantive, and
methodological knowledge about causes, effects, extent, dimensionality and meas-
urements of environmental concern in a cross-national comparative perspective.
On the other hand, comparative cross-national survey studies come with a wide
range of methodological challenges and problems. In recent years, the research
on comparative survey research joined forces in the research programme ‘3MC’
(Multicultural, Multinational, and Multiregional Contexts; Harkness et al.,
2010). The following brief description summarizes some of the major methodo-
logical problems that cross-national surveys face. These differences stem from
different sources which are classified into four problem areas: survey design;
sampling issues; country-specific characteristics; and differences in understand-
ing the meaning of concepts (see Bachleitner et al., 2014; Braun, 2014; Harkness
et al., 2010, and Nasif et al., 1991 for a discussion of these and many other issues
in detail).
Survey design
When conducting a cross-national survey with dozens of participating countries
all over the world, it is often difficult to implement the same survey mode in
all countries. Reasons for this may be the coverage of specific communication
media (e.g. Internet connection) or the financial costs of computer-assisted per-
sonal interviews. Some countries implement mixed-mode surveys, while others
conduct mono-mode surveys. Moreover, field time is often not possible to syn-
chronize, thus differences between nations may be biased by different time points
of conducting the survey. In addition, the workload of interviewers may be very
different across countries, which is connected to financial issues, as well as the
availability of skilled interviewers. These differences quickly lead to measure-
ment-error issues. A very serious threat to validity of comparative surveys is
the problem of fabricated data. Blasius and Thiessen (2012, 2015), for instance,
report on fabricated data in international surveys like the World Value Survey
2005–2008 and the Programme for International Student Assessment (PISA)
2009. This problem arises mainly as a result of the fact that in some international
survey programmes, countries have to reach specific levels of minimum response
rates which originally were meant to ensure high-quality data in all countries. The
lack of careful documentation in some countries, as well as high variations in data
quality management, are connected to these problems. Even the adequate design
186 Jochen Mayerl
of response scales may be problematic; e.g. some languages are written from left
to right and others from right to left, or depending on cultural values and norms,
bipolar scales may be inappropriate and unipolar scales are preferred (e.g. when
disagreement is socially undesirable).
Sampling Issues
A big issue is that random sampling cannot be done in the same way in all
countries. Some countries use data from resident registration offices, others use
random walk techniques, random digit dialling, or randomized last-digit tech-
niques via telephone. This variation also means that some countries draw random
samples on an individual level and some on a household level. In the case of the
latter technique, a second random procedure such as the ‘next-birthday method’
must be implemented, but it is not guaranteed that these methods are imple-
mented in all countries. Even worse, some countries do not do random sampling,
but instead use quota sampling or related non-random samples. Countries also
differ in their levels of effort to reduce non-responses (follow-up attempts,
refusal conversion, variation in the dates of contacts), and the number of con-
tacts in case a target person is left out of the sample. Good sample-management
systems (see above) use paradata to enhance these survey sampling methods
(e.g. Olson, 2013). There are also strong differences in the need to make the
effort since countries vary strongly in the overall willingness from residents to
participate in surveys (see Stoop et al., 2010, for the case of the European Social
Survey). Thus, depending on social norms and values within cultural contexts,
there is a high variation in the percentage of unit non-responses across coun-
tries. Last but not least, cross-national surveys should be aware of a selection
bias since countries which take part in an international study are not selected by
random. Nevertheless, inferences are often made towards a population such as
continents, cultural entities, or even ‘the world’, which could be biased owing to
this selection bias. Overall, these issues lead to serious problems with regard to
the representativeness of the data.
Country-specific characteristics
Countries differ in social norms and values, thus leading to country-specific lev-
els of non-response, levels of acquiescence (Franzen and Vogl, 2012), and the
extent of individual tendency towards socially desirable responses, as well as
the extent and direction of trait desirability, meaning that some issues might be
highly sensitive in one country but not in another (e.g. topics like sexuality, alco-
hol etc.). In addition, cross-national surveys are faced with country-specific levels
of literacy and education. This sometimes excludes modes of surveys (e.g. postal
and Internet surveys), but also leads to different education-specific response
styles. Thus answers may highly vary owing to a lack of knowledge regarding
the subject in question.
Understanding and the subjective meaning of constructs

This type of threat to cross-national surveys again is connected with the country-
specific norms, values and cognitive associations people have in mind when
responding to survey questions. The point is that people may understand the
meaning of a question in different ways depending on their cultural background.
Translation techniques of forward-and-backward translation are not sufficient to
solve these problems, because these translations adhere narrowly to literal transla-
tions of exact wording, and often fail to take into account the semantics of a question
for other languages and cultural backgrounds. This raises the issue of measurement
equivalence: are we measuring the same thing when we are adopting the same sets
of indicators to measure a specific latent construct? Thus, is the meaning of the
indicators and the constructs the same across countries? To answer these ques-
tions, Multiple Group Confirmatory Factor Analyses (MGCFA) are conducted,
which will be the focus of the subsequent sections. In addition more comparative
qualitative work is needed to get a deeper understanding of national differences of
subjective meaning of terms and constructs used in cross-national surveys.
Another even broader distinction of types of functional equivalence in com-
parative research was proposed by Bachleitner et al. (2014), who categorize
equivalence issues into problems of equivalence of basic assumptions (analytical
and theoretical level), content and construct equivalence, method equivalence,
interpretation equivalence and application equivalence. Therefore, how research-
ers deal with results in terms of interpretation and application, as well as which
underlying theoretical understanding of a model is given, should also be consid-
ered. The latter was discussed previously in the section ‘Environmental concern
in Europe and worldwide’, as environmental concern is conceptualized and
operationalized in such different ways in the literature that results of empirical
comparisons may depend heavily on these theoretical assumptions.
Obviously, it is highly risky to assume that survey data from different countries
can be compared validly and reliably. Comparative research tries to find evidence
for substantial differences between countries, but needs to eliminate or control for
the stated methodological threats to valid comparisons. In short, are we compar-
ing incomparable outcomes between countries and come to wrong conclusions
about national differences? Validity of research results greatly depends on how
we deal with these methodological challenges.
Measurement equivalence of environmental concern

in cross-national surveys
In this chapter, we will subsequently focus on the problem of measurement
equivalence (aka measurement invariance) of environmental concern across
Europe and worldwide. Measurement equivalence asks whether the measure-
ment models operationalized in surveys are the same across different countries.
Thus, in cross-national MGCFA it is all about the empirical test of whether the
188 Jochen Mayerl
measurement model can be treated to be equal across groups (here: nations) or

not. These tests are called ‘tests of measurement equivalence’ and ‘tests of facto-
rial invariance’, respectively (e.g. Kline, 2011; Davidov et al., 2014; and Urban
and Mayerl, 2014).
Measurement equivalence includes the structure of a measurement scale
(so-called ‘configural invariance’); the weights resp. factor loadings of the
indicators within a construct (so-called ‘metric invariance’); and the means of
the indicators (so-called ‘scalar invariance’). Table 8.1 summarizes different
Table 8.1 Concepts of factorial invariance (FIV) in cross-national studies.
Factorial Term of Factorial Short description

Invariance (FIV) Invariance Assumption
FIV1 configural factorial no equality constraints but the same

invariance measurement model in all groups
Comparative perspective: Do respondents from
different countries conceptualize the latent
constructs in the same way?
FIV2 metric (metric- (partial) equal factor loadings
(precondition: partial) factorial Comparative perspective: Are the constructs
FIV1) invariance consistent in their weighting (‘mix’) of the
indicators to ensure the same meaning of
these constructs?
FIV3 metric (metric- (partial) equal factor loadings and equal
(precondition: partial) factorial measurement errors
FIV1 and FIV2) invariance plus Comparative perspective: Are the random
measurement errors equal across countries?
FIV4 (partial) scalar equal factor loadings and (partial) equal
(precondition: invariance intercepts
FIV1 and FIV2) (mean Comparative perspective: is the mean structure
structure) of the indicators identical to allow latent
mean comparison between countries?
FIV5 latent mean equal factor loadings, intercepts and latent
(precondition: invariance means
FIV1, FIV2 and (mean structure) Comparative perspective: are the latent means
FIV4) in different countries equal or not?
FIV6 factor variance factor variances are treated to be equal
(precondition: invariance across groups
FIV1 and FIV2) Comparative perspective: are the
measurement models equal even at the level
of latent variances?
FIV7 strict invariance measurement equivalence: all possible
(precondition: restrictions
all FIV above) Comparative perspective: are the
measurement models completely invariant
across countries?
Bold type shows minimum factorial invariance assumptions needed for cross-national comparison of latent
means
types of measurement equivalence which can be tested via Multiple Group

Confirmatory Factor Analysis (MGCFA). It is important to note that these
measurement invariance assumptions must be treated in cumulative order.
Configural factorial invariance (FIV1) simply means that the same indicators
are used to operationalize the latent construct in all countries. If this does not bear
empirical testing, respondents of different countries conceptualize the latent con-
structs in different ways, making all comparisons of results impossible.
The assumption of metric factorial invariance (FIV2) assumes equal factor

loadings; i.e. equal weights of the indicators across all countries. If this is not the
case, comparisons of correlates or causal effects would result in erroneous conclu-
sions about national differences.
For comparing means of latent constructs across groups – e.g. latent means
of environmental concern across countries – the mean structure of the indicators
must be the same as well. Since indicators are dependent variables (see Figure 8.1),
the mean information refers to intercepts. The equivalence of indicator intercepts
is called scalar invariance (FIV4). Scalar invariance indicates that cross-national
differences of means of observed variables are the result of differences in latent
means at a construct level.
The cumulative characteristic of these factorial invariance models imply
that metric invariance (FIV2) needs configural invariance (FIV1) as a precondi-
tion. Furthermore, scalar invariance (FIV4) needs configural (FIV1) and metric
invariance (FIV2) as necessary preconditions. To gain valid comparisons in cross-
national studies, configural (FIV1) and metric invariance (FIV2) are needed to
compare effects and co-variances across countries, and in addition, scalar invari-
ance (FIV4) is needed to compare latent means across countries.
In ‘classic’ comparative studies, it is simply assumed that the measurement
equivalence of measures of environmental concern (and related constructs) is
given. Recently, environmental research began to pay attention to empirical
tests of measurement equivalence via MGCFA. Urban and Ščasný (2012) stud-
ied the measurement equivalence of a six-item environmental concern scale in
nine OECD-countries and found partial metric invariance but no scalar invari-
ance. Schultz et al. (2005) analysed the measurement equivalence of different
scales of environmental concern in six countries and found metric invariance.
Milfont et al. (2010) confirmed configural equivalence of first- and higher-order
environmental concern in three countries. Milfont et al. (2006) tested a scale of
environmental problems for Europeans versus Asian New Zealanders and found
partial metric and partial scalar invariance. Boeve-de-Pauw et al. (2012) studied
differences of environmental concern between men and women and found partial
metric and scalar invariance. They also showed that comparing sum scores with
t-tests indicated significant mean differences, but when applying partial invari-
ant MGCFA, latent mean differences were no longer significant. Thus, classic
statistical approaches may result in methodological artefacts. Ippel et al. (2014)
studied the invariance of Inglehart’s post-materialism scale across ten European
countries and found that metric invariance is given but scalar invariance is not.
190 Jochen Mayerl
Marquardt-Pyatt (2012) reports that metric and scalar invariance are given in her
measures of environmental concern with ISSP 2000.
Overall, it seems that latent constructs related to environmental topics often
reach (partial) metric invariance but seldom scalar invariance. In addition, most
of these studies compared measures of environmental concern with a very limited
number of countries or groups. Therefore, the next section illustrates the test of
measurement invariance of an environmental scale across Europe and worldwide,
with a higher number of countries using ISSP 2010 data. Is it possible to state that
environmental concern can be measured and compared across a wide range of cul-
tures globally, or at least in Europe? Is it possible to compare aggregated means of
environmental concern across countries?
Empirical study: measurement equivalence of

environmental concern in ISSP 2010
Data and measurement
To study environmental concern in a cross-national comparison, data from the
International Social Survey Programme (ISSP) 2010 were used (ISSP Research
Group, 2012). ISSP 2010 collects data on attitudes and values of citizens from 32
countries worldwide. The overall sample size is 43,589 respondents. Participating
countries are (ordered alphabetically) Argentina, Austria, Belgium, Bulgaria,
Canada, Chile, Croatia, Czech Republic, Denmark, Finland, France, Germany,
Great Britain, Israel, Japan, Latvia, Lithuania, Mexico, New Zealand, Norway,
Philippines, Russia, South Korea, Slovak Republic, Slovenia, South Africa,
Spain, Sweden, Switzerland, Taiwan, Turkey, and the USA.
The questionnaire contains several questions on environmental concern
which can be conceptualized on the basis of classic tripartite attitude theory
(Rosenberg and Hovland, 1960). This model distinguishes three attitudinal
dimensions: affective, cognitive and conative. Indicators of the affective dimen-
sion refer to emotions such as fear, worry, anger etc. Cognitive indicators are
connected to individual knowledge and refer to subjective beliefs and rather
rational statements of individuals. Indicators of the conative dimension refer to
the behavioural level of intended or actual (self-reported) behaviour. A classic
scale of environmental concern based on this tripartite model was introduced by
Maloney and Ward (1973). Dunlap and Jones (2002) proposed another scale of
environmental concern, which derived from Maloney and Ward’s measurement
model.
To avoid measurement errors owing to word bias in questionnaires (negative
versus positive phrasing), which typically lead to two-dimensional constructs
reflecting these two wording dimensions (e.g. Spector et al., 1997; Dunlap et al.,
2000), only indicators of environmental concern using unilateral (i.e. the same)
wording are considered. Table 8.2 reports the seven indicators used in the present
empirical study. In addition, overall mean and standard deviations of these indica-
tors, aggregating all 32 countries, are reported.
Franzen and Meyer (2010) and Franzen and Vogl (2013) used a similar set of
nine indicators from ISSP 1993, 2000 and 2010 to operationalize environmental
concern as a sum score index (based on 9 indicators). Most indicators of the
present study are part of this scale (see Table 8.2). Reliability analysis of the
present scale results in a value of Cronbach’s Alpha = 0.72 (based on 7 indica-
tors). This finding is very similar to the reliability measures reported by Franzen
and Meyer (2010) and Franzen and Vogl (2013). Furthermore, when estimating
reliability separately per country, Cronbach’s Alpha is on average 0.71 (i.e. the
mean of reliabilities of all 32 countries), with the lowest Alpha value of 0.50 in
Taiwan and the highest Alpha value of 0.82 in Great Britain. Thus, the reliability
of the used scale of environmental concern is quite acceptable but varies notably
across countries.
Measurement equivalence: multiple group confirmatory

factor analysis (MGCFA)
When applying the measurement scale of environmental concern shown in
Table 8.2, a traditional statistical approach would be to compute an additive
index (i.e. sum score) out of these seven indicators for each individual, then
calculate a mean value for each country to get a cross-national ranking of
environmental concern. However, this approach is based on implicit, highly
Table 8.2 ISSP 2010 indicators of environmental concern used in this study.
Question Dimension Rating scale Mean SD
Q10a We worry too much affective 1 (Agree strongly)– 5 2.99 1.164

about future (Disagree strongly)
environment
Q10c People worry too affective 1 (Agree strongly)– 5 3.05 1.082
much about harming (Disagree strongly)
environment
Q13a Too difficult for conative 1 (Agree strongly)– 5 3.13 1.183
someone like me (Disagree strongly)
to do much about
environment
Q13c More important things conative 1 (Agree strongly)–5 3.05 1.098
to do than protect (Disagree strongly)
environment
Q13d No point unless conative 1 (Agree strongly)– 5 3.06 1.233
others do the same (Disagree strongly)
Q13e Many claims about cognitive 1 (Agree strongly)– 5 3.15 1.119
environment (Disagree strongly)
exaggerated
Q13f Hard to know whether cognitive 1 (Agree strongly)– 5 3.10 1.074
the way I live is (Disagree strongly)
helpful or harmful
192 Jochen Mayerl
unrealistic and not empirically tested assumptions. First, it assumes that all
indicators are weighted identically within the construct. This could be solved
by classic Principal Component Analysis (sometimes called ‘exploratory fac-
tor analysis’), using the estimated factor loadings as weights, which give the
strength of the relationship between the latent construct and the manifest indi-
cator. This, however, does not solve the subsequent problems. Second, it is
assumed that all indicators are measured with zero measurement error. This is
highly unrealistic if latent variables such as environmental concern are opera-

tionalized, especially when using survey data from 32 countries. Third, the
weights (i.e. factor loadings) of each indicator are supposed to be the same in
all countries. This is a core assumption needed for comparative cross-national
studies, assuming that the construct consists of the same ‘mixing’ of indicators
in all countries. If this requirement were not met, one would compare rela-
tionships and levels of environmental concern across countries, but would be
unsure if the constructs were comparable. Fourth, when comparing aggregated
construct means at a national level (i.e. latent means), the means of the indica-
tors must be the same throughout all investigated countries.
Overall, it is important to note that the ‘classic’ way of comparing the
means of aggregated sum scores assumes implicitly that all four of these
assumptions are empirically given. If this is not the case, results of mean
comparisons may be completely wrong! In short, this is the main reason why
Multiple Group Confirmatory Factor Analysis (MGCFA) will be used to deal
with these problems.2 MGCFA solves the first and second problems by esti-
mating factor loadings and measurement errors. It is important to note that
MGCFA does not solve the third and fourth problems, but rather gives insight
into whether these assumptions are met or not. From a comparative research
perspective, correlates and latent means of environmental concern at the latent
construct level are allowed to be compared between countries only if there is
empirical evidence that the measurement is uniform throughout all countries.
If not, it would be like comparing apples to pears.
Therefore, measurement equivalence is our first criterion in evaluating the
measurement model. It is crucial to compare different levels of factorial invar-
iance (FIV) which can be ordered in a cumulative way (see Table 8.1; Urban
and Mayerl, 2014, pp. 167ff). MGCFA is a powerful tool to test these factorial
invariance assumptions empirically. In practice, it is often very difficult to specify
a measurement model which holds the required equivalence assumptions, espe-
cially regarding the intercepts of the indicators (scalar invariance). But what if
the equivalence assumptions are empirically falsified? One practical solution is
to specify partial invariance, allowing some factor loadings or some intercepts to
be free in some groups, while all other loadings with respect to intercepts are still
the same in all groups (Byrne, Shavelson and Muthen, 1989). Adopting partial
invariance, one possible solution would be to set a cut-off rule, that at least more
than 50% of the indicators must be equal across groups (Urban and Mayerl, 2014,
p. 169). However, when specifying partial invariance, the comparability of results

between groups is questionable (Beuckelaer and Swinnen, 2011; Davidov et al.,
2014, p. 66); but, when the alternative would be to skip the comparative analysis,
it is a fairly accepted way to deal with equivalence issues (Urban and Mayerl,
2014, pp. 225f.).3
Another evaluation criterion within MGCFA is to consider model fit indices
(see Brown and Cudeck, 1992; Hu and Bentler, 1995, 1999; Kline, 2011; Urban
and Mayerl, 2014, for more details on fit indices). Goodness-of-fit indices
compare the data from the observed variance-covariance matrix and the model-
based expected variance-covariance matrix (including mean structure). In other
words, they test whether the model fits the data or not. To evaluate the models
of environmental concern across countries, it is a precondition that the model
fit is acceptable. Many goodness-of-fit measures were proposed, but the most
important are CFI (Comparative Fit Index), SRMR (Standardized Root Mean
Square Residual) and RMSEA (Root Mean Square Error of Approximation)
with its 90% confidence interval. Suffice it to say that CFI should be higher
than 0.95 (good fit), or at the very least higher than 0.92 (Hair et al. 2010), and
SRMR and RMSEA should be lower than 0.05 (very good fit) or lower than 0.08
(acceptable fit). In the case of RMSEA, the upper limit of the 90% confidence
interval should be lower than 0.10. But these are only rules of thumb, which of
course vary across the literature.
Another evaluation criterion is the convergent validity of the measurement
model. As a common criterion, factor loadings should be significant and, in terms
of standardized values, at least above 0.5. A composite measure of convergent
validity is the so-called AVE (Average Variance Extracted), which is the mean
of all squared factor loadings of all indicators of the construct. When stating that
standardized factor loadings should be higher than 0.5 − at least the mean of all
factor loadings − the AVE should be at least higher than 0.25.
When adopting these three evaluation criteria − measurement equivalence,
model fit and convergent validity − it is important to note that generally, in prac-
tice, a model is accepted even when equivalence is empirically wrong, as long
as the model fit is still acceptable. This means that we force measurements to be
equal across groups or nations, even when this equality is empirically false, as
long as the model acceptably fits the data.4 It is all about comparability! When
measurement equivalence is not given and the model fit is poor, this simply
leads to bad news: it is no longer possible to compare the results across groups.
When ignoring this, the results of the comparison would be highly susceptible
to complete error, would be misleading and strongly biased. Thus, it is all about
the question of whether or not we are able to compare results related to latent
variables across groups.
Figure 8.1 shows the measurement model of environmental concern which will
be tested empirically with ISSP 2010 data. As can be seen in Figure 8.1a, the basic
measurement model includes seven indicator variables without error correlations.
194 Jochen Mayerl
A modified second model (Figure 8.1b) introduces six error correlations to specify
systematic measurement errors and to gain a better model fit.5 These error cor-
relations were specified in all countries to ensure strict configural equivalence for
comparative cross-national analyses.6
A robust Maximum Likelihood estimator (MLM) was used throughout the sta-
tistical analyses with software Mplus 7.7
Measurement equivalence of environmental concern

in Europe and worldwide
Subsequently, we applied MGCFA to test the measurement model and its meas-
urement equivalence across all 20 European countries in the first step (models
‘E.x’), and across 32 countries around the world in the second step (models ‘W.x’).
Starting with analysis of European countries, Table 8.3 reports the model fit and
composite convergent validity (in terms of AVE) of our environmental concern
scale distinguished into four models. In model E.1 and model E.2, factor load-
ings are constrained to be equal across the countries, but everything else is free.
Env.
Conc.
Q10a Q10c Q13a Q13c Q13d Q13e Q13f
(a) E1 E2 E3 E4 E5 E6 E7
Env.
Conc.
Q10a Q10c Q13a Q13c Q13d Q13e Q13f
E1 E2 E3 E4 E5 E6 E7
(b)
Figure 8.1 Measurement model of environmental concern (mean structure not shown):

a) without error correlations; and b) with error correlations.
Model E.2 introduces the error correlations reported in Figure 8.1b, whereas
model E.1 does not include these error correlations (Figure 8.1a). While these
two models are not enough to compare latent means of environmental concern
across countries, they test our basic measurement model, asking whether metric-
measurement equivalence is given in the European subsample.
In terms of model fit, as can be seen in Table 8.3, the fit is poor in model
E.1 (CFI<0.95 (even <0.90); RMSEA>0.08; SRMR>0.05). In contrast, the
fit is good − or at least acceptable − in model E.2 (CFI>0.95; RMSEA<0.08;

SRMR<0.08). Thus, when accepting error correlations as a result of other related
sources shared by specific indicators, it is possible to state that the measurement
model can be treated as the same in terms of the internal ‘mixture’ of indica-
tors across Europe. This is in line with other empirical studies on measurement
equivalence of environmental scales (see ‘Environmental concern in Europe and
worldwide’), and showing that metric invariance holds up in empirical testing,
thus allowing the comparison of causal relationships of environmental concern
to other constructs, predictors and outcomes.
As model E.2 was acceptable, model E.3 additionally introduces intercepts as
equal across countries, given that scalar invariance is a necessary condition to
compare latent means. By this, it is guaranteed that mean differences stem from
corresponding latent construct levels (Davidov, 2009; Urban and Mayerl, 2014).
If model E.3 (scalar invariance) works, this would be the best model to com-
pare latent means of environmental concern across Europe. However, looking at
Table 8.3, it is obvious that scalar invariance does not meet the requirement of
empirical testing, as the model fit is unacceptable poor (CFI<0.95 (even <0.80);
RMSEA>0.08; SRMR >0.08). This shows that the latent means of environ-
mental concern are − from a very strict point of view − not comparable across
countries in Europe!
Model E.4 relaxes the strict comparability criterion of full scalar invariance
by not including the intercept invariance constraints of three out of seven indi-
cators in each country. This is done by relaxing the three intercepts which are
most different with respect to the mean of all other countries. At least three
out of seven unconstrained intercepts follow the line that more than 50% of
the indicators should be equal when estimating partial invariance. As reported
in Table 8.3, the fit of model E.4 may be acceptable with a marginal fit in
terms of CFI (CFI>0.92) and an acceptable fit in terms of the other fit indices
(RMSEA<0.08; SRMR<0.08). Thus, it is possible to compare the latent means
of environmental concern across Europe with model E.4, but it must be borne
in mind that the comparability is reduced and results may be somewhat biased.
In terms of convergent validity, the mean AVE (i.e. the mean AVE across all
European countries) holds the stated minimum criterion of a mean AVE>0.25
in all four models, meaning that standardized factor loadings are, on average,
higher than 0.5. Nevertheless, AVE is lower than 0.25 in several countries
(see Table 8.3): in our final model E.4, in seven out of twenty countries, con-
vergent validity is low.
196 Jochen Mayerl
Table 8.3 Evaluation of measurement model across Europe.
Model E.1 Model E.2 Model E.3 Model E.4

Metric Metric Metric and Metric and
invariance invariance scalar invariance partial scalar
without error with error with error invariance with
correlations correlations correlations error correlations
Model RMSEA 0.088 0.060 0.119 0.073

Fita (90% C.I.) (0.086–0.091) (0.057–0.063) (0.116–0.121) (0.070–0.076)
CFI 0.862 0.956 0.754 0.921
SRMR 0.073 0.052 0.100 0.066
Chi2 3,856.800 1,379.022 6,563.774 2,321.397
df; p 394; 0.000 274; 0.000 388; 0.000 328; 0.000
AVE: Mean Mean: 0.300 Mean: 0.273 Mean: 0.264 Mean: 0.271
AVE across Hi: GBritain: Hi: GBritain: Hi: GBritain: Hi: GBritain:
countries; 0.403 0.367 0.360 0.367
highest–lowest Lo: Turkey: Lo: Spain: Lo: Spain: Lo: Turkey:
value; 0.198 0.181 0.168 0.178
amount of AVE<0.25: 5 AVE<0.25: AVE<0.25: AVE<0.25: 7/20
countries with out of 20 7/20 8/20
AVE < 0.25 countries
a
criteria for acceptable model fit: CFI>0.95 (marginal>0.92); RMSEA<0.05 (at least <0.08); SRMR
<0.05 (at least<0.08)
When cross-national comparisons of environmental concern already come with

strong methodological problems in Europe, especially in terms of scalar measure-
ment equivalence, it is even more questionable if environmental concern can be
compared between countries across the world. Table 8.4 reports the model fit and
convergent validity of environmental concern, when using all 32 ISSP countries
and the same four model specifications which were used before.
According to Table 8.4, the goodness-of-fit model W.1 without error cor-
relations is not acceptable (CFI<0.90; RMSEA>0.08; SRMR<0.08), but taking
model W.2, metric-measurement equivalence comes along with quite an accept-
able fit (marginal fit: CFI>0.920; acceptable: RMSEA<0.08; SRMR<0.08).
Thus, using model W.2, relationships of environmental concern to other con-
structs can be compared across the world, and in this sense, people from all
nations give the same basic meaning to the construct (or at least have a strong
tendency to do so).
Model W.3 with scalar invariance and Model W.4 with partial scalar invari-
ance are not acceptable (CFI<0.90; RMSEY>0.08; SRMR = 0.122 resp. 0.074).
Thus, a latent mean comparison is not possible with these models. In addition, the
results reveal that convergent validity in terms of mean AVE across all countries
is quite acceptable. However, looking at country level, validity is not given in a
third to half of the countries in the study, thus revealing a considerable problem
of measurement validity.
Table 8.4 Evaluation of measurement model across the world.
Model W.1 Model W.2 Model W.3 Model W.4

Metric Metric Metric and Metric and
invariance invariance scalar invariance partial scalar
without error with error with error invariance with
correlations correlations correlations error correlations
Model RMSEA 0.087 0.061 0.132 0.084

Fita (90% C.I.) (0.085–0.089) (0.058–0.063) (0.130–0.134) (0.082–0.086)
CFI 0.842 0.946 0.642 0.876
SRMR 0.073 0.054 0.122 0.074
Chi2 6,493.479 2,440.855 13,918.250 5,154.598
df; p 634; 0.000 442; 0.000 628; 0.000 532; 0.000
AVE: Mean Mean: 0.279 Mean: 0.251 Mean: 0.242 Mean: 0.248
AVE across Hi: GBritain: Hi: GBritain: Hi: GBritain: Hi: GBritain:
countries; 0.402 0.367 0.360 0.362
highest–lowest Lo: Taiwan: Lo: Taiwan: Lo: Taiwan: Lo: Taiwan:
value; 0.122 0.111 0.080 0.102
amount of AVE<0.25: AVE<0.25: AVE<0.25: AVE<0.25:
countries with 11/32 15/32 17/32 16/32
AVE < 0.25
a
criteria for acceptable model fit: CFI>0.95 (marginal>0.92); RMSEA<0.05 (at least <0.08); SRMR
<0.05 (at least<0.08)
Upon further inspection of the results, model W.4 does not fit in three countries
(each contribution to Chi2>400): Bulgaria, Taiwan, and Japan. By excluding these
three countries (model W.4.1), the model fit is quite acceptable with a marginal fit
in terms of CFI = 0.920, and an acceptable fit according to RMSEA = 0.069 and
SRMR = 0,061. On average, the AVE value is 0.257 and thus acceptable, but 12 out of
29 countries show AVE values lower than 0.25. All in all, model W.4.1 will be used in
the following section for a comparison of latent means across the ISSP 2010 countries.
Comparing latent means of environmental concern

On the whole, latent means of environmental concern might be compared across
Europe (using partial scalar invariance model E.4), as well as compared world-
wide (using a partial scalar invariance model W.4.1). But three shortcomings must
be kept in mind. Firstly, the model fit is only marginally acceptable, demonstrating
the remarkable differences between specified model structure and the observed
data. These goodness-of-fit results show that substantive interpretation of model
results is problematic, depending on how strictly the fit indices and their thresh-
olds are applied. Secondly, it was shown that scalar invariance is only given as
a partial invariance, thus reducing the comparability of a country’s latent means.
Thirdly, the construct of environmental concern is low in convergent validity in 7
out of 20 countries (Europe), and 12 out of 29 countries (worldwide).
198 Jochen Mayerl
Results of the worldwide latent mean comparison with model W.4.1 are given
in the following. A separate analysis of European countries reveals very similar
results in terms of ranking within this subgroup of countries (not shown).
When comparing latent means, MGCFA requires the specification of a reference
group. The latent mean of this group is fixed to zero, and the reported latent means
of all other groups (countries) indicate the latent mean difference between each of
these groups and the reference group (including p-values of significance tests of
these differences). In our case, we treat Austria as the reference group.

In Table 8.5 the results of the latent mean comparison using model MGCFA
W.4.1 are given, with high latent mean values indicating a higher latent mean than
Austria (within the metric of a 5-point rating scale). Respectively, negative values
indicate a lower latent mean relative to Austria. Reported p-values indicate the sig-
nificance level of these differences to Austria. In addition, results from convergent
validity measures are given to evaluate which countries had measurements with low
validity. The last two columns show the results of manifest means from the ‘classic’
statistical approach of computing a manifest additive index out of seven indicators.
The results show that environmental concern is highest in Canada and
Northern Europe (significant difference from Austria: Canada, Sweden,
Table 8.5 Comparison of mean environmental concern across the world: latent and
manifest approaches (reference group is Austria).
Latent Mean Comparison Manifest Mean

(model W.4.1; reference group is Austria) Comparison
(additive index of
seven indicators)
Country Latent Mean p (difference convergent Manifest

(difference to Austria) validity (AVE) Mean
to Austria) (italic: AVE <0.25)
Canada 0.129 0.000 0.285 3.476

Sweden 0.107 0.000 0.323 3.405
Finland 0.088 0.017 0.305 3.398
Switzerland 0.083 0.019 0.222 3.416
Denmark 0.072 0.061 0.294 3.430
Norway 0.058 0.099 0.250 3.294
Spain 0.008 0.819 0.178 3.185
Austria 0 — 0.334 3.396
Germany -0.095 0.012 0.287 3.254
Slovenia -0.138 0.000 0.216 3.178
New -0.139 0.000 0.288 3.244
Zealand
Argentina -0.146 0.000 0.195 3.136
Mexico -0.159 0.000 0.266 3.032

France -0.171 0.000 0.334 3.166
Turkey -0.174 0.000 0.182 3.018
Russia -0.186 0.000 0.239 3.019
Croatia -0.201 0.000 0.281 3.098
Belgium -0.202 0.000 0.338 3.158

USA -0.209 0.000 0.257 3.216
Slovak -0.222 0.000 0.310 3.069
Republic
Chile -0.243 0.000 0.230 3.088
Lithuania -0.245 0.000 0.271 2.988
Czech -0.262 0.000 0.263 3.008
Israel -0.279 0.000 0.218 3.072
South -0.290 0.000 0.124 3.049
Korea
Great -0.353 0.000 0.366 3.022
Britain
Latvia -0.372 0.000 0.240 2.924
South -0.705 0.000 0.164 2.755
Africa
Philippines -0.811 0.000 0.200 2.494
Finland, Switzerland with p<0.05, and with marginal significance in Denmark

and Norway with p<0.10). At the lower end of these latent means are the
Philippines, South Africa, Latvia, Great Britain and South Korea. It is interest-
ing to observe that countries at the lower end of environmentalism stem from
different continents and geographic directions worldwide. Thus, there is only
a tendency, but not a strong pattern, of northern and western countries to rank
high in environmentalism vs. southern and eastern countries which rank lower,
as has been argued in some of the literature (see ‘Environmental concern in
Europe and worldwide’).
As these latent means are statistically significantly different from the refer-
ence group (p<0.05), and the goodness of fit of a model with equal latent means
(FIV5 according to Table 8.1) is unacceptably poor (CFI<0.90; RMSEA>0.08;
SRMR>0.08), the factorial invariance of latent means can be falsified. Thus, when
assuming that the partial scalar invariant model (model W.4.1) is acceptable, it
can be shown that latent means of environmental concern indeed differ across
countries in Europe and all over the world.
The AVE values remind us that some of these latent means are the result of
a measurement model with problematic validity, thus asking the question about
200 Jochen Mayerl
taking the mean estimates seriously or not. In addition, the regional patterns are
quite similar for the latent and manifest approach, but there are also some remark-
able differences between the latent MGCFA and the classic manifest sum score
approach in the classification of some countries (e.g. Spain, Mexico, Turkey,
Russia, Belgium, USA, Chile, Lithuania, Israel, South Korea, Great Britain). The
most extreme differences appear in the ranking of Turkey, Mexico and Russia,
which are strongly underrated in manifest rankings, whereas the USA is strongly
overrated in manifest estimation. This emphasizes that it is useful to estimate

latent MGCFA, thus taking measurement errors into account and testing for meas-
urement equivalence − and showing that the manifest approach would lead to
(even more) biased results.
Conclusion
The aim of this chapter was to discuss threats to validity when comparing
the empirical results on environmental concern across countries, and to con-
duct an empirical application of measurement equivalence as a prerequisite of
cross-national studies. Undoubtedly we need comparable data to gain a deeper
understanding and insight into the rise and fall of worldwide environmentalism.
However, as discussed, comparative research faces very serious methodological
issues, which raise doubts about the validity of comparative conclusions regard-
ing worldwide environmental concern.
Within this context, the chapter’s focus was on why and how to test for measure-
ment equivalence of cross-national survey data. The good news for comparative
environmental research is that the chapter shows empirically that it is possible to
reach measurement models with acceptable goodness of fit for specifications of
configural and full metric invariance of a seven-indicator latent construct of envi-
ronmental concern all over the world, using ISSP 2010 data from 32 countries.
Such a model enables comparisons of causes and effects of environmental con-
cern across countries. Nevertheless, every cross-national study, using its specific
data with a specific measurement model of environmental concern, must recheck
for metric-measurement invariance before interpreting and comparing environ-
mental relationships across countries.
When trying to compare aggregated latent means of environmental concern
across countries, the prerequisite of strict scalar invariance of the indicators was
not given empirically with ISSP data. At least, partial scalar invariance reached
a marginally acceptable level of goodness of fit, thus allowing comparison of
latent means, albeit in tendency. The results showed that northern and western
countries are higher in their mean environmental concern. However, at the lower
end of the ranking, countries from all continents and geographic directions can
be found. Moreover, these results strongly depend on: 1) we accept the partial
scalar invariance model with a marginal fit; and 2) we accept comparability when
only partial scalar invariance is given but not full scalar invariance, thus allowing
a subset of indicators to vary freely across countries in terms of their intercepts.
Further research is needed to evaluate the impact of partial invariance on the

comparability issue.
In addition, while the basic patterns of country mean comparison are much
the same for MGCFA latent mean estimation and the ‘classic’ approach via com-
puting sum scores, remarkable differences have been found for some countries.
This finding underscores the importance of MGCFA to: 1) test for measurement
equivalence; and 2) take different indicator weights (i.e. factor loadings) and ran-
dom measurement errors, as well as systematic measurement errors into account.

Overall, empirical results of cross-national surveys are very important for
understanding environmental issues all over the world. Awareness, caution and
attention to methodological pitfalls must be given when performing comparative
research, and testing for measurement equivalence must be established whenever
empirical comparisons are done.
Notes
1 Throughout this chapter, ‘cross-national’, ‘cross-country’, ‘cross-cultural’ and ‘cross-
context’ are treated to be synonymous for the sake of simplicity.
2 Common applications of CFA use at least two latent constructs, thus giving the possibil-
ity to test for convergent as well as discriminant validity. In this study, only one construct
(environmental concern) is studied to avoid problems of model fit owing to problematic
measurement of the second construct, which is not in the focus of this chapter. Therefore,
we centre on the convergent validity of environmental concern and its comparability
across nations.
3 Other possibilities to deal with non-invariance are: 1) use a subgroup of countries that
show equivalence and skip the non-equivalent countries; 2) use a subset of indicators for
which the equivalence assumption holds. This is a good way to deal with the problem
when indicators are indeed interchangeable and the content of the latent construct is not
changed when skipping items; and 3) try to explain the non-invariance with additional
predictor variables, e.g. in a multilevel context using country-level predictors.
4 This means that we use a model when its fit indices indicate an acceptable fit, independ-
ent of whether the equivalence constraints are empirically wrong or not. Equivalence
might indeed be wrong in most cases (indicated by Chi2-difference tests), but as long as
the model fit still is acceptable, one should use the most comparable model, this being
the model with equality constraints (see Urban and Mayerl, 2014, pp. 225f.).
5 These error correlations are based mainly on categorization into affective, cognitive and
conative indicators, but also on other similarities of the wording of the indicators like
exaggeration or perceptions of the dispensability of the environmental topic.
6 Single-country analyses of this measurement model show that model fit is acceptable in
all 32 countries under study (CFI>0.95 in 31 countries; CFI>0.94 in 1 country; RMSEA
<0.05 in 26 countries; RMSEA<0.08 in 6 countries; SRMR<0.05 in all 32 countries).
A test of the tripartite model as a three-dimensional factor (affective, cognitive, conative)
structure showed that model fit is not satisfying (CFI<0.94) and is worse than the fit of the
model with error correlations (Fig. 8.1b). The same is true for a two-factor-solution on
the basis that indicators stem from two different item blocks in the questionnaire design
(question 10 and question 13), but again the model fit was not satisfying (CFI<0.94).
Thus we exclude these alternative models from the subsequent analyses.
202 Jochen Mayerl
7 Reported models are estimated with list-wise deletion, but re-estimations with a Full
Information Maximum Likelihood (FIML) estimator lead to similar results of model fit,
showing the results are stable with respect to missing data problems.
References
Bachleitner, R., Weichbold, M., Aschauer, W. and Pausch, M. (2014) Methodik und
Methodologie interkultureller Umfrageforschung. Wiesbaden: Springer.

Best, H. and Mayerl, J. (2013) Values, beliefs, attitudes: an empirical study on the structure
of environmental concern and recycling participation. Social Science Quarterly 94 (3).
pp. 691–714.
Beuckelaer, A. de and Swinnen, G. (2011) Biased latent variable mean comparisons due
to measurement noninvariance: a simulation study (pp. 117–148). In Davidov, E.,
Schmidt, P. and Biliet, J., (ed.). Cross-Cultural Analysis. Methods and Applications.
NY/Hove: Taylor and Francis.
Blasius, J. and Thiessen, V. (2015) Should we trust survey data? Assessing response
simplification and data fabrication. Social Science Research 52. pp. 479–493.
Blasius, J. and Thiessen, V. (2012) Assessing the Quality of Survey Data. London: Sage.
Boeve-de Pauw, J., Jacobs, K. and Van Petegem, P. (2012) Gender differences in environ-
mental values: an issue of measurement. Environment and Behavior 46 (3). pp. 373–397.
Braun, M., (2014) Interkulturell vergleichende Umfragen (pp. 757–766). In Baur/Blasius
(ed.). Handbuch Methoden der empirischen Sozialforschung. Wiesbaden: Springer.
Brechin, S. R. (1999) Objective problems, subjective values, and global environmental-
ism: evaluating the postmaterialist argument and challenging a new explanation. Social
Science Quarterly 80 (4). pp. 793–809.
Brechin, S. R. and Kempton, W. (1994) Global Environmentalism: A Challenge to the
Postmaterialism Thesis? Social Science Quarterly 75 (2). pp. 245–269.
Brown, M. W. and Cudeck, R. (1992) Alternative ways of assessing model fit. Sociological
Methods and Research 21. pp. 230–258.
Byrne, B. M., Shavelson, R. J. and Muthen, B. (1989) Testing for the equivalence of
factor covariance and mean structures: the issue of partial measurement invariance.
Psychological Bulletin 105. pp. 456–466.
Davidov, E. (2009) Measurement equivalence of nationalism and constructive patriot-
ism in the ISSP: 34 countries in a comparative perspective. Political Analysis 17(1).
pp. 64–82.
Davidov, E., Meuleman, B., Cieciuch, J., Schmidt, P. and Billiet, J. (2014) Measurement
equivalence in cross-national research. Annual Review of Sociology 40. pp. 55–75.
Diekmann, A. and Franzen, A. (1999) The wealth of nations and environmental concern.
Environment and Behavior 31 (4). pp. 540–549.
Dunlap, R. E. (1994) International attitudes towards environment and development
(pp. 115–126). In Bergesen, H. O. and Parmann, G. (eds.). Green Globe Yearbook
of International Co-operation on Environment and Development. Oxford: Oxford
University Press.
Dunlap, R. E. and Jones, R. E. (2002) environmental concern: conceptual and measure-
ment issues (pp. 482–524). In Dunlap, R. E. and Michelson, W. (eds.). Handbook of
Environmental Sociology. Westport, London: Greenwood Press.
Dunlap, R. E. and Mertig, A. G. (1995) Global concern for the environment: is affluence a
prerequisite? Journal of Social Issues 51 (4). pp.121–137.
Dunlap, R. E., Van Liere, K. D., Mertig, A. G. and Jones, R. E. (2000) Measuring endorse-
ment of the new ecological paradigm: a revised NEP scale. Journal of Social Issues 56.
pp. 425–442.
Fairbrother, M. (2013) Rich People, Poor People, and Environmental Concern: Evidence
across Nations and Time. European Sociological Review 29 (5). pp. 910–922.
Franzen, A. (2003) Environmental attitudes in international comparison: an analysis of the
ISSP Surveys 1993 and 2000. Social Science Quarterly 84 (2). pp. 297–308.
Franzen, A. and Meyer, R. (2004) Climate change in environmental attitudes? An analysis

of the ISSP 2000. Zeitschrift für Soziologie 33 (2). pp. 119–137.
Franzen, A. and Meyer, R. (2010) Environmental attitudes in cross-national perspective: a
multilevel analysis of the ISSP 1993 and 2000. European Sociological Review 26 (2).
pp. 219–234.
Franzen, A. and Vogl, D. (2012) Acquiescence and the willingness to pay for environmen-
tal protection: a comparison of the ISSP, WVS, and EVS. Social Science Quarterly 94
(3). pp. 637–659.
Franzen, A. and Vogl, D. (2013) Two decades of measuring environmental attitudes:
a comparative analysis of 33 countries. Global Environmental Change 23 (5).
pp. 1001–1008.
Guha, R. and Martinez-Alier, J. (1997) Varieties of Environmentalism: Essays North and
South. London: Earthscan.
Hair, J. F., Black, W. C., Babin, B.J. and Anderson, R. E. (2010) Multivariate Data Analysis
(7th ed.). Upper Saddle River, NJ: Pearson.
Harkness, J. A., Braun, M., Edwards, B., Johnson, T. P., Lyberg, L., Mohler, P. P., Pennell,
B.-E. and Smith, T. (eds.) (2010) Survey Methods in Multinational, Multiregional, and
Multicultural Contexts. Hoboken, NJ: Wiley.
Hu, L. T. and Bentler, P. M. (1995) Evaluating model fit (pp. 76–99). In Hoyle, R. H.
(eds.). Structural Equation Modeling: Concepts, issues, and applications. Thousand
Oaks, CA: Sage.
Hu, L. and Bentler, P. M. (1999) Cutoff criteria for fit indexes in covariance structure
analysis: conventional criteria versus new alternatives. Structural Equation Modeling:
A Multidisciplinary Journal 6. pp. 1–55.
Inglehart, R. (1995) Public support for environmental protection: objective problems and
subjective values in 43 societies. Political Science and Politics 28 (1). pp. 57–72.
Ippel, L., Gelissen, J. P. T. M. and Moors, G. B. D. (2014) Investigating longitudinal and
cross cultural measurement invariance of Inglehart’s short post-materialism scale.
Social Indicators Research 115. pp. 919–932.
doi:10.4232/1.11418
Kline, R. B. (2011) Principles and Practice of Structural Equation Modeling (3rd edition).
New York/London: Guilford.
Maloney, M. P. and Ward, M. P. (1973) Ecology: Let’s Hear from the People, an
Objective Scale for the Measurement of Ecological Attitudes and Knowledge. American
Psychologist 28. pp. 583–586.
Marquart-Pyatt, S. T. (2012) Contextual influences on environmental concerns cross-
nationally: a multilevel investigation. Social Science Research 41 (5). pp. 1085–1099.
Mayerl, J. and Best, H. (2014) Theoretische und statistische Modellierung von Cross-
Pressures in Kontextanalysen. Kölner Zeitschrift für Soziologie 66. pp. 135–154.
204 Jochen Mayerl
Milfont, T., Duckitt, J. and Cameron, L. D. (2006) A cross-cultural study of environmental

motive concerns and their implications for proenvironmental behaviour. Environment
and Behavior 38 (6). pp. 745–767.
Milfont, T., Duckitt, J. and Wagner, C. (2010) The higher order structure of environmental
attitudes: a cross-cultural examination. Interamerican Journal of Psychology 44 (2).
pp. 263–273.
Nasif, E. G., Al-Daeaj, H., Ebrahimi, B. and Thibodeaux, M. S. (1991) Methodological
Problems in Cross-Cultural Research: An Updated Review. Management International

Review 31 (1): 79–91.
Olson, K. (2013) Do non-response follow-ups improve or reduce data quality? A review of
existing literature. Journal of the Royal Statistical Society: Series A 176 (1). pp. 129–145.
Pampel, F. C. (2013) The varied influence of SES on environmental concern. Social
Science Quarterly 951. pp. 57–75.
Rosenberg, M. J. and Hovland, C. I. (1960) Cognitive, affective, and behavioural com-
ponents of attitudes (pp. 1–14). In Hovland, C. I. / Rosenberg, M. J. (eds.). Attitude
Organization and Change. New Haven, CT: Yale University Press.
Schultz, P. W. (2001) The structure of environmental concern: concern for self, other
people, and the biosphere. Journal of Environmental Psychology 21 (4). pp. 327–339.
Schultz, P. W., Gouveia, V. V., Cameron, L. D., Tankha, G., Schmuck, P. and Franěk,
M. (2005) Values and their Relationship to Environmental Concern and Conversation
Behaviour. Journal of Cross-Cultural Psychology 36 (4). pp. 457–475.
Spector, P. E., Van Katwyk, T. T., Brannick, M. T. and Chen, P. Y. (1997) When two
factors don’t reflect two constructs: how item characteristics can produce artifactual
factors. Journal of Management 23. pp. 659–677.
Stoop, I., Matsuo, H., Koch, A. and Billiet, J. (2010) Paradata in the European Social
Survey: Studying nonresponse and adjusting for bias (pp. 407–421). In American
Statistical Association (eds.). Joint Statistical Meeting 2010: Processings Section on
Survey Research Methods. Alexandria: American Statistical Association.
Urban, D. and Mayerl, J. (2014) Strukturgleichungsmodellierung. Ein Ratgeber für die
Praxis. Wiesbaden: Springer.
Urban, J. and Ščasný, M. (2012) Exploring domestic energy-saving: the role of environ-
mental concern and background variables. Energy Policy 47. pp. 69–80.
Chapter 9
Youth environmental
consciousness in Europe
The influence of psychosocial factors
on pro-environmental behaviour
Leena Haanpää
Introduction
Environmental consciousness is inevitably a matter of knowledge, attitudes and
behaviour. There is an extensive body of research focusing on these relation-
ships regarding adults (e.g. Bamberg & Möser 2007; Diekmann & Preisendörfer
2003; Kollmus & Agyeman 2002; Oreg & Katz-Gerro 2006). However, research
focusing on youth environmental consciousness has been given less attention,
especially in a cross-cultural research setting. Previous studies focus mainly
on individual-level factors within a limited student data in a separate nation
(Levine & Strube 2012), and often on one or two components of environmental
consciousness (Alp et al. 2006; Bruni, Chance & Schultz 2012; Fisman 2005;
Hausbeck, Milbrath & Enright 1992). As pointed out in earlier studies (Levine &
Strube 2012; Oreg & Katz-Gerro 2006), it is important to focus on more than a
single predictor when trying to understand pro-environmental behaviour.
This study addresses the individual conditions of environmental conscious-
ness among European youth by using the well-documented Theory of Planned
Behaviour (TPB; Ajzen 1991). Besides the TPB, Inglehart’s (1997) post-
materialism index of 17 European countries is applied to provide a more complete
picture of cultural context influencing environmental consciousness. Thus, this
study focuses on cultural values, environmental attitudes, intentions and behav-
iours of European youth. The data are drawn from the International Social Survey
Programme 2010 (ISSP 2010; European countries, total sample size 23,513
respondents), focusing on the environment, and the research population is lim-
ited to respondents between the ages of 15 and 29. A special study interest lies
in focusing on the relationships between the various psychosocial elements on
pro-environmental behaviour. These elements can be seen as one of the key com-
ponents in bringing about more a hopeful future for the environment.
Youth environmental consciousness

Various terms are used when discussing the human–nature relationship; i.e. the
orientation an individual has towards the natural environment. One of the terms
206 Leena Haanpää
focusing on this relationship is ’environmental consciousness’; a concept widely

used in environmental sociology. As such, the term environmental consciousness
can be treated as a matter of world view tracing back to different paradigmatic
conceptualizations of the relationship between humans and the nature. To put it
simply, there are two fundamental distinctions between these world views on how
humans assess the importance of nature: anthropocentric and ecocentric world
views. Various later theoretical approaches have been developed to capture the
essence of sociological thinking on these prevalent world views, of which the most
well-known ones are the human exemptionalism paradigm (HEP) and the new
environmental paradigm (NEP) (Dunlap & Catton 1994).1
The first emphasizes the uniqueness of mankind over the other creatures of the
world, while the second stresses human dependency on other species and natural
resources.
Environmental consciousness is a broad concept that often incorporates
studies related to behavioural orientation and psychological constructs. These
studies underline the multidimensional nature of the concept: cognitive,
affective and evaluative factors influencing pro-environmental behaviour.
Special attention to environment-behaviour relationship was given already
in the 1980s by Buttel (1987), who paid attention to the role of attitudes and
values. Later on, studies have emphasized the strong interconnections among
beliefs, values, attitudes and knowledge affecting pro-environmental behav-
iour (Sánchez & Lafuente 2010; Schultz 2001; Stern & Diaz 1994). Thus,
environmental consciousness is a concept through which the growth of envi-
ronmental significance in Western societies has been explained (Rannikko
1996). The concept ties together socio-psychological aspects with structural
elements which, in this study, are understood as the combination of cognitive
structure of pro-environmental behaviour and attitude-behaviour congruence.
In other words, this study builds upon previous work that has demonstrated
the existence of linkages between value-mediated attitudes and beliefs (world
views) promoting pro-environmental behavioural patterns (e.g. Stern & Diaz
1994). At an individual level, environmental consciousness refers to a personal
commitment to environmental issues, whereas collective-level consciousness
deals with, for example, national environmental movements.
It has been argued that the young generation differs from previous ones
in their environmental consciousness (Kanchanapibul et al. 2014). They are
shown to be more environmentally and socially conscious than older genera-
tions, for example, in their consumption behaviour (Hume 2010). Previous
research on youth environmental attitudes indicates that attitudes have an
influence on pro-environmental behaviour (Alp et al. 2008; Bruni et al. 2012;
Johnson & Manoli 2011; Meinhold & Malkus 2005), although this impact is
often small (Kollmus & Agyeman 2002). Positive environmental attitudes do
not necessarily lead to pro-environmental behaviour because there are other
influencing factors, one of which is behavioural intention, which has a part to
Youth environmental consciousness 207
play as a mediator of the attitudes to that very behaviour. According to Fishbein

and Ajzen, ‘behavioural intentions are indications of a person’s readiness to
perform a behaviour’ (2010, p. 39). Intentions are determined by personal atti-
tudes towards behaviour and have ‘considerable predictive validity’ (Fishbein
and Ajzen 2010, p. 68; see also Ajzen, Brown & Carvajal 2004). An abundant
body of research utilizing the theory of planned behaviour (Ajzen 1991; there
is more about this theory in the following section) shows, for example, that stu-
dents’ positive environmental attitudes influence their sustainability intention,

which then affects behaviour (Halder et al. 2013; Swaim et al. 2013).
There is also evidence that recognition of environmental problems in early
childhood may have a significance on how people as adults perceive environ-
mental issues (Strife 2012). In this study, environmental attitudes are measured
by items of environmental concern and perception of threat. Environmental
concern is an affective dimension of environmental attitude, resulting from the
awareness and beliefs which individuals have regarding harmful consequences
(Schultz 2001; Bruni et al. 2012). Environmental concern refers to the emo-
tions people have about environmental problems (Strife 2012). Xiao and Dunlap
(2007) emphasized the multifaceted nature of environmental concern which may
change over time and across populations. Considering this argument, not all
people have strong environmental concern, but other conflicting attitudes may
outweigh environmental concern. One of these is an experience of threat: for
example, one may think that the gains of industrial productivity exceed indus-
trial pollution and therefore, the perception of threat is low (Jorgenson & Givens
2014). To a certain extent, environmental attitudes are connected to values,
beliefs and world views of the environment and to social structures in which
individuals are embedded. If this social context is not considered, research fails
to take into account individual, social and institutional constraints; the combin-
ing external and internal factors (Blake 1999).
Willingness to behave pro-environmentally has proved to mediate a relation-
ship between environmental attitudes and pro-environmental behaviour (Oreg &
Katz-Gerro 2006). Theories like the Theory of Planned Behaviour (Ajzen 1991),
the Model of Responsible Environmental Behaviour (Hines et al. 1986–87), and
the Value–Action Gap (Blake 1999) have argued that attitudes do not directly
predict pro-environmental behaviour, but there are incoming factors which mod-
erate the relationship between attitudes and actions. That is why intention between
attitudes and behaviour needs to be taken into account. As intention, this study
focuses on willingness to make sacrifices for the environment.
Theories of planned behaviour and post-materialism

This study utilizes two theories widely used in environmental studies, the theory
of planned behaviour (Ajzen 1991) and post-material values thesis developed by
Inglehart (e.g. Inglehart & Abramson 1999).
208 Leena Haanpää
Theory of planned behaviour

The theory of planned behaviour (TPB) has been applied extensively in research
focusing on peoples’ attitudes and behaviour (see Oreg and Katz-Gerro 2006;
Kollmus & Agyeman 2002). TPB aims to explain certain behaviour mediated
through behavioural intentions. Thus the theory predicts behaviour that a person
has planned and has an intention to put into action. According to the theory, inten-
tion displays the strength of how strongly an individual is willing to try, and how
much effort they need to use in order to carry out some sort of behaviour. The
main rule is that the stronger the intention in question, the more likely the behav-
iour is to be realized.
TPB explains behaviour by means of beliefs, attitudes and intentions. According
to the theory, attitudes or beliefs do not determine behaviour directly, rather they
influence behavioural intentions. TPB enables the prediction and measurement of
the behaviour; for example, saving or reusing water, if a person is willing to make
sacrifices in their own standard of living for environmental reasons (Ajzen 1991).
Behavioural intentions are influenced by attitudes, subjective norms2 and per-
ceived behavioural control. Personal attitudes reflect, for example, how positive
or negative an adolescent is towards the use of public transport instead of private-
car use (cutting back on driving a car). Attitudes, in turn, are formed by world
views and beliefs that a person has towards an issue. The model also contains
factors of perceived behavioural control; internal and external limitations control-
ling an action. These limitations and control factors define how easy or difficult an
action is to accomplish. An adolescent may feel that using public transport is not
an option because it takes too much time. It points out what Ajzen (1991) in his
theory of planned behaviour suggests; cognitive self-regulation plays an impor-
tant part in it, as well as resources and opportunities.
Post-material values thesis and objective

problems–subjective values (OPSV) thesis
The rationale lying behind the post-material values thesis builds upon Maslow’s
Hierarchy of Needs (Maslow 1954). Industrialization, economic growth and
achievement motivation have traditionally been regarded as key elements of
modernity, whereas postmodernity de-emphasizes the instrumental rationality.
‘Postmodernization is a shift in survival strategies. It moves from maximiz-
ing economic growth to maximizing survival and well-being through lifestyle
changes’ (Inglehart 1997, p. 66). The shift from materialist values (economic
and physical security) to post-material values (freedom, self-expression and
quality of life) becomes crucial in postmodernity (Inglehart 1977; Inglehart
1995; Inglehart & Abramson 1999). It has been argued that once a certain stage
of economic wealth has been achieved, less basic, higher level post-material
needs become more valuable (Brechin & Kempton 1997; Guha 2000; Kidd &
Lee 1997). Such a higher-order need is, for example, environmental concern
(Jorgenson & Givens 2013). This leads people to show that they care about the
environment and are also capable and willing to pay for it. In other words, wealth
brings, alongside growth, the willingness to protect the environment.
Proponents of the post-material thesis have then shown that environmentalism
is, to a certain extent, ‘a function of actual environmental conditions’ (Abramson
1997, p. 21): people are motivated about environmental protection by direct
experience. Acknowledgement that not only people from affluent countries are
capable of perceiving environmental concern has brought forth a new explana-

tion: the objective problems–subjective values (OPSV) thesis, which proposes
that in poor countries environmentalism is spawned by a citizen’s direct experi-
ence of environmental hazards (not in ref list. Plse insert 1999; Inglehart 1995).
Environmental concern is thus seen as a subjective value caused by real environ-
mental degradation (Inglehart 1995). According to the OPSV thesis, the level of
the economy does, nonetheless, play a decisive role in making a difference in the
perception of environmental concern.
After the presentation of the OPSV thesis a wide scholastic debate emerged
over its usefulness. The main criticism against these hypothesis concerns the
anomaly of empirical testing. As pointed out by several scholars (Abramson
1997; Brechin & Kempton 1997; Dunlap & Mertig 1997; Dunlap & York 2008;
Martínez-Alier 1995), in contrary to the hypothesis, empirical testing of the
OPSV thesis has shown high levels of public concern for the environment in poor
countries, even higher than in affluent countries. The conflicting results among
different scholars then led to an uncertainty of how to explain environmentalism
as a global phenomenon (Brechin 1999). However, the major opponents (Brechin
1999; Brechin & Kempton 1997; Dunlap & Mertig 1997) agreed with what Kid
and Lee had stressed: that the OPSV is not to be understood as an aggregate-level
method but is ‘grounded at the individual level and in environmental concern’
(1997, p. 38). Bearing in mind the criticism on the usefulness of the thesis in
explaining global environmental concern Inglehart’s post-materialism thesis
offers an individual-level scoring on cultural values that serve as links to environ-
mental concern. According to prior empirical evidence, there should be a positive
correlation between post-materialism and environmental concern (Kidd & Lee
1997; Lee & Kidd 1997). In addition, the post-materialism index used as a con-
text variable may assess our understanding of pro-environmental behaviours if
direct or indirect effects are found between the two.
Aims of the present study

This study follows the theory of planned behaviour (Ajzen 1991) and aims to
predict youth pro-environmental behaviour through behavioural intentions
(Figure 9.1). A previous study utilizing ISSP environmental data and TPB
proposed a model which included the national-cultural level claiming that cul-
tural circumstances work at an aggregate level, as well as at an individual level
(Oreg & Katz-Gerro 2006). The study of Oreg and Katz-Gerro focused on adults,
210 Leena Haanpää
Post-material Environmental
values concern Pro-
Behaviour
environmental
intention
behaviour
Risk
perception
Perceived
behavioural
control
Figure 9.1 Model predicting pro-environmental behaviour (cf. Oreg & Katz-Gerro 2006).
not on youth. However, it offers a research setting that is also applicable in this
study when it comes to the combining cultural values at an individual level.
They also pointed out that social context in which the individuals are embed-
ded in should be taken into account in the analysis to demonstrate hypothesized
mediated effects. Although this work does not focus on country-level values,
post-material values are useful at the individual level, as noted above.
This study aims to identify the extent to which individual-level psychoso-
cial factors have an effect on behavioural intentions and pro-environmental
behaviours. Following Figure 9.1, the model suggests that both direct and indi-
rect relationships (which are estimated simultaneously, although not all links
are presented in the figure to keep it simple) exist between psychosocial fac-
tors and pro-environmental behaviour. Furthermore, according to the TPB, the
incoming factor intention itself has a mediator role with regard to behaviour,
while other variables have an indirect influence through it to pro-environmental
behaviour. This model suggests that Ingleharts’s post-materialism index func-
tions as a cultural context, precedes concern for the environment and also has a
mediated effect on pro-environmental behaviour. Behavioural intention – that is,
willingness to pay for the environment – is preceded by individual environmen-
tal attitudes, which in this study are performed by two measures: environmental
concern and risk perception. These two attitudinal elements are expected to have
indirect effects on pro-environmental behaviour. Individual efficacy, the ability
to contribute to environmental protection (i.e., perceived behavioural control),
is, in turn, expected to precede and have a direct effect on intention and pro-
environmental behaviour.
In addition, beyond the hypothesized effects, direct effects between environmen-
tal attitudes (environmental concern and risk perception) and pro-environmental
behaviour are anticipated. The existence of such effects would reveal the direct
significance of personal attitudes to the very behaviour which is often neglected.
The matter of interest is not to validate or devalue the various existing measures of
environmental concern or endorsement (for more about measures see e.g. Dunlap
et al. 2000), but to add to prevalent models with new knowledge about the condi-
tions of youth environmental consciousness. At the same time, the suitability of
the ISSP questionnaire for this kind of study will be evaluated.
Data, methods and variables

The data come from the International Social Survey Programme: Environment
III – ISSP 2010 (ISSP Research Group 2012), from which a total 19 European
countries were selected: Austria, Belgium, Bulgaria, Croatia, the Czech Republic,
Denmark, Finland, France, Germany, Great Britain, Latvia, Lithuania Norway,
Russia,3 Slovakia, Slovenia, Spain, Sweden and Switzerland.
The survey was conducted between 2009 and 2011. The age of the respondents
included in this study varied between 15 and 29 years.4 After this selection of
respondents, the total amount of study sample was 4,114 respondents. In 2010, the
theme of the ISSP survey was the environment, and the respondents were asked
to express their opinions and attitudes concerning different environmental issues.
Besides attitudes, the questionnaire also contained questions which measured
both the respondents’ behavioural aspects and their perceptions of environmental
problems; questions related to, for example, environment choices regarding pur-
chase behaviour or worry about the state of the environment.
The theoretical model was tested using a structural equation model (SEM)
constructed with Mplus5 (Muthén & Muthén 1998–2011). All the used variables
originated from ISSP questionnaire.
Post-material value scores were formed of two questions in which the
respondents were asked to choose firstly, one thing which they thought should
be their country’s most important thing to do, and secondly, to choose the next
important thing. Both the questions had four alternatives of which two concerned
material priorities (‘Maintain order in the nation’, ‘Fight rising prices’) and two
post-material priorities (‘Give more say in government decisions’, ‘Protect
freedom of speech’). The first choice was scored 0 or 2: materialist priorities
were given 0 points and post-material priorities 2 points. The second choice
was scored 0 (material priorities) or 1 (post-material priorities; see also Oreg &
Katz-Gerro 2006).
The concept of environmental consciousness was operationalized using items
evaluating environmental attitudes (items measuring environmental concern and
perceived threat), intentions (willingness to make sacrifices) and items of pro-
environmental behaviour measured on a four- or five-point Likert scale.
Environmental concern was formed of two items addressing the extent to
which adolescents are concerned about the environment. These measured atti-
tudes towards the environment and were labelled as ‘We worry too much about
212 Leena Haanpää
the environment and not enough about jobs’ and ‘We worry too much about pro-
gress harming the environment’. The original scale varied from 1 (strongly agree)
to 5 (strongly disagree). The items reliability alpha coefficient was 0.55 and their
zero-order correlation was 0.38 (p < 0.01).
Risk perception was measured by seven items which addressed general
threat and the level of concern for the environment. The items in the ques-
tionnaire asked for adolescents’ opinions about the danger to the environment
from: air pollution caused by cars (pollution from cars’ risk perception); air
pollution caused by industry (pollution from industry risk perception); pesti-
cides and chemicals (pesticides in farming risk perception); rivers, lakes and
streams (pollution in waterways risk perception); a rise in the world’s tempera-
ture caused by climate change (climate-change risk perception); the genetic
modification of certain crops (GM crop risk perception); and nuclear power sta-
tions (nuclear power risk perception). Regarding these items the original scale
varied from positive to negative: 1 (extremely dangerous) to 5 (not dangerous
at all), and was reverse-coded before being included in the analysis. The reli-
ability alpha coefficient of the subscale of general threat was 0.80.
Perceived behavioural control was formed of two items tapping on cognitive
self-efficacy. These were: ‘It is just too difficult for someone like me to do much
about the environment’ and ‘There is no point in doing what I can for the environ-
ment unless others do the same’. The original scale varied from 1 (strongly agree)
to 5 (strongly disagree). The items reliability alpha coefficient was 0.52 and their
zero-order correlation was 0.35 (p < 0.01).
Behavioural intention was assessed by three items measuring adolescents’
willingness to pay higher prices for the environment, willingness to pay higher
taxes for the environment, and willingness to accept cuts in the standard of liv-
ing for the environment. Response options varied between 1 (very willing) and 5
(very unwilling) and were thus reverse-coded. The reliability alpha coefficient of
behavioural intention was 0.83.
Pro-environmental behaviour was measured as follows: the question set con-
sidered private-sphere environmentally friendly behaviour such as recycling
(how often one recycles), buying organic food (buy fruit and vegetables grown
without pesticides or chemicals), energy-saving behaviour (reduce the consump-
tion of energy or fuel at home, save or re-use water) and consumerist behaviour
(avoid buying certain products). One item in this question set was left out; that
is, the one focusing on cutting back on driving a car. This decision was made
because some of the respondents in the data were underage and did not have a
driving licence, so their answers were not comparable with those that had or could
have had a licence based on their age. Again the scale was reverse-coded from
1 (always) to 4 (never). The reliability alpha coefficient of pro-environmental
behaviour items was 0.77.
The analyses, as well as the reliability calculations (Cronbach’s alpha and zero-
order correlations), were conducted with SPSS version 21. The indicators were
all continuous, thus analyses were conducted using Maximum Likelihood (ML)
estimation to perform exploratory factor analysis (EFA) and a structural equation

model (SEM). SEM analysis was chosen to examine associations between the
environmental consciousness construct for their advantages. SEM provides flexi
bility for model relationships among multiple predictors and criterion variables,
and at the same time constructs unobservable latent variables. Furthermore, it
allows statistical testing and measurement assumptions against empirical data. To
evaluate the fit of the hypothesized model, four commonly recommended fit indi-
ces were used. The adequate fit is defined as CFI>0.9, TLI>0.9, RMSEA<0.08,
and SRMR<0.08 (Byrne 2011).
To perform a SEM, two kinds of models were performed (Byrne 2011). The
first requirement was to specify a measurement model whereby the unmeas-
ured latent variables were scaled onto their related observed indicator variables
(e.g. environmental risk perception, pro-environmental behaviour). All the item
loadings were statistically significant (p <0.001; β ≥ 0.343) for their theoretical
constructs, which allowed their use as indicators in a structural model. In the
full SEM model, regression structures among the latent variables were specified
and the relationships between the latent constructs and single-item variables
were drawn.5
Results
The descriptive statistics are presented as the number of responses, means and
standard deviations for each variable (Table 9.1). The perception of risk concern-
ing air pollution from industry combined the opinions of European youngsters
the most. A third of the respondents (77%) considered pollution from industry at
least very dangerous for the environment. This variable also showed lower stand-
ard deviation than other variables: that is, more consistency among the youth
with respect to other variables. Items measuring pro-environmental behaviour,
on the other hand, show that most of the young people are not prepared to make
environmentally friendly consumption decisions. One-quarter of them (25%)
would often or always avoid buying certain products for environmental reasons.
The results also point out that willingness to make sacrifices for environment is
not self-evident for younger people. The proportion of those fairly or very will-
ing to pay higher taxes is only 20% of the total. In general, the results draw a
picture of youth that are concerned about the environment’s state at their attitu-
dinal level, but do not show special interest in taking a stand for it in their own
consumer behaviour.
Figure 9.2 confirms the estimated, positive links between the tested latent
and single-item indicators and proves all the anticipated and also positive direct
effects on factors of behavioural intention and pro-environmental behaviour. In
addition, a direct effect between risk perception and pro-environmental behav-
iour was confirmed; an effect which was estimated as indirect according to the
theoretical premises. The model fit was acceptable (χ2 = 1,752.892, df = 156;
CFI = 0.923; TLI = 0.906; RMSEA = 0.05; SRMR 0.058).
214 Leena Haanpää
Table 9.1 Number of respondents, means and standard deviations of variables used in
SEM analysis.
Variable N M SD
Post-materialism 3,812 1.28 0.979
Environmental concern
We worry too much about the environment & 3,973 3.08 1.168
not enough about jobs

We worry too much about progress harming 3,923 3.20 1.056
environment
Perceived threat
Pollution from cars risk perception 4,035 3.57 0.857
Pollution from industry risk perception 4,034 4.04 0.796
Pesticides in farming risk perception 3,972 3.75 0.914
Pollution in waterways risk perception 4,012 3.83 0.923
Climate-change risk perception 3,958 3.75 1.015
GM crops risk perception 3,740 3.38 1.108
Nuclear power risk perception 3,952 3.80 1.127
Perceived behavioural control
Too difficult for me to do anything about the 4,033 3.25 1.158
environment
No point for me to do unless others do the 4,013 3.24 1.214
same
Behavioural intention
Willingness to pay higher prices for the 3,966 2.65 1.138
environment
Willingness to pay higher taxes for the 3,951 2.39 1.131
environment
Willingness to accept cuts in standard of living 3,951 2.70 1.184
for the environment
Pro-environmental behaviour
How often recycle 3,810 2.92 1.040
Buy fruit and vegetables grown without 3,849 2.21 0.962
pesticides or chemicals
Reduce the consumption of energy or fuel at 4,030 2.16 0.930
home
Save or re-use water 4,042 1.99 0.943
Avoid buying certain products 4,002 1.97 0.865
N = 20
All the reported effects were significant at p < 0.001. As estimated, post-
material values had a quite strong direct effect on environmental concern and a
weak indirect association with pro-environmental behaviour. This indirect effect
was mediated through environmental concern and intentions (β 0.023, p < 0.001).
Post-material 0.21 Environmental 0.29

values concern Pro-
Behaviour 0.39
environmental
intention
0.15 behaviour
Risk 0.15
perception
0.31 0.27
Perceived
behavioural
control
Figure 9.2 SEM results of path model. Values are reported as standardized coefficients.
Post-material values accounted for 4% of the variance of environmental concern.

The two indicators of environmental attitudes proved to have direct effects on
behavioural intentions; a stronger association was noted from environmental
concern to intention, while risk perception had a direct relationship with pro-
environmental behaviour. The final model did not include an estimation of the
direct effect between environmental concern and pro-environmental behaviour
since it lost its statistical significance. As anticipated, perceived behavioural con-
trol (self-efficacy) had a strong direct effect on behavioural intentions and on
pro-environmental behaviour. Behavioural intentions – i.e. willingness to pay for
the environment – proved to have a strong direct effect on pro-environmental
behaviour. Latent structures of environmental concern, environmental risk per-
ception and perceived behavioural control explained a significant proportion of
behavioural intention: 22% of the variance. Accordingly, perceived behavioural
control, environmental risk perception and behavioural intention accounted for
34% of the variance of pro-environmental behaviour.
The structural equation model revealed quite strong relationships between
different indicators of youth environmental consciousness. The strongest direct
effects were observed between behavioural intentions and pro-environmental
behaviour, perceived behavioural control and intentions, and perceived behav-
ioural control and pro-environmental behaviour.
Discussion
In this study on youth environmental consciousness in Europe, in a sample of more
than 4,100 youngsters between the ages of 15 and 29, representing 19 European
countries, a model built upon the survey questions of ISSP Environment 2010 III
216 Leena Haanpää
data was shown to have an acceptable fit. Studying young people in the context
of quite a wide age range, given that the youngest presumably still live at home,
while other young adults may have already formed their own families, creates
a research setting that places quality requirements on the survey questionnaire.
Bearing this in mind, this study aimed at building a comprehensive model of
the conditions of youth environmental consciousness. In that work theories of
planned behaviour and post-materialism were utilized. Ajzen’s (1991) theory
on planned behaviour was complemented by cultural value orientations, formed

from Inglehart’s post-materialism thesis on materialist and post-materialist val-
ues (Inglehart 1997; Inglehart & Abramson 1999). Inclusion of the cultural value
context in which European adolescents are embedded as part of the model com-
plemented the psychosocial dimension of youth environmental consciousness,
and as the study results showed, post-material value orientations had quite a
strong direct relationship with environmental concern and also an indirect albeit
weak effect on pro-environmental behaviour. The result as such supports what
lies behind the thesis: postmodern value orientations are connected to higher lev-
els of environmental concern. The results also gave evidence of their associations
towards pro-environmental behaviour: post-material values had a mediated effect
for the latent variable of pro-environmental behaviour, which suggests that indi-
vidual value orientations have, to some extent, an indirect influence on youth
pro-environmental behaviours.
The results also indicate a very strong direct effect of behavioural intention on
pro-environmental behaviour; that is, willingness to pay for the environment was
positively related to behaviour aimed at diminishing environmentally harmful
actions, measured in the ISSP questionnaire by items of recycling, buying organic
food, energy-saving and consumerist behaviour.
Furthermore, the strong direct effects from intention to behaviour, as well
as indirect, mediated effects through intention to behaviour gave support to the
theory of planned behaviour. The interpretation of the results according to the
TBP strongly emphasizes that youth environmental intentions – in this case, will-
ingness to pay for the environment – really do manifest in sound environmental
behaviour.
An important notion is also that pro-environmental attitudes had a significant
effect on intentions, especially when it comes to adolescents’ level of environ-
mental concern. In addition to the direct positive effects of environmental risk
perception – that is, youth opinions about different factors threating the state of
the environment –, this environmental attitude had also a direct effect on pro-
environmental behaviour. This indicates that by affecting youth consciousness
of environmental issues positively it is possible to bring about positive change
towards pro-environmental behaviours among young people.
Some limitations concerning this study should be noted. The ISSP 2010 survey
for the environment forms an extensive body of data with many European coun-
tries represented. However, respondents’ age differences may pose challenges for
the interpretation of different questions: for instance, a 15-year-old adolescent
responding to questions about the country’s highest priorities may be a difficult

task compared to, for example, a middle-aged respondent who has more expe-
rience of politics and culture. The utilized sample represents European youth.
However, it must be noticed that they do not form a homogenous group but are
embedded in their own cultural and national contexts.
The applicability of the questionnaire in testing theories like the theory of
planned behaviour is naturally limited to some extent. In this study, an effort
was made to conceptualize one component of the TPB; namely, subjective norm.
Inclusion of one possible item to measure subjective norm, however, decreased
the total model fit indices, although the item itself was found to have statistically
significant relationships. This indicates that elaborating empirical constructs of
the secondary data is not always possible.
The utility of the research results lies mainly in the field of environmental
education. As the study results point out, taking a positive stand on environmen-
tal issues in everyday life routines may result in favourable attitudes that further
pro-environmental youth behaviour. In this sense, schools and homes are the most
prominent environments for young people, where such behaviours are possible
to actualize. Questions about lifestyles are also shaped by underlying values.
Contemporary Western consumerism is one obvious manifestation of adolescents’
lifestyles that is often connected to rampant materialism and mass consumption.
Extravagant lifestyles are claimed to have led to excessive and wasteful consump-
tion. Such a viewpoint brings to the fore environmentally harmful elements of
consumption; the fact that consumption has an intrinsic value in people’s life by
becoming a principal activity of leisure time and a search for happiness.
More research is needed to indicate the influence of aggregate-level national
context to youth environmental consciousness. Considering the age group, con-
ventional aggregate-level indicators may not work in predicting environmental
consciousness. Additional studies of the usefulness, for example, of post-material
values at the national level are needed. Do post-material values work in stud-
ies concerning youth as they did in the case of adults (cf. Oreg & Katz-Kerro
2006)? When it comes to adolescents, the evaluation of European environmen-
tal education programmes would form an aggregate-level context to deepen our
understanding of the elements affecting youth environmental consciousness.
Notes
1 See Dunlap 2002 for a more profound discussion on the competing theoretical orientations.
2 Subjective norm was left out of the final model because the ISSP questionnaire does not
contain a well-fitting item to measure it. See also Discussion.
3 Russia was included but Turkey was excluded following the most commonly accepted
geographical extent of Europe (http://en.wikipedia.org/wiki/List_of_sovereign_states_
and_dependent_territories_in_Europe).
4 2% (n = 84) of respondents of the sample population were under 18 years.
5 Model fit was revised by freeing specific parameters measuring the same type of behav-
ior or attitudes; e.g. energy-saving behavior or risk perception.
218 Leena Haanpää
References
Abramson, PR 1997, ‘Post-materialism and environmentalism: a comment on an analysis
and a reappraisal’, Social Science Quarterly, vol. 78, no 1, pp. 21–29.
Ajzen, I 1991, ‘The theory of planned behavior’, Organizational Behavior and Human
Decision Processes, vol. 50, no. 2, pp. 179–211.
Ajzen, I, Brown, TC & Carvajal, F 2004, ‘Explaining the discrepancy between intentions
and actions: the case of hypothetical bias in contingent valuation’, Personality and
Social Psychology Bulletin, vol. 30, no. 9, pp. 1109–1121.

Alp, E, Ertepinar, H, Tekkaya, C & Yilmaz, A 2006, ‘A Statistical analysis of children’s envi-
ronmental knowledge and attitudes in Turkey’, International Research in Geographical
and Environmental Education, vol. 15, no. 3, pp. 210–223.
Bamberg, S & Möser, G 2007, ‘Twenty years after Hines, Hungerford, and Tomera: a new
meta-analysis of psychosocial determinants of pro-environmental behaviour’, Journal
of Environmental Psychology, vol. 27, no. 1, pp. 14–25.
Blake, J 1999, ‘Overcoming the “value–action gap” in environmental policy: tensions
between national policy and local experience’, Local Environment, vol. 4, no. 3,
pp. 257–278.
Brechin, SR & Kempton W 1997, ‘Beyond postmaterialist values: national versus indi-
vidual explanations of global environmentalism’, Social Science Quarterly, vol. 78,
no. 1, pp. 16-20.
Bruni, CB, Chance, RC & Schultz, PW 2012, ‘Measuring values-based environmental
concerns in children: An environmental motives scale’, The Journal of Environmental
Education, vol. 43, no. 1, pp. 1–15.
Buttel, FH 1987, ‘New directions in environmental sociology’, Annual Review of Sociology,
vol. 13, no. 1, pp. 465–488.
Byrne, B 2011, Structural equation modelling with mplus: Basic concepts, applications,
and programming, Routledge, New York, Hove.
Diekmann, A & Preisendörfer, P 2003, ‘Green and greenback: The behavioural effects of
environmental attitudes in low-cost and high-cost situations’, Rationality and Society,
vol. 15, no. 4, pp. 441–472.
Dunlap, RE 2002, ‘Paradigms, theories, and environmental sociology’. In Sociological
theory and the environment: classical foundations, contemporary insights, eds. RE
Dunlap, FH Buttel, P Dickens & A Gijswijt, Rowman & Littlefield, Lanham, Boulder,
New York, Oxford, pp. 329–350.
Dunlap, RE & Catton, WR JR. 1994, ‘Struggling with human exemptionalism: the rise,
decline, and revitalization of environmental sociology’, The American Sociologist,
vol. 25, no. 1, pp. 5–30.
Dunlap, RE & Mertig, AG 1997, ‘Global environmental concern: an anomaly for postma-
terialism’, Social Science Quarterly, vol. 78, no. 1, pp. 24–29.
Dunlap, RE, Van Liere, KD, Mertig, AG & Jones, RE 2000, ‘Measuring endorsement of
the new ecological paradigm: A revised NEP scale’, Journal of Social Issues, vol. 56,
no. 3, pp. 425–442.
Dunlap, RE & York, R 2008, ‘The globalization of environmental concern and the limits of
the postmaterialist values explanation: Evidence from four multinational surveys’, The
Sociological Quarterly, vol. 49, no. 3, pp. 529–63.
Fishbein, M & Ajzen, I 2010, Predicting and changing behaviour: the reasoned action
approach, Psychology Press, Taylor & Francis, New York.
Fisman, L 2005, ‘The effects of local learning on environmental awareness in children: an

empirical investigation’, Journal of Environmental Education, vol. 36, no. 3, pp. 30–50.
Guha, R 2000, Environmentalism: a global history, Longman, New York.
Halder, P, Pietarinen, J, Havu-Nuutinen, S, Pelkonen, P, Chang, C-Y, Prokop, P & Usak,
M 2013, ‘Knowledge, perceptions, and attitudes as determinants of youths’ intentions
to use bioenergy: a cross-national perspective’, International Journal of Green Energy,
vol. 10, no. 8, pp. 797–813.
Hausbeck, KW, Milbrath, LW & Enright, SM 1992, ‘Environmental knowledge, aware-

ness and concern among 11th-grade students: New York state’, The Journal of
Environmental Education, vol. 24, no. 1, pp. 27–34.
Hines, JM, Hungerford, HR & Tomera, AN 1986–87, ‘Analysis and synthesis of
research on responsible pro-environmental behavior: a meta-analysis’. The Journal of
Environmental Education, vol. 18, no. 2. pp. 1–8.
Hume, M 2010, ‘Compassion without action: examining the young consumers’ consump-
tion and attitude to sustainable consumption’, Journal of World Business, vol. 45, no. 4,
pp. 385–394.
Inglehart, R 1977, The silent revolution: Changing values and political styles in advanced
industrial society, NJ, Princeton University Press, Princeton.
Inglehart, R 1995, ‘Public support for environmental protection: Objective problems and
subjective values in 43 societies’, Political Science & Politics, vol. 28, no. 1, pp. 57–72.
Inglehart, R 1997, Modernization and postmodernization: Cultural, economic, and politi-
cal change in 43 societies. Princeton University Press, New Jersey.
Inglehart, R & Abramson, PR 1999, ‘Measuring post-materialism’, American Political
Science Review, vol. 93, no. 3, pp. 665–677.
III – ISSP 2010. Gesis Data Archive, Cologne. ZA5500 Data file version 2.0.0, doi:
10.4232/1.11418.
Johnson, B & Manoli, CC 2011, The 2-MEV Scale in the US: A measure of children’s
environmental attitudes based on the Theory of Ecological Attitude’, Journal of
Jorgenson, AK & Givens, JE 2014, ‘Economic globalization and environmental concern:
a multilevel analysis of individuals within 37 nations’, Environment and Behavior,
vol. 46, no. 7, pp. 848–871.
Kanchanapibul, K, Lacka, E, Wang, X & Chan, HK 2014, ‘An empirical investigation of
green purchase behaviour among the young generation’, Journal of Cleaner Production,
vol. 66, no. 1, pp. 528–536.
Kidd, Q & Lee, AR 1997, ‘Post-materialist values and the environment: a critique and
reappraisal’, Social Science Quarterly, vol. 78, no. 1, pp. 1–15.
Kollmus, A & Agyeman, J 2002, ‘Mind the gap: why do people act environmentally
and what are the barriers to pro-environmental behavior?’, Environmental Education
Research, vol. 8, no. 3, pp. 239–260.
Lee, AR & Kidd, Q 1997, ‘More on post-materialist values and the environment’, Social
Science Quarterly, vol. 78, no. 1.pp. 36–43.
Levine, DS & Strube, MJ 2012, ‘Environmental attitudes, knowledge, intentions and
behaviors among college students’, The Journal of Social Psychology, vol. 152, no. 3,
pp. 308–326.
Martínez-Alier, J 1995, ‘The environment as a luxury or “too poor to be green”?’ Ecological
Economics, vol. 13, no. 1, pp. 1–10.
220 Leena Haanpää
Maslow, A 1954, Motivation and personality, Harper, New York, NY.

Meinhold, JL & Malkus, AJ 2005, ‘Adolescent environmental behaviors: can knowledge,
attitudes, and self-efficacy make a difference?’, Environment and Behavior, vol. 37,
no. 4, pp. 511–532.
Muthén, LK & Muthén, BO, 1998–2011, Mplus user’s guide, 6th edn, Muthén & Muthén,
Los Angeles, CA.
Oreg, S & Katz-Gerro, T (2006). ‘Predicting proenvironmental behavior cross-nationally:
values, the theory of planned behavior, and value-belief-norm theory’, Environment

and Behavior vol. 38. No. 4, pp. 462–483.
Rannikko P 1996, ‘Local environmental conflicts and the change in environmental
consciousness’, Acta Sociologica, vol. 39, no. 1, pp. 57–72.
Sánchez, MJ & Lafuente, R 2010, ‘Defining and measuring environmental consciousness’,
Revista Internacional de Sociología, vol. 68, 3, pp. 731–755.
Schultz, WP 2001, ‘The structure of environmental concern: concern for self, for other
people, and the biosphere’, Journal of Environmental Psychology, vol. 21, no. 4,
pp. 327–339.
Stern, PC & Dietz, T 1994, ‘The value basis of environmental concern’, Journal of Social
Issues, vol. 50, no. 3, pp. 65–84.
Strife, SJ 2012, ‘Children’s environmental concerns: expressing ecophobia’, Journal of
Swaim, JA, Maloni, MJ, Napshin, SA & Henlye AB 2013, ‘Influences on student inten-
tion and behavior toward environmental sustainability’, Journal of Business Ethics,
vol. 124, no. 3, pp. 465–484.
Xiao, C & Dunlap, R 2007, ‘Validating a comprehensive model of environmental concern
cross-nationally: a US-Canadian comparison’, Social Science Quarterly, vol. 88, no. 2,
pp. 471–493.
Chapter 10
Dilemmas of sustainability
in turbulent times
The Portuguese case in context
João Guerra, Luísa Schmidt and

Susana Valente
Introduction
This chapter considers a group of data marked by multiple dilemmas and
perplexities that challenge both citizens and decision-makers in troubled times.
The financial crisis which, since 2008, has proved to have social, economic and
ecological effects,1 has given rise to increasing difficulties in bringing together
socio-economic needs (i.e. ensuring or improving quality of life for everyone)
and environmental needs (i.e. ensuring that ecological limits are not exceeded).
Hence we face a double challenge. On the one hand, it is increasingly
acknowledged that the present socio-economic model ‘cannot deliver well-being,
environmental stability and social justice in a world where poverty and hunger
occur simultaneously with overconsumption’ (Leahy, Healy & Murphy 2014,
p. 105). However, ‘the institutional and technological locked-in situation into
which the western path of economic development ( . . . ) has led our societies’
(Van Griethuysen 2009, p. 6) imposes a situation where citizens and decision-
makers are hostages to the market and to growth. On the other hand, in line with
what has been claimed for two decades (WCED 1987), several documents focus-
ing on the sustainable development and green economy narratives have come to
light, pointing to alternative economic solutions capable of safeguarding envi-
ronmental and social values (Bina 2013; Ferrão 2014).
There are thus two distinct narratives – unlimited growth (continuing business as
usual) and sustainable development (investment in renewal and prevention) – that
diverge in a context of socio-economic crisis. Austerity policies have aggravated
social problems (weakening of the welfare state, unemployment, social exclusion).
At the same time, environmental policies and programmes have been relaxed or
neglected. Given these circumstances, present-day societies face difficult dilem-
mas. The urgent ‘needs of the moment’ are seldom confronted without negative
consequences in the present, let alone in the long run and with regard to future
generations.
In the light of this scenario, in which pessimism, disbelief and mistrust go
side by side with a narrative of hope, change and the opportunity for sustain-
ability, it is important to understand the public point of view and the factors that
222 João Guerra et al.
contribute to differing opinions. This chapter aims to analyse a group of data

from ISSP Environment III (ZA5500: International Social Survey Programme:
Environment III – ISSP 2010),2 taking different socio-geographical contexts into
account and focusing on the Portuguese example – one of the economies most
severely affected by the ongoing sovereign debt crisis since 2008/2009 – but also
looking at other European regions and the international context.
With its Southern-European character, Portugal occupies a sui generis position
within Europe as a whole. It has some of the characteristics that Mediterranean

and Latin cultures bring to democratic practices and political systems, but at the
same time is curiously similar to the former Eastern bloc countries in a number
of ways. In both Iberian and Eastern countries democracy is relatively recent,
and has emerged in a context of economic weakness. This has specific practi-
cal implications for environmental values and national sustainability performance
(Schmidt, Nave & Guerra 2006).
Therefore, viewing the Portuguese case in a comparative and contextual
approach, we will analyse the impact of the economic crisis in Europe on attitudes
to the sustainable development model which, according to the normative stance
in the Brundtland Report, Our Common Future (WCED 1987), should include its
environmental, economic and social aspects.
Methods and purposes

In order to pursue our goals, we chose a contextual set of countries based on
their availability in the ISSP 2010 dataset, on Portugal’s European status, and
also on a long historical process that potentially involves other cultural prox-
imities: America, and most particularly Latin America, which in turn is a useful
point of comparison to the European situation. Accordingly, we selected a set
of 27 countries, grouped into five different geographical areas: North America
(Canada and USA; n = 2,423), Latin America (Argentina, Chile and Mexico;
n = 4,203), Western Europe (Austria, Belgium, Denmark, Finland, France,
Germany, Iceland, Netherlands, Norway, Sweden, Switzerland, and the United
Kingdom; n = 15,318), Eastern Europe (Bulgaria, Croatia, Czech Republic, Latvia,
Lithuania, Russia, Slovakia, and Slovenia; n = 9,524) and Southern Europe (Portugal,
and Spain; n = 3,582)3.
Grouping countries in this way is not to deny their differences. It only seeks to
simplify the analysis and to embrace some common but not necessarily essential
conditions for the comparison, for exmaple: an authoritarian past and a legacy of
social distrust; aggravated economic circumstances in turbulent times; tradition and
common sociocultural characteristics. Moreover, to achieve a more robust set of con-
textual features (independent variables), we also used a set of international indicators
that may assist us in understanding underlying inequities in adopting the sustainabil-
ity case: Gross National Income per capita (GNI), the Human Development Index
(HDI), the Coefficient of Human Inequality (CHI) and the unemployment rate, all
taken from the Human Development Report 2013 (UNDP 2013).
Dilemmas of sustainability 223
Since it is not practical to cover all aspects of sustainability in an article of this

kind, our main purpose is to test – at national and regional level – the influence of
these contextual variables on three of the many possible sustainability dilemmas
faced by citizens in times of economic crisis. Each of these dilemmas results from
an apparent and supposedly difficult choice between alternative courses of action
or arguments, taken from a set of variables in the ISSP 2010 database, and whose
relationship will be tested, amongst others, by F test (differences of means) and
R2 test (coefficient of determination).

The first type, dilemmas of knowledge and concern, results from a pano-
ply of variables including: 1) the extent of self-assessed knowledge of the causes
of and solutions to environmental problems; 2) general environmental concern;
and 3) a general evaluation of national environmental performance (e.g. ‘this
country is not doing enough to protect the environment’). The different levels of
environmental knowledge and concern are mediated by diverse socio-economic
circumstances, and by the greater or lesser confidence in the national ability to
overcome environmental problems.
The second type, dilemmas of environment and economy emerges from
the analysis of: 1) the importance attributed to the environment (one of the most
important issues for the nation); 2) general environmental concern versus rela-
tive environmental concern, taking into account the idea that there is ‘too much
concern about the environment and not enough about prices and jobs’; 3) to what
extent responders agree with two different statements concerning the relationship
between the environment and the economy: a) economic growth always harms
the environment; and b) economic growth is needed to protect the environment.
Here we endeavour to identify and rethink the dichotomy between these two
areas, analysing both their independence and clashes and trying to understand
the effect of economic crisis on citizens’ motivation and willingness to protect
the environment.
Finally, under dilemmas of predispositions and impositions, we question
the best way of leading and promoting the spread of environmental protection
practices which make citizens hesitate between coercion and free will. For this,
we will use variables that: 1) adopt different approaches to protect the environ-
ment for business and for citizens (heavy fines, tax incentives, more information/
education); and 2) present two dichotomized policy measures to protect the envi-
ronment (let citizens and businesses decide for themselves, pass laws applicable
to citizens and businesses).
Dilemmas of knowledge and concern

Several studies have revealed and confirmed the social importance of knowl-
edge, information and environmental literacy as factors influencing attitudes,
opinions and values. We do not intend to relate them to a change in attitudes, but
rather to analyse how the environmental literacy rate conditions the positioning
of citizens towards basic but significant related dilemmas, such as the evaluation
of the state of the environment (worrying or not worrying) and trust in institu-
tions (trust or mistrust).
There is a great amount of information available at present, although it is
complex, contradictory and not always accessible, and demands effort and an
exceptional ability to process and filter in order to turn it into practical knowl-
edge. In this part we will analyse citizens’ ability to evaluate the present situation
(for instance, what do people know about the causes of and solutions to environ-
mental problems), considering the complexity of the issues and the fact that a lot
of information does not necessarily mean good information.
In the ISSP survey, two questions were considered to measure respondents’
self-assessment of their knowledge of the causes of and solutions to environ-
mental problems. In Portugal respondents tend to concentrate on intermediate
categories: they neither consider themselves totally ignorant nor assume clear
knowledge. However, there is a greater concentration of answers on the cat-
egories corresponding to lower levels of knowledge. At the same time, there
is a tendency to acknowledge knowing more about the causes than about the
solutions (Figure 10.1).
The greater struggle is not so much to understand the causes of a problem but
rather to imagine a possible solution to it. This stems from a complexity of factors
such as the absence of a basic culture or the fragmentation of information pro-
cesses, which generate weakness and uncertainty (Schmidt et al. 2010).
Based on both variables, we have created an ‘Environment Knowledge
Index’ which measures assumed knowledge of the causes of and solutions to
environmental problems. According to Figure 10.2, the Southern-European
countries (average = 2.84) stand out negatively in relation to their Western-
European counterparts (average = 3.27), showing lower levels of knowledge.
The results show greater similarities between the Iberian levels and those of
some of the Eastern-European countries with lower rates (Russia, Slovakia,
Lithuania, and Czech Republic), and also those of Latin-American countries
such as Chile and Argentina.
5.0%
2.1%
Causes 13.7% 17.3% 42.7% 19.2%

3.7%
2.9%
Solutions 16.8% 22.4% 39.6% 14.6%
1 Know very little 2 3 4 5 Know a lot DK/NA
Figure 10.1 Extent of self-assessed knowledge of causes of and solutions to environmental

problems in Portugal (ISSP 2010, N = 1,022).
F = (26,33.379) 73,521; p < 0.001
3.54
3.45
3.40
3.38
3.33
3.27
3.27
3.24
3.22
3.21
3.19
3.18
3.18
3.18
3.16
3.13
3.11
3.08
3.01
3.01
2.95
2.96
2.92
2.91
2.90
2.90
2.89
2.84
2.82
2.82
2.81
2.76
2.71
W. Europe
RU
SK
CL
LT
ES
PT
CZ
AR
L. America
US
BG
E. Europe
LV
MX
Total
N. America
BE
IS
GB
NL
DE
HR
AT
SE
DK
CH
FR
CA
SI
FI
NO
S. Europe
Figure 10.2 Environmental knowledge index (causes and solutions) within ISSP countries
(ISSP 2010, N = 35,051).
In Figure 10.3a the correspondence between self-assessed knowledge of the

causes of and solutions to environmental problems (R2 = 0.846) is, unsurprisingly,
very clear, but the distance between the two Southern-European countries and
Western Europe, as well as their proximity to Eastern Europe and Latin America,
is more interesting for our analysis. This may result from historically determined
similarities (e.g. an authoritarian past), from culture (post-colonial ties with Latin
America), but certainly also from common economic difficulties. Alongside them,
some of the most affluent nations like Norway, Finland and Canada, as well as
Western-European and North-American countries, stand out for their positive levels,
showing higher rates of causes and solutions knowledge of environmental problems.
This means that, when considering indicators such as GNI per capita, HDI
and the Coefficient of Human Inequality, we find correlations which confirm
the interdependency of the extent of self-assessed knowledge in different coun-
tries and their socio-economic circumstances. We may therefore conclude that
the higher the GNI per capita, the higher the stated environmental knowledge,
which may imply that the population’s environmental literacy increases overall
with the country’s available wealth (Figure 10.3b). The same tendency can be
observed regarding HDI; that is to say, generally, countries with a high Human
Development Index show higher levels of self-assessed environmental knowledge
(Figure 10.3c). Lastly, this correlation reverses when considering the Coefficient
of Human Inequality; that is, the greater a country’s inequality, the lower the
respondents’ stated knowledge (Figure 10.3d).
In this context, does the level of environmental concern mirror the level of
knowledge? According to the results in Figure 10.4, the percentage of respondents
(regardless of their nationality) who say they are not at all or hardly concerned
Knowledge of causes versus knowledge of solutions
3.60 NO
Causes of Environmental Problems
FI
CA
3.40
CH
FR
SI
DK
DE
W.Europe
3.20 SE
N.America NL
AT
TS
HR
LVBE UK
3.00 BG
US E.Europe
AR MX
PT
RU ES CZ L.America
2.80
LT CL S.Europe
SK
2.60
2.20 2.40 2.60 2.80 3.00 3.20

(a) Solutions to Environmental Problems
Notes: y = 0.24 + 1.03*x; R² = 0.846
Environmental knowledge versus GNI per capita
3.40 NO
FI
SI
CA
Environmental Knowledge Index
3.20
FR
DK AT
HR W.Europe CH
SE
3.00 NL
UK DE
TS N.America
MX
LV BE
BG AR
2.80 E.Europe
CZ US
PT
L.America S.Europe
LT SK ES
2.60 CL
RU
2.40
10.0000 20.0000 30.0000 40.0000 50.0000 60.0000 70.0000

(b) Gross national income (GNI) per captia 2013
Notes: y = 2.25 + 1.03*x; R² = 0.368

Environmental knowledge versus HDI
3.4000 NO
FI
CA
3.2000 SI
FR
CH
AT DK SE
HR NL
W.Europe
3.0000
DE
UK
MX LV N.America
BE IS
E.Europe
L.America
2.8000 PT S.Europe
US
BG AR CZ ES
CL
2.6000 RU LT
SK
2.4000
.750 .800 .850 .900 .950

(c) Human Development Index (HDI 2013)
Notes: y = 0.6 + 2.69*x; R² = 0.400
Environmental knowledge versus Human inequality
3.4000 NO
FI
SI
CA
3.2000
FR
W.Europe
DK CH HR
3.0000 NL AT
DE UK
IS
BE LV N.America
E.Europe BG L.America
2.8000
AR
CZ PT S.Europe US
SK CL
2.6000 LT ES RU
2.4000
5.000 10.000 15.000 20.000 25.000

(d) Coefficient of human inequality 2013
Figure 10.3 Environmental knowledge by GNI per capita, HDI and inequality within ISSP
countries (ISSP 2010, N = 35,051; UNDP, 2013).
Notes: y = 0.3,16 + −0.02*x; R² = 0.252
100% 5
90%
80% 4
4.10
4.07
4.01
4.03
3.91
70%
3.85
3.86
3.84
3.78
3.78
3.74
3.75
3.72
3.67
3.61
3.61
3.65
3.63
3.62
3.62
3.58
3.55
3.53
3.43
3.41
3.41
60% 3.37 3
3.37
3.35
3.25
3.10
50%
2.86
40% 2
30%
20% 1
10%
0% 0
CZ
SK
HR
LT
SE
DK
IS
E. Europe
NO
AT
RU
MX
UK
LV
NL
CL
BG
W. Europe
DE
FI
FR
L. America
ES
BE
CH
S. Europe
US
N. America
PT
AR
SV
CA
1 Not at all concerned 2 3 4 5 Very concerned Mean
Figure 10.4 General environmental concern by ISSP country and region (ISSP 2010,
N = 35,051).
with the environment is very small. Nevertheless, Portuguese respondents follow

the Canadians, Slovenians and Argentineans, showing very high concern (aver-
age > 4). In the opposite position are the Czechs, Slovaks, Croatians, Lithuanians
and, probably for different reasons, the Danes and the Swedes.
Also, according to Figure 10.4, significant differences are revealed when
comparing respondents according to their greater geographical regions. North
America (3.91), Southern Europe (3.85), to whose figures Portugal contributes
significantly, Latin America (3.75) and, to a lesser degree, although still above
the total average, Western Europe, stand out for their higher levels of concern.
Eastern Europe is the only region below the global average, with a figure of 3.41.
The results shown in Figure 10.5 help us to understand this relationship
between knowledge and environmental concern. It is possible to identify a group
of countries whose respondents are hardly concerned and have very low levels of
environmental knowledge (Lithuania, Slovakia, Czech Republic, and the average
of the Eastern-European countries). In contrast, in the upper right-hand corner we
can see another group of countries whose respondents are very concerned, and
who also show high levels of knowledge (Canada, Switzerland, Slovenia, France,
Finland . . . ). In the upper left-hand and lower right-hand corners we see coun-
tries with mixed behaviour – high literacy and low concern (Austria, Denmark,
Iceland, Sweden and Norway standing out) – and the group that includes Portugal
adding extreme concern and a low level of knowledge (Argentina, USA, Spain,
Portugal . . . ). The relationship between the two variables is apparent and
statistically significant – r (44.340) = 0.299, p < 0.001 – although in Figure 10.5
Self-assessed knowledge versus general concern

3.40 NO
FI
SI
CA
3.20
FR
W.Europe CH
HR DK NL
3.00 IS AT DE BE
SE UK
N.America
MX LV
2.80 CZ E.Europe L.America AR
CL US PT
BG
LT S.Europe
SK
2.60 RU ES
2.40
3.00 3.30 3.60 3.90 4.20

General Environmental Concern
Figure 10.5 Environmental knowledge and general environmental concern (ISSP 2010,

N = 35,051; UNDP, 2013).
Notes: y = 2.48 + 0.12*x; R² = 0.027
the mediation of the different national contexts does not permit a clear and une-
quivocal trend line.
Nevertheless, it is possible to identify significant statistical evidence of the
correlation between levels of environmental knowledge – r (41.367) = -0.035;
p < 0.001 – and concern with the environmental conditions – r (43.772) = -0.205;
p < 0.001 – when relating them to the level of confidence in the national aptitude to
solve environmental problems.
Figure 10.6 shows clear and widespread evidence of respondents’ criticism
of the performance of each of their states as far as environmental protection is
concerned. The percentage of respondents who make a negative assessment of
national performance on environmental protection exceeds one quarter in almost
every case, except for the Netherlands. In the wider socio-geographical contexts,
we can see that only Western Europe (38.7%) is below the global average (52.3%).
Latin America (72.7%), followed closely by the Iberian countries with 69.2%
(Southern Europe), are the most critical of their states’ environmental protection
actions, whereas Eastern Europe (58.7%), followed by North America (54.8%),
show intermediate levels, but are still above the global average.
If we look at the same data focusing on each country’s results it becomes
apparent that the level of affluence of the country influences the distribution of
84.2%
75.7%
72.7%
72.0%
71.8%
71.0%
70.1%
69.2%
66.3%
64.0%
62.9%
61.0%
58.3%
57.8%
56.2%
54.8%
53.5%
52.7%
50.6%
44.7%
43.6%
40.1%
38.7%
37.1%
36.0%
35.7%
35.8%
32.1%
31.1%
29.9%
27.7%
23.2%
W. Europe
S. Europe
CH
NL
FI
DE
SE
NO
UK
DK
CZ
AT
BE
SK
US
LV
N. America
HR
CA
FR
PT
RU
LT
SV
CL
ES
MX
L. America
AR
BG
IS
E. Europe
Figure 10.6 To protect the environment, this country is doing too little by ISSP country
and region (ISSP 2010, N = 35,051).
responses, at least in part. It is in rich countries, such as the Netherlands, Finland

and Switzerland that less criticism is made of the state’s action and effectiveness
in relation to environmental protection. On the opposite side are the less affluent
and more critical countries, led by Bulgaria and including Portugal, with 62.9%.
Concern with the environment also stems from a lack of confidence in institutions.
This tends to be greater in poorer countries where the economic crisis is felt more
intensely, such as in Southern Europe.
In short, although concern with the environment is a global phenomenon, the
Portuguese stand out with above-average levels of concern. In the end, this is
a distinctive cultural trait, identified in previous studies (Schmidt, Trüninger &
Valente 2004; Valente & Ferreira 2014). Nevertheless, whereas in Portugal high
levels of concern with the environment correspond to low levels of environmental
literacy, globally this correlation is neither univocal nor constant.
In Western Europe, particularly in the Scandinavian countries, the somewhat
low level of concern may be related to the presence of greater trust in institu-
tions (Listhaug & Ringdal 2008) and a higher level of environmental literacy; two
circumstances that seem to favour a more confident attitude towards problem-
solving. On the other hand, in Southern Europe, and especially in Portugal, high
levels of concern and low levels of environmental knowledge go hand in hand
with high levels of institutional mistrust, as well as with inconsistent informa-
tion policies (Schmidt & Guerra 2010). As a result, the ability to manage and
filter the available information is scarce. Caught between deep concern and a high
level of mistrust of the State respondents, especially the Portuguese, seem divided
between impotence resulting from their admitted lack of knowledge, and inaction
stemming from the alleged national inability to deal with the problems.
Dilemmas of environment and economy

Environment and economy have mainly been seen as antagonistic fields.
However, with the dawning of the sustainable development narrative, as well as
of a set of proposals based on it, environment and economy have started to be
approached through an integrated perspective. A balance between the multiple
interests at stake has been urged, at least since the publication of the Bruntland
Report (WCED 1987). Nevertheless, these two areas are not always perceived
socially as convergent. Former antagonisms seem to re-emerge, especially during
troubled times.
If concern with the environment is presented in an abstract manner – ‘Generally
speaking, how concerned are you about environmental issues?’ – the answer tends
to be straightforward and unrelated to the level of national affluence or the priority
given to environmental issues (Figure 10.7a).
However, the concern-priority correlation changes significantly when pressing
issues such as employment and cost of living mediate concern for the future of the
environment. As shown in Figure 10.7b, Portugal’s position – as well as that of
many other countries – changes dramatically, appearing well below previous con-
cern levels.4 In other words, in the presence of other socio-economic urgencies,
sharpened in times of crisis, high environmental concern lessens quite remark-
ably, especially in less affluent countries.
Let us then go further in our analysis by introducing other variables. Taking
environment and economy to be opposing forces we will now analyse the
respondents’ reaction to the statement ‘economic growth always harms the
environment’. Figure 10.8 shows that the perception of economic growth being
harmful to the environment is greater in Latin America; probably owing to the
still few convincing examples of compatibility between economy and environ-
ment. Portugal (2.95) is near the Eastern-European countries’ average (2.92)
and very far from Western Europe (2.79), and even further from North America
(2.56). In this last region, as in the Nordic countries, the dichotomy of envi-
ronment and economy seems to have faded away, perhaps as the result of a
more successful process of ecological modernization that made the compatibility
hypothesis more credible (Mol, Sonnenfeld & Spaargaren 2009).
Nevertheless, it should be noted that the concurrence average never gets close
to the scale’s highest level (5). This suggests a general and pervasive refusal of
a dichotomist vision that opposes environment and economy in a simplistic way.
This approach has been gradually replaced by the sustainable development per-
spective, exhibiting more moderate principles and integrating both spheres. Thus,
the ideas of sustainable development and the green economy seem to be gain-
ing ground and, despite fleeting reversals, have already won a place in the social
imagination and are likely to extend to several socio-geographical contexts.
Comparing the results of ISSP Environment III (2010–12) with ISSP
Environment II (2000) (see, for instance, Schmidt & Valente 2004) the Portuguese
case is one of the most interesting (Figure 10.9). It is among the Portuguese
General concern versus environmental importance
4.20
CA
AR SI
N.America
General Environmental Concern
PT S.Europe
3.90 CH
L.America US BE
ES DE
BG UK FR FI
MX
3.60
CL LV RU NL W.Europe
E.Europe AT
LT IS SE NO
3.30 DK
HR
SK
3.00
CZ
0.0000 0.1000 0.2000 0.3000 0.4000

(a) Environment as one of the most important issues
Notes: y = 3.63 + -‘0.01*x; R² = ns
Too much worry versus environmental importance

3.40
LV
Too much worry about environment and
SI FR
not enough about prices and jobs
3.20 LT UK
ES RU BE
HR CZ FI
E.Europe SK
3.00 BG US MX NL
S.Europe W.Europe
CL
L.America
2.80 AP IS N.America
PT AT CH
DE NO
SE
CA
2.60
DK
2.40
0.0000 0.1000 0.2000 0.3000 0.4000

(b) Environment as one of the most important issues
Figure 10.7 Importance of environment concern and other priorities (ISSP 2010,

N = 35,051).
Notes: y = 3.12 + -1.12*x; R² = 0.249
F(34, 45.696) = 125.953; p < 0.001
3.29
3.28
3.22
Minimum 1 – Maximum 5
3.18
3.06
3.04
3.01
3.00
2.96
2.95
2.95
2.94
2.92
2.91
2.90
2.89
2.83
2.82
2.81
2.79
2.75
2.73
2.72
2.67
2.59
2.59
2.56
2.56
2.55
2.52
2.49
2.31 IS
NO
DK
US
N. America
CA
SE
ES
UK
HR
LV
S. Europe
W. Europe
NL
BE
DE
RU
FR
CZ
E. Europe
FI
PT
AT
LT
CH
SK
SV
AR
BG
L. America
CL
MX
Figure 10.8 Economic growth always harms the environment (ISSP 2010, N = 35,051).
respondents that the most perceptible decrease in agreeing with the statement
‘economic growth always harms the environment’ takes place. This decrease in
agreement – from 60% to 35% – suggests that, at least in Portugal, economic
growth is no longer seen with the former distrust.
Thus, as far as Portugal is concerned, there is a persistent high appraisal of
‘economic growth’ as an increased necessity in a crisis context, which, to a cer-
tain extent, overshadows (but does not deny) the environmental values. Actually,
as several studies have shown, the progress of the sustainable development
and green economy narratives is undoubtedly real. The belief in a new green
economy occurs mainly amongst the younger generations, who have the higher
literacy levels and skills to process the environmental information discussed
above (Ferrão 2014; ed. Schmidt & Delicado 2014).
0.31
0.22
0.10
0.06
0.05
0.03
0.03
0.02
0.02
0.02
0.02
−0.01
−0.04
−0.06
−0.07
−0.09
−0.11
−0.14
−0.15
−0.76
PT
RU
US
MX
BG
CHL
NL
ES
SVN
UK
SE
LV
CL
DK
CA
NO
DE
FI
AT
CZ
Figure 10.9 Economic growth harms the environment. Evolution between 2000 and
2010–12 (ISSP 2010, N = 35,051; ISSP 2000, N = 28,389).
The results presented in Figure 10.10 confirm this standard of response.

Portugal (3.63) comes close to the average of the Eastern-European countries
(3.19), and is above the average of the Western-European (3.19), North-American
(3.30), Latin-American (3.36) and Southern-European (3.47) countries, exhibit-
ing a significant difference between the Iberian countries. Yet, it is noteworthy
that the lowest levels of agreement with the notion that environmental protection
needs economic growth are found in rich countries such as Switzerland, Finland
and Norway, which have already reached economic and environmental well-
being. At the opposite end we find countries such as Bulgaria, Czech Republic,
Russia and Latvia, where the integration of environment and economy is still to be
achieved and where the levels of institutional trust are very low.
In short, present-day experience shows that the population’s disregard for envi-
ronmental problems reflects the economic crisis and poverty. However, in absolute
terms, concern with environmental problems is still high, because the govern-
ment’s lack of investment imposed by austerity measures leads to decisions that
neglect quality of life and the environment. The Portuguese, as with other people
affected by the crisis, feel this situation and exhibit paradoxical positions. On the
one hand, they prioritize the environment and disregard the economy, and on the
other hand, they value the economy and growth as essential means of ensuring
both social and environmental well-being.
In Portugal, as in other countries, the results show a decrease, varying across
generations and social groups, of the perception of the economy as harmful and
increased recognition of its potential to protect the environment. It is the coun-
tries going through economic crisis or with low indicators of development and
affluence, such as the Southern and Eastern-European countries, which stress that
environmental protection depends on economic growth. On the other hand, it is in
Western Europe and in countries with a higher GNI per capita (where the standard
F (34, 45.899) = 99.696; p < 0.001

3.83
3.76
3.75
3.73
Minimum 1 – Maximum 5
3.66
3.63
3.62
3.60
3.50
3.48
3.47
3.47
3.41
3.40
3.36
3.36
3.35
3.34
3.32
3.32
3.31
3.30
3.24
3.19
3.19
3.18
3.07
3.07
3.04
2.97
2.71
2.70 CH
FI
NO
CA
AT
NL
IS
W. Europe
BE
FR
N. America
CL
SE
DK
UK
DE
L. America
AR
ES
MX
S. Europe
US
HR
LT
SK
PT
E. Europe
LV
RU
CZ
BG
SV
Figure 10.10 Country needs economic growth to protect environment (ISSP 2010,

N = 35,051).
of living is stable) that this perspective is less marked. This means that economic
security is an important condition, although not a sufficient one, for ensuring the
population’s unequivocal support for environmental-protection policies. Despite
a tendency to subscribe to the sustainability and green economy narrative, seen
as a hopeful route, the threat of economic insecurity produces social hesitation,
even though it is known that short-term decisions may have unwanted and lasting
environmental side effects.
Dilemmas of predispositions and impositions

According to Adger & Jordan (2009), the sustainable development concept
embodies two greater and all-encompassing aspects: 1) results – the global quality
of sustainability that includes human well-being and the well-being of the ecosys-
tem; and 2) processes – the way and the means that lead to those results, that is to
say, sustainability governance. Bearing this in mind, it is important to understand
which interventions and strategies (processes) are chosen to conduct and support
environmental protection (results).
Thus, the dilemmas under analysis focus on the greater or lesser accept-
ance of states’ coercive means of protecting the environment (for instance, laws
and fines), of the collection of non-compulsory taxes, which are dependent on
institutionalized public policies, and of the State’s regulatory power (such as
tax breaks), or mobilization and persuasion relying mainly on the public’s will-
ingness to contribute to the common good (such as informative and enabling
educational activities).
According to Figure 10.11, there is a clear tendency to support less coer-
cive solutions for citizens and more coercive ones for businesses. Portuguese
respondents (and consequently the Southern Europeans) are those who most
value information and education, instead of coercive measures aimed at citizens
(66.3%). On the other hand, Eastern Europe supports heavy fines for citizens and,
above all, for businesses (37% and 46.9%, respectively). Latin-American coun-
tries follow suit (30.6% and 47.8%, respectively). As for the Western-European
and North-American countries, they seem to reject coercive measures and to sup-
port environmental information and education as well as tax breaks – ‘Use the
tax system to reward people/businesses that protect the environment’ – stressing
the importance of accountability and reflecting green taxation policies already in
place in many of those countries.
On the whole, despite differences between geographical contexts, levying
fines on those who cause environmental damage is the preferred measure
(39.2% among respondents of the countries represented here), followed by tax
breaks that reward good environmental performance (32.7%) and lastly, infor-
mation and education (22.7%). Amongst those who object the most to heavy
fines are countries with a heavy tax burden, such as Portugal and many other
countries affected by the economic crisis. It is also in these countries that we
find greater mistrust of the end use of revenue from fines. The highest levels of
4.2% 4.4% 2.3% 2.5% 1.8% 1.7%

5.9% 6.1% 5.7% 5.1% 5.3% 6.4%
15.1%
22.5% 23.6%
30.6% 29.1% 35.0%
41.9% 45.9%
43.6% 47.5%
57.9%
31.9% 66.3%
18.6%
33.9%
26.5% 39.8%
26.4%
17.7% 21.8%
39.2% 38.3% 16.6%
46.9% 47.8% 15.3%
37.0% 36.1% 36.2%
31.7% 30.6% 30.6%
23.3%
13.1% 12.8% 16.6%
EE- EE- WE- WE- NA- NA- LA- LA- SE- SE- PT- PT-
Citizens Business Citizens Business Citizens Business Citizens Business Citizens Business Citizens Business
Eastern Europe Western Europe North America Latin America Southern Europe Portugal
Heavy fines Tax system rewards More information/education DK/NA
Figure 10.11 To protect the environment: best approach for citizens and for businesses
(ISSP 2010, N = 35,051).
support for tax breaks are found in the Western-European and North-American
countries. In these regions, a more conciliatory approach to environmental
protection needs and market needs is apparent. This perspective feeds on the
gradual implementation of green taxation measures and the ensuing tax breaks.
It seems unquestionable that in Portugal and Southern Europe support prevails
for procedures based on increased environmental information and education. This
reveals a greater trust in the power of the dissemination of information and envi-
ronmental knowledge. At the same time, it shows a preference for measures with
less impact on citizens’ budgets.
We endeavoured to assess whether respondents believe that, as far as envi-
ronmental protection policies are concerned, government action should be more
coercive or more tolerant, both for citizens and businesses. According to the
results shown in Figure 10.12 and, to an extent, contradicting the previous
results, more coercive measures for citizens largely surpass the other options
(‘government should pass laws to make ordinary people protect the environ-
ment, even if it interferes with people’s rights to make their own decisions’).
Analysing this seeming contradiction, it is the Southern-European and Latin-
American countries that support tougher governmental action (above 70%).
At the opposite end North American and Western-European responses favour-
ing more coercive measures do not exceed 58.6%. These countries stand out,
on the one hand, for their greater support of free will or citizen responsibility
(‘government should let ordinary people decide for themselves how to protect the
environment, even if it means they don’t always do the right thing’) and, on the
3.5%
8.0% 6.8% 7.6% 8.3% 5.0% 6.8%
12.5% 10.7% 11.6%
18.4%
24.8%
73.5% 78.8% 80.4%

66.8%
79.2% 58.6% 84.4%

51.6% 80.6% 81.7% 89.0% 89.9%
23.6% 23.0% 20.7% 18.9% 17.7% 14.6%

12.7% 8.8% 8.7% 2.6% 3.3%
6.7%
NA- NA- WE- WE- EE- EE- LA- LA- PT- PT- SE- SE-
Citizens Business Citizens Business Citizens Business Citizens Business Citizens Business Citizens Business
North America Western Europe Eastern Europe Latin America Portugal Southern Europe
Let citizens and entreprises decide for themselves Pass laws to citizens and entreprises NA/DK
Figure 10.12 In order to protect the environment, government should . . . (ISSP 2010,

N = 35,051).
other hand, for their higher percentages of ‘don’t know/no reply’, perhaps reflect-
ing a refusal to concede the dichotomy presented. It may be a way of expressing
an intermediate position, integrating both forms of environmental governance.
Respondents’ attitudes to businesses reveal an acceptance of stricter govern-
mental behaviour: ‘government should pass laws to make businesses protect the
environment, even if this interferes with businesses’ rights to make their own
decisions’. These measures seem to be perceived as a means of protecting the envi-
ronment; an interpretation that cuts across all socio-geographical contexts. It is once
again in Southern Europe where people are more peremptory in defending State
intervention to make businesses act in a way that ensures environmental protection.
We will now take a look at the distribution of respondents when crossing diver-
gent options of environmental governance, such as ‘more information/education
for ordinary people and for business’, and ‘laws that make citizens and businesses
respect the environment’. Figure 10.13 shows that Portuguese respondents stand
out in supporting quite noticeably both more information and stronger laws to
protect the environment. Crossing the ‘more information and education’ option
with GNI per capita shows that responses do not seem to be strongly related to
affluence, even if Eastern-European countries (with a lower GNI per capita) value
information and education the least.
In sum, the best way to intervene for environmental protection seems to be a
hard-to-solve dilemma for most respondents, and especially for the Portuguese.
On the whole, the answers reveal respondents divided between support for
more information and education on the one hand, and reinforcement of stricter
More information/education versus coersive laws
1.20
PT
More information and education
1.00
S.Europe
IS
NO L.America
AR ES
0.80 US FR
MX CL
DX
SI
N.America SE
FI
0.60 W.Europe
UK NL BE CH
CA
LV LT DE
HR E.Europe
0.40 CZ SK AT
BG
RU
0.20
(a)
0.50 0.60 0.70 0.80 0.90
(a) Laws to protect environment
More information/education versus GNI

1.20
PT
More information and education
1.00 S.Europe
IS
AR DK NO
0.80 L.America ES FR US
MX SI FI
CL
UK N.America
SE
0.60 LV W.Europe BE NL CH
LT
HR DE
CZ CA
0.40 E.Europe AT
SK
BG RU
0.20
10,0000 20,0000 30,0000 40,0000 50,0000 60,0000 70,0000

(b) Gross national income (GNI) per capita 2013
Figure 10.13 More information/education on environmental protection versus coercive

laws and GNI (ISSP 2010, N = 35,051).
measures on the other. Despite demanding stricter laws and a tougher State to
secure environmental protection, many respondents are unsure about accepting
the higher costs of that protection. Thus, they support more liberal measures
for themselves and stricter measures for others (businesses). This seeming
contradiction cuts across the regions, but is more apparent in more vulnerable
socio-geographical contexts, driven by the effects of the financial crisis. Such is
the case in Portugal, where citizens feel that the imposed solutions to the finan-
cial crisis result essentially from the sharpening of their subjective and objective
deprivation (Ferreira 2011) on the one hand, and on the other, have little trust in
business leaders, or in political power and its counterparts (Schmidt & Guerra
2010; Schmidt, Nave & Guerra 2006) and therefore feel they cannot do without
laws that are meant to protect the common good.
Concluding Remarks
In the multilayered troubled times experienced in Europe since 2008, the imbal-
ance between economic, social, environmental and political aspects is acutely
challenging. This challenge grows bigger in the face of the failure of the present
growth model – the ‘business as usual’ narrative – in reconciling these aspects, and
the weakness of the alternative model – the sustainability narrative – whose dif-
ficulties of execution have been highlighted by several authors. Michael Redclift
(2005) states that it is the superficial consensus of the debate around sustainable
development, converging on multiple parallel and contradictory discourses, that
actually hampers its progress. After all, it is the ‘white blackbird that no one has
ever seen’ and despite its rhetoric, the ‘decolonization’ process of the social imag-
inary is far from finished (Latouche 2004; 2005). This image postulates economic
growth as an indisputable value, perpetuating misunderstandings and contra-
dictions as reflected in the results presented here. However, successes in some
contexts seem to have increased public support for the alternatives related to sus-
tainable development, including so-called green growth, especially in Northern
Europe, as shown by those countries’ indicators.
As for Eastern and Southern Europe, there is also a wish for a sustainable bal-
ance between the different areas of development. However, at the same time, their
willingness to change is still low. In the Portuguese case, three major factors help
to explain this state of affairs:
1 The crisis and austerity context to which the country has been subjected
since 2010 has centred citizens’ concerns on a daily life marked above all by
economic scarcity. This tends to overshadow the environmental problems,
although it does not do away entirely with concerns about them, partly owing
to the government’s disregard of environmental policies.
2 The lack of environmental information, knowledge and culture leads to
greater difficulty in perceiving an alternative path. It is not understood how to
make sustainability and a green economy a fact and this blocks the citizen’s
power to go beyond the ‘business as usual’ view (although some younger

segments of the population have a greater desire to ‘make a difference’).
3 At the heart of these issues lies respondents’ inability to anticipate truly sus-
tainable development, because they do not trust the State. They find it weaker
each day, with no capacity to fulfil strategic functions while ensuring greater
social and environmental justice.
However, although it is true that accumulated mistrust persists and affects the
ability to think of an alternative future, to many respondents the State continues
to be the necessary assurer of the protection and defence of the environmental
common good. In that sense, confidence in, and the credibility of governmental
institutions emerge as fundamental elements of sustainability; more so because
the socio-economic status quo to which we were accustomed ‘failed in such a
spectacular fashion that it is unclear when, if ever, it will recover in anything like
its previous form’ (Sim 2010, p. x). This global disruption creates new dilemmas
for today’s societies and particularly for the socio-geographical contexts where
the system’s breakdown is greater, and on that account, the environmental dilem-
mas are more important to citizens.
An alternative programme to the present socio-economic model cannot be
easily put in place. We can only hope that the newer generations – with better
education, more information and environmental skills – can create the space to
implement the dynamics of economic, political, environmental and social moder-
nity; in short, dynamics that lead to a sustainable future.
Notes
1 With the ongoing financial crisis, government policy has focused on the need for eco-
nomic growth, and economic hardship has increasingly been used as a justification for
delaying action on critical environmental issues (Tienhaara 2009).
2 The data sets from the Netherlands and Portugal were delivered to the Archive after the
integrated file had already been prepared in 2014. The years in which fieldwork was car-
ried out in these countries vary from 2009 (e.g. Slovak Republic) to 2012 (e.g. Portugal).
3 It is a grouping based on a limited number of available countries, which – as in the case
of Southern Europe – is far from representing the whole region.
4 In order to measure the level of environmental concern, agreement with the sentence
‘worry too much about the future of the environment and not enough about prices and
jobs today’ was inverted. 1 corresponds to the highest degree and 5 to the lowest.
References
Adger, Neil and Jordan, Andrew (2009) (eds.). Governing Sustainability. Cambridge:
Cambridge University Press.
Bina, Olivia (2013). ‘The green economy and sustainable development: an uneasy bal-
ance between the imperatives of crisis management and persisting contradictions’.
Environment and Planning C-Government and Policy, 31: 1023–1047.
Ferreira, António C. (2011), ‘A sociedade de austeridade: Poder, medo e direito do tra-

balho de exceção’, Revista Crítica de Ciências Sociais, Onde para o mercado? 95:
119–136.
Ferrão, João (ed.) (2015). Que economia queremos? Lisbon: Fundação Francisco Manuel
dos Santos.
doi:10.4232/1.11418.
Latouche, Serge (2004). Survivre au développement: De la décolonisation de l’imaginaire
économique à la construction d’une société alternative. Paris: Mil et une Nuit.
Latouche, Serge (2005). Décoloniser l’imaginaire: La pensée créative contre l’économie
de l’absurde. Lyon: Parangon.
Leahy, Ann, Healy, Seán & Murphy, Michelle (2014). The European Crisis and its Human
Cost: A Call for fair and Alternatives and Solutions. Brussels: Caritas Europa.
Listhaug, Ola & Ringdal, Kristen (2008). ‘Trust in political institutions’. In Nordic Social
Attitudes in a European Perspective. Ervasti, Heikki, Fridberg, Torben, Hjerm, Mikael &
Ringdal, Kristen (eds.). Cheltenham: Edward Elgar, pp. 131–151
Mol, Arthur P.J., Sonnenfeld, David A., and Spaargaren, Gert, (eds.) 2009. The Ecological
Modernisation Reader: Environmental Reform in Theory and Practice. London and
New York: Routledge.
Redclift, Michael. (2005). ‘Sustainable development (1987–2005): an oxymoron comes of
age’. Sustainable Development, 13 (4): 212–227.
Schmidt, Luísa & Delicado, Ana (Eds.) Ambiente, Alterações Climáticas, Alimentação e
Energia: A opinião dos portugueses. Lisbon: Imprensa de Ciências Sociais.
Schmidt, Luísa & Valente, Susana (2004), Factos e opiniões: uma abordagem transna-
cional ao desenvolvimento sustentável. Lisbon: Imprensa de Ciências Sociais.
Schmidt, Luísa and Guerra, João (2010). ‘Da Governança Global à Sustentabilidade Local:
Portugal e o Brasil em Perspectiva Comparada’. Revista de Ciências Sociais, 41 (2):
101–124.
Schmidt, Luísa, Nave, Gil & Guerra, João (2006), ‘Who’s afraid of Local Agenda 21?
Top-down and bottom-up perspectives on local sustainability’, International Journal of
Environment and Sustainable Development, 5 (2): 181–198.
Schmidt, Luísa, Nave, Gil & Guerra, João (2010), Educação Ambiental: Balanço e
Perspectivas para uma Agenda mais Sustentável. Lisbon: Imprensa de Ciências
Sociais.
Schmidt, Luísa, Trüninger, Mónica & Valente, Susana (2004). ‘Problemas ambientais,
prioridades e quadro de vida’, in Ferreira de Almeida, João (ed.), Os Portugueses e o
Ambiente: II Inquérito Nacional às Representações e Práticas dos portugueses sobre o
Ambiente, Oeiras, Celta: 65–172.
Sim, Stuart, (2010), The End of Modernity: What the Financial and Environmental Crisis
is really telling us. Edinburgh: Edinburgh University Press.
Tienhaara, Kyla (2009), ‘A Tale of Two Crises: What the Global Financial Crisis Means
for the Global Environmental Crisis. Amsterdam: Global Governance Working Paper
No. 37. The Global Governance Project. Available at www.glogov.org (accessed 14
March, 2016).
UNDP (2013), Human Development Report 2013: The Rise of the South: Human Progress
in a Diverse World. New York: United Nations Development Programme.
Valente, Susana & Ferreira, José G. (2014), Ambiente: das preocupações às práti-
cas in Schmidt, Luísa & Delicado, Ana (eds.) Ambiente, Alterações Climáticas,
Alimentação e Energia: A opinião dos portugueses. Lisbon: Imprensa de Ciências
Sociais, 31–74.
Van Griethuysen, Pascal (2009) ‘Why are we growth-addicted? The hard way towards
degrowth in the involutionary Western development path’, Journal of Cleaner
Production (xxx): 1–6.
WCED (1987) Our Common Future. Report of the World Commission on Environment
and Development. Geneva: United Nations.
Index
Abramson, P. 59, 209 Ajzen, I. 207, 208, 216

ACA see Attitudes of Concern and altruistic values xix, 61
Activation index AMS see Attitudes of Material Sacrifice
activism 8, 91–110, 112–115, 133; index
cognitive mobilization 100–101; Ančić, Branko 8, 157–181
data and methods 93–94; degrowth Appelbaum, R.P. 34
160; economic causality 96–99; Argentina: environment and economy
globalization of 151; individual-level dilemmas 232–234; environmental
variables 135–137; mobilization across concern 71, 79, 190, 198; environmental
countries 95–96; post-materialism knowledge 224, 225–230; policies 238;
99–100; religion 102–106; social capital risk/threat perceptions 19–22
101–102; see also behaviour; donations; Asia 78, 79
membership of pro-environmental attitudes xviii, 2, 31, 133, 206; degrowth
groups; petitions; protests potential 157, 160, 161, 163–164;
Adger, Neil 235 environmental concern 57, 64–66, 190,
affluence xx, 114; awareness linked to 65; 210–211; environmental cultures 93;
degrowth potential 168–169, 170, 172, hierarchical causal chain model 184;
173–174, 175, 176–177; environmental public-sphere environmental behaviours
activism 96, 108, 128–129; 139, 143–147, 149, 150; risk/threat
environmental concern 56–57, 58–60, perceptions 16, 28; Theory of Planned
73, 80–82, 160–161, 183; environmental Behaviour 208; tripartite model of 190;
utilization 61–62; public-sphere values and 61; youth environmental
environmental behaviours 138–139, 145, consciousness 206–207, 211, 216; see
149, 150; relative wealth 121, 129–130; also environmental concern
risk/threat perceptions 17, 23, 27, 28, 48; Attitudes of Concern and Activation
views on state environmental protection (ACA) index 163–164, 167–170,
229–230; willingness to pay 138 173–174, 175
age: awareness, efficacy, and willingness Attitudes of Material Sacrifice (AMS)
to pay 72, 74; environmental activism index 164, 165, 167–170, 173–174, 175
94, 103, 113, 137; environmental Australia: environmental concern 71, 79;
concern 67, 160–161, 183–184; risk/ risk/threat perceptions 17, 18, 19–22
threat perceptions 17, 24–26, 28, 32–33, Austria: degrowth potential 171–174;
37, 42–43, 46; United States xvii environment and economy dilemmas
agriculture 63, 70, 76, 80, 142 232–234; environmental activism
air pollution: Asia 78; Environmental 95, 97, 104, 116–117, 119, 121, 124,
Performance Index 142; regulation 126–128; environmental concern 71,
of 67, 80; risk/threat perceptions 13, 79, 190, 198; environmental knowledge
17–23, 27, 36, 40–41, 139, 212, 214 225–230; policies 238; public-sphere
244 Index
environmental behaviours 140; risk/ Brand, Karl-Werner xviii

threat perceptions 18, 19–22, 35, 40, 47, Brechin, S.R. 56–57, 60
49–51 Brody, S. 63
awareness 7–8, 63, 65, 72–82; affluence Bulgaria: degrowth potential 171–174;
and 161; CEEs 4; confirmatory factor environment and economy dilemmas
analysis 66–67, 68; degrowth potential 232–234; environmental activism
176–177; environmental activism 95–97, 104, 106, 116–117, 119–121,
113–114, 124–125, 129, 135; risk/threat 124, 126–128; environmental concern
perceptions 33, 45 71, 79, 190, 197; environmental

knowledge 225–230; human
Bachleitner, R. 187 development 162; policies 238; public-
Balžekianė, Aistė 7, 31–55, 111 sphere environmental behaviours 140;
Beck, Ulrich xxi, 4, 5, 13, 14, 16, 28, 33 risk/threat perceptions 19–22, 35, 40,
behaviour 2, 31, 111–112, 129, 206; 47, 49–51; salience of environmental
attitudes linked to 210–211; CEEs 4; issues 118; trust 102
degrowth potential 157; hierarchical Butkevičienė, Eglė 8, 111–132
causal chain model 184; private-sphere Buttel, Frederick H. 6, 206
118–120, 123–125, 126, 129–130, 133,
136; public-sphere 118–120, 123–125, Calvinism 105
126, 129, 133–151; religion influence on Canada: degrowth potential 175;
115; Theory of Planned Behaviour xix, environment and economy dilemmas
205, 207–208, 209, 216, 217; values and 232–234; environmental concern 71,
61; youth environmental consciousness 79, 135, 190, 198; environmental
206–207, 210, 212, 213, 214, 215, 216; knowledge 225–230; policies 238; risk/
see also activism threat perceptions 19–22
behavioural spill-over effect 135–136 car use 119, 142, 147
Belgium: degrowth potential 171–174; Catholicism 102–104, 107
environment and economy dilemmas Central and Eastern-European countries
232–234; environmental activism (CEEs) 8, 78, 79; environmental
95–97, 104, 116–117, 119, 121, 124, activism 104, 106; environmental
126–128; environmental concern 71, sociology 3–4; religion 122; risk/
79, 190, 199, 200; environmental threat perceptions 39–40, 51; salience
knowledge 225–230; policies 238; of environmental issues 118; see also
public-sphere environmental behaviours Eastern Europe
140; religion 107; risk/threat perceptions CFA see confirmatory factor analysis
18, 19–22, 23–26, 35, 40, 47, 49–51; Chase-Dunn, C. 163
salience of environmental issues 118 CHI see Coefficient of Human Inequality
beliefs xix, 113, 133, 206; hierarchical children 18, 24–26
causal chain model 184; social context Chile: environment and economy dilemmas
207; Theory of Planned Behaviour 208 232–234; environmental concern 71,
Benighaus, C. 32 79, 190, 199, 200; environmental
Best, H. 184 knowledge 224, 225–230; policies 238;
Bickerstaff, K. 28n1 risk/threat perceptions 19–22
biodiversity: Ecosystem Vitality 35; Choi, S.O. 114
Environmental Performance Index 142 Christianity 94, 102–106, 107, 115
biospheric values xix, 61, 135 citizenship 101
Blasius, J. 185 climate change xviii, 6, 15, 114; personal
Boeve-de Pauw, J. 189 interests 62, 63; religion impact on
Botetzagias, Iosif 8, 133–154 behaviour 115; risk/threat perceptions
Bourdieu, Pierre 107 17–23, 27–28, 36, 41, 142, 212, 214;
Bozonnet, Jean Paul 8, 91–110 UN Framework Convention on 158;
Brajdić Vuković, M. 160, 161, 165, 167, 169 United States xvii
Index 245
Coefficient of Human Inequality (CHI) 232–234; environmental activism

222, 225, 227 95, 97, 104, 116–117, 119, 121, 124,
cognitive factors 32, 33, 38, 44, 52 126–128; environmental concern 71,
cognitive hierarchy model 61 79, 190, 198; environmental knowledge
cognitive mobilization 100–101, 106 225–230; policies 238; public-sphere
confirmatory factor analysis (CFA) xxi, environmental behaviours 141; risk/threat
66–70, 182, 201n2; see also Multiple perceptions 19–22, 35, 40, 47, 49–51;
Group Confirmatory Factor Analysis salience of environmental issues 118
conservation 63 developed countries: degrowth 159, 160;

context 33–35, 133, 207, 210 environmental concern 62, 63, 66;
core and periphery regions 162–163, ‘green economy’ 158; public-sphere
169–170, 171–173, 175, 176–177 environmental behaviours 149
Croatia: degrowth potential 171–174; developing countries: environmental
democratic institutions 122; concern 57, 59, 60, 63, 66; ‘green
environment and economy dilemmas economy’ 158
232–234; environmental activism Diekmann, A. 65, 138
95–97, 106, 116–117, 119, 121, 124, Dietz, R. 160
126–128; environmental concern 71, Dietz, Thomas xix, 16, 57
79, 190, 199; environmental knowledge Dolenec, D. 160, 161, 169, 172, 175
225–230; policies 238; public-sphere Domazet, Mladen 8, 157–181
environmental behaviours 141; risk/threat donations 8, 95–96, 113, 119, 129, 133–134,
perceptions 19–22, 35, 39, 40, 47, 49–51; 139–150; affluence relationship 138;
salience of environmental issues 118 behavioural spill-over effect 136;
cultural lag 34–35, 49–50 environmental cultures 93; individual-level
culture 8, 91–92, 99, 101, 102, 106 variables 135; socio-demographics 137
Czech Republic: degrowth potential Douglas, Mary 16, 92
171–174; democratic institutions 122; Dunlap, Riley E. xvii–xxiv, 2–3, 4, 56–57,
environment and economy dilemmas 60, 111, 138, 183, 190, 207
232–234; environmental activism Durkheim, E. 105
95–97, 104, 106, 116–117, 119, 121, dynamics 52
124, 126–128; environmental concern
71, 79, 190, 199; environmental Eastern Europe 222, 239; awareness,
knowledge 224, 225–230; policies 238; efficacy, and willingness to pay 79;
public-sphere environmental behaviours environment and economy dilemmas
140; religion 122; risk/threat perceptions 231, 232–234; environmental activism
19–22, 35, 40, 47, 49–51; salience of 96, 129; environmental knowledge
environmental issues 118 224, 225–230; policies 235–238; risk/
threat perceptions 39–40, 51; salience
Dake, K. 15 of environmental issues 118; see also
degrowth 157–177; conceptualizing Central and Eastern-European countries
159–161; environment-development ecological communication 4–5
trade-offs 165–166, 170–175, 176–177; Ecological Footprint (EF) 166, 167, 176
individual concern and sacrifice 163– ecological modernization 3, 14–15, 27,
165, 167–170; research design 162–167 168, 177, 231
Demaria, F. 159 economic crisis 8, 34, 35, 221–222, 231,
democracy 114–115, 160, 222 234, 239, 240n1
Democracy Barometer 115, 121–122 economic growth 34, 38, 59, 223;
Democracy Index 34, 38, 39, 46–48, 52 degrowth potential 157–158, 160–
democratization 13, 14, 28, 114 161, 165–166, 168, 171–174, 175,
demonstrations see protests 176; ecological modernization 168;
Denmark: degrowth potential 171–174; environment and economy dilemmas
environment and economy dilemmas 223, 231–235; modernity 208
246 Index
economic indicators 34, 38, 39, 51 61–62; relative exposure to 63–64; as

economic progress 166, 172–174, 176–177 trigger for environmental concern 81;
Economist Intelligence Unit 34, 38 values 60–61
Ecosystem Vitality 35, 46–48, 142 environmental efficacy 7–8, 65, 72–82;
ecosystems 160, 167 affluence and 161; confirmatory factor
education: awareness, efficacy, and analysis 66–67, 68–69; degrowth
willingness to pay 72, 74; environmental potential 166; environmental activism
activism 94, 100, 101, 103, 108, 113, 136; public-sphere environmental
137; environmental concern 67, 183–184; behaviours 142, 149; risk/threat

need for environmental education 217, perceptions 17, 24–26, 27–28; youth
235–236, 237, 238; public-sphere environmental consciousness 212, 214,
environmental behaviours 146, 149; risk/ 215
threat perceptions 17, 24–26, 28, 37, 43, environmental good and bad 14–15
46; sampling issues 186; United States xvii Environmental Health 35, 46–48, 52
EF see Ecological Footprint Environmental Performance Index (EPI):
efficacy 7–8, 65, 72–82; affluence and political freedom 148; public-sphere
161; confirmatory factor analysis environmental behaviours 142, 144,
66–67, 68–69; degrowth potential 166; 145, 146; risk/threat perceptions 35, 38,
environmental activism 136; public- 39, 46–48
sphere environmental behaviours environmental preservation 57, 58, 61–62
142, 149; risk/threat perceptions 17, environmental utilization 57, 58, 61–62
24–26, 27–28; youth environmental EPI see Environmental Performance Index
consciousness 212, 214, 215 Ester, Peter xviii
egoistic values xix, 61 Eurobarometer 185
employment status 72–73, 74, 80 European Social Survey (ESS) 122, 185
energy consumption 6, 119; degrowth 158; European Union (EU) 6, 17, 20, 23, 27
youth environmental consciousness 212, European Value Study 185
214, 216; see also renewable energy everyday problems 22, 23–26, 27
environmental concern xviii, xix, xxi, 31,
56–88, 135, 207; affluence and xx, 56–57, factorial invariance (FIV) 188–189,
58–60, 73, 80–82, 160–161, 183; Attitudes 192–193, 195–197, 200–201
of Concern and Activation index 163–164, Fairbrother, Malcolm 70, 183
167–170, 173–174, 175; definition of fines 235–236
183; degrowth potential 176; dilemmas Finland: degrowth potential 171–174;
of knowledge and 223–230; environment environment and economy dilemmas
and economy dilemmas 223, 231–235; 232–234; environmental activism
globalization of 151; as a higher-order 95–97, 104, 116–117, 119, 121, 124,
need 208–209; intra-attitudinal dimension 126–128; environmental concern 71,
64–66; local versus global dimension 79, 190, 198; environmental knowledge
58–60; measurement equivalence 225–230; policies 238; public-sphere
182–183, 184, 187–200; objective environmental behaviours 141; risk/
problems-subjective values approach 138, threat perceptions 19–22, 36, 40, 47,
209; risk/threat perceptions 32, 44, 45, 46, 49–51; salience of environmental issues
52; values versus interest 60–64; youth 118
environmental consciousness 210–212, Finseraas, H. 113, 114, 115
214, 215, 216 Fischhoff, B. 15
environmental consciousness, concept of Fishbein, M. 207
205–206; see also youth environmental FIV see factorial invariance
consciousness flooding 78
environmental cultures 93 fossil fuels 14–15, 158
environmental degradation 56–57, 65, 67, France: degrowth potential 171–174;
80; local problems 60; personal interests environment and economy dilemmas
Index 247
232–234; environmental activism Gross, Matthias 1–10

95–97, 104, 116–117, 119–121, 124, gross domestic product (GDP): awareness,
126–129; environmental concern 71, efficacy, and willingness to pay 75,
79, 190, 199; environmental knowledge 80; degrowth potential 157, 159,
225–230; environmental sociology 169; environmental activism 96, 121,
3, 6; policies 238; public-sphere 122; environmental concern 70, 161;
environmental behaviours 140; risk/ political freedom 148; public-sphere
threat perceptions 19–22, 36, 40, 47, environmental behaviours 142–143,
49–51; rural areas 5 144, 145, 146; risk/threat perceptions

Franzen, A. 6, 59, 65, 67, 111, 114, 134, 17, 34, 35, 38, 39, 46–48, 52
138, 160–161, 177n2, 191 gross national income (GNI) 222, 225,
Freymeyer, R.H. 111, 113, 134, 137, 138 226, 234–235, 237, 238
Guagnano, Gregory A. xix
Gallup xix Guerra, João 8, 221–242
GDP see gross domestic product Guha, R. 184
Gelissen, J. 138 Gupta, N. 33
Gemenne, François 157
gender: awareness, efficacy, and Haanpää, Leena 8, 205–220
willingness to pay 72, 74; environmental habit 133
activism 113, 137; environmental habitus 107
behaviours 111; environmental concern Hadler, Markus 6, 7, 13–30, 111–112, 114,
67, 160–161; risk/threat perceptions 17, 118
24–26, 28, 32–33, 37, 42–43, 46, 51–52; Hall, T.D. 163
United States xvii Haller, M. 6, 111–112, 114, 118
genetically modified foods/crops 17–23, Hatch, A. 111, 113
36, 40, 41, 142, 212, 214 HDI see Human Development Index
Germany: degrowth potential 171–174; Health of the Planet Survey (HOP) xx, xix,
environment and economy dilemmas 183
232–234; environmental activism Healy, Seán 221
95–97, 104, 116–117, 119–121, 124, Heberlein, Thomas A. xviii–xix
126–128; environmental concern 71, Henwood, K.L. 32–33
79, 190, 198; environmental knowledge HEP see human exemptionalist paradigm
225–230; environmental sociology 3, 4; hierarchy of needs 96, 208
green model 91; policies 238; public- high risks 22, 23–26, 27–28
sphere environmental behaviours 140; Hoban, T.J. 33, 40
religion 103, 107; risk/threat perceptions Homer, P. 61
19–22, 23–24, 26, 36, 40, 47, 49–51; HOP see Health of the Planet Survey
salience of environmental issues 118 Human Development Index (HDI) 222;
Gijswijt, August 6 degrowth potential 162, 166, 167;
Givens, J.E. 59 environmental knowledge 225, 227; risk/
globalization xx, 60, 111, 151, 183 threat perceptions 34, 35, 38, 39, 46–48, 52
GNI see gross national income human exemptionalist paradigm (HEP)
goodness-of-fit 193, 194, 195–197, 201n4 2–3, 206
government efficacy 136, 142, 144, 145, Hungary: environmental activism 104;
147, 149 environmental concern 71, 79
government intervention 237 Hunter, L.M. 60, 111, 113
Greece 6 hysteresis xxi, 8, 92, 93, 107–108
‘green economy’ 158, 168, 221, 231, 233,
239–240 Iceland: environment and economy
green parties xix, 4, 135, 138 dilemmas 232–234; environmental
Greenspan, I. 135, 137 knowledge 225–230; policies 238; risk/
grid and group theory 16 threat perceptions 17, 19–22
248 Index
III see Inequality-adjusted Income ISSP see International Social Survey

Index Programme
impact-oriented behaviours 93 Italy 79
income: awareness, efficacy, and
willingness to pay 72, 74, 80, 82; Japan: degrowth potential 175;
environmental activism 94, 99, 103, environmental concern 71, 79, 190, 197;
137; environmental concern 67, risk/threat perceptions 19–22
160–161; public-sphere environmental Johnson, A. 111, 113
behaviours 143, 148; see also Johnson, B.E. 111, 113, 134, 137, 138
affluence Jones, Nikoleta 8, 133–154
individualism 100 Jones, R.E. 183, 190
inequalities 13, 14, 16, 159, 161; Jordan, Andrew 235
Coefficient of Human Inequality 222, Jorgenson, A.K. 59
225, 227; degrowth potential 175
Inequality-adjusted Income Index (III) Kahle, L. 61
166–167, 170 Katz-Gerro, T. 209–210
Inglehart, Ronald xxi, 78, 111, 114, Keller, C. 32, 33
205, 207; factorial invariance 189; Keller, R. 28
materialism 59; objective problems- Kemmelmeier, M. 138
subjective values approach 60, 137–138; Kidd, Q. 209
postmodernization 208; values 57, 92, Kim, S. 114
99, 209, 216 Kim, Y.H. 138
institutions 121–122, 123, 125–129, 130, knowledge: dilemmas of 223–230;
224, 229–230, 240 environmental activism 100–101,
intentions 206–207, 208, 210, 211–215, 113–114, 135; risk/threat perceptions
216 33, 38, 44, 45, 46
intent-oriented behaviours 93 Kraemer, Klaus 7, 13–30
International Social Survey Programme Krol, G. 138
(ISSP) xix, xx, 1, 2, 6, 185, 209, Kuznets theory 96
222; country contexts 33; degrowth Kvaløy, B. 113, 114, 115
potential 157, 161, 162–167, 175, 176;
environment and economy dilemmas Lange, Hellmuth 3
231–235; environmental activism 94, Latin America 222; environment and
106, 111, 114, 116–118; environmental economy dilemmas 231, 232–234;
concern 58, 81, 82, 160–161, 183, environmental knowledge 225–230;
190–200; environmental knowledge policies 235–238; see also South
224–230; measurement equivalence America
8, 190–200; membership of pro- Latvia: degrowth potential 171–174;
environmental groups 112; public- environment and economy dilemmas
sphere environmental behaviours 134, 232–234; environmental activism 95,
139; risk/threat perceptions 14, 17–26, 97, 104, 106, 116–117, 119–121, 124,
27, 31–32, 35–37; social capital 101, 126–128; environmental concern 71,
102; youth environmental consciousness 79, 190, 199; environmental knowledge
211–217 225–230; policies 238; public-sphere
Ippel, L. 189 environmental behaviours 141; risk/threat
Ireland: environmental concern perceptions 19–22, 36, 40, 47, 49–51;
79; public-sphere environmental salience of environmental issues 118
behaviours 140 Leahy, Ann 221
Islam 94, 104, 105–106 Lee, A.R. 209
Israel: environmental concern 71, 79, 190, Lichtenstein, S. 15
199, 200; risk/threat perceptions 18, lifestyle: Attitudes of Concern and
19–22 Activation index 164, 167–168;
Index 249
public-sphere environmental behaviours Mostafa, M.M. 138

142, 149; youth environmental multilevel modelling xx, xxi, 57, 66, 70–82
consciousness 217 Multiple Group Confirmatory Factor
Lima, M.L. 33 Analysis (MGCFA) 183, 187–189,
Listhaug, O. 113, 114, 115 191–194, 198–200, 201
Lithuania: degrowth potential 171–174; Murphy, Michelle 221
environment and economy dilemmas Muslims 94, 104, 105–106
232–234; environmental activism 95,
97, 104, 106, 111, 116–117, 119–121, NAM see norm-activation model
124, 126–128; environmental concern national cultures 8, 92, 94, 106, 108
71, 79, 190, 199, 200; environmental natural resources: degrowth 159;
knowledge 224, 225–230; policies 238; Ecosystem Vitality 35; Environmental
public-sphere environmental behaviours Performance Index 142; risk/threat
141; risk/threat perceptions 19–22, 36, perceptions 17–23
40, 47, 49–51; salience of environmental nature 91, 105
issues 118 Nawrotzki, R.J. 67, 138
Luhmann, Niklas 4–5 Nelson, E. 63
NEP see new environmental paradigm
Macias, T. 63 Netherlands: environment and economy
Malesios, Chrysovalantis 8, 133–154 dilemmas 232–234; environmental
Maloney, M.P. 190 concern 71, 79; environmental
marital status 17–18, 24–26 knowledge 225–230; lifestyles and
Marquart-Pyatt, S.T. 65, 66–67, 81, 134, consumption 3; policies 238; public-
183, 190 sphere environmental behaviours 140;
Marshall, B. 82 risk/threat perceptions 17, 18, 19–22
Martinez-Alier, J. 184 new environmental paradigm (NEP) 2–3,
Marxism 35 135, 206
Maslow, A.H. 96, 208 New Zealand: environmental concern 71,
Mayerl, Jochen 8, 182–204 79, 190, 198; risk/threat perceptions 18,
measurement equivalence 8, 182–183, 184, 20–22
187–200 non-governmental organizations (NGOs)
Mediterranean countries 91, 222; see also 114, 135, 136–137
Southern Europe Nordic countries: environment and
membership of pro-environmental economy dilemmas 231; environmental
groups 8, 93, 95, 111–132, 136–137; activism 96, 104; green model 91;
cognitive mobilization 100; public- religion 103; risk/threat perceptions 51;
sphere environmental behaviours 142, see also Scandinavia
144–147, 149, 150 norm-activation model (NAM) xviii
Mertig, A.G. 111, 138, 183 norms xix, 133; hysteresis 107; sampling
methodological problems 185–190 issues 186, 187; social change 34–35;
Mexico: environment and economy trust 102
dilemmas 232–234; environmental North America 222; awareness, efficacy,
concern 71, 79, 190, 199, 200; and willingness to pay 79; environment
environmental knowledge 225–230; and economy dilemmas 231, 232–234;
policies 238; risk/threat perceptions 18, environmental knowledge 225–230;
19–22 policies 235–238; see also Canada;
Meyer, R. 6, 111, 114, 160, 161, 177n2, United States
191 Norway: degrowth potential 171–174;
MGCFA see Multiple Group Confirmatory environment and economy dilemmas
Factor Analysis 232–234; environmental activism
Milfont, T. 189 95, 97, 104, 116–117, 119, 121, 124,
Morkevičius, Vaidas 8, 111–132 126–128; environmental concern 71,
250 Index
79, 190, 198; environmental knowledge political trust 37, 44–45, 46

225–230; policies 238; public-sphere pollution 14–15; Asia 78; Environmental
environmental behaviours 140; risk/threat Performance Index 142; local problems
perceptions 20–22, 36, 40, 47, 49–51; 60; regulation of 67, 80; risk/threat
salience of environmental issues 118 perceptions 17–23, 27, 36, 40–41, 139,
nuclear power/waste: personal interests 63; 142, 212, 214; youth environmental
risk/threat perceptions 13, 15, 17–23, consciousness 213, 214
27–28, 36, 40–41, 142, 212, 214 Portugal: case study 221–242;
environment and economy dilemmas

objective problems-subjective values 231–235; environmental concern 71, 79;
(OPSV) approach 60, 62, 63, 81–82, environmental knowledge 224,
137–138, 209 225–230; environmental sociology
Oceania 78, 79 6; policies 235–239; public-sphere
O’Connor, R. 62 environmental behaviours 141; risk/
O’Neill, D. 160 threat perceptions 17, 20–22
OPSV see objective problems-subjective post-materialism xix–xx, xxi, 111,
values approach 208–209; affluence 59–60;
Oreg, S. 209–210 environmental activism 94, 99–100,
104, 106, 108, 135; environmental
Pampel, F.C. 60, 67 concern 56–58, 61–62, 67, 73, 74, 78,
Parsons, Talcott 4 81–82, 160–161, 183; hierarchical
party identification xvii, 160–161 causal chain model 184; public-sphere
perceived behavioural control 210, 212, environmental behaviours 142, 144,
214, 215 145, 147, 149; risk/threat perceptions 7;
personal interests 57, 58, 61, 62–64, 66, 82 youth environmental consciousness 205,
pesticides: awareness, efficacy, and 211, 214, 215, 216
willingness to pay 73, 75, 80; postmodernity 208, 216
Environmental Performance Index 142; poverty 57, 175, 234
private-sphere environmental behaviours Principal Component Analysis 192
119; risk/threat perceptions 17–23, 36, private-sphere environmental behaviours
41, 139–142, 212, 214 118–120, 123–125, 126, 129–130, 133, 136
petitions 8, 95–96, 113, 119, 129, prosperity hypothesis 8, 111, 160, 177n2;
133–134, 139–150; affluence see also affluence
relationship 138; behavioural spill- protected areas 70, 76, 80
over effect 136; cognitive mobilization Protestantism 102–106, 107, 108
100; environmental awareness and protests 8, 95–96, 119, 129, 133–134,
knowledge 113–114; environmental 139–150; affluence relationship
cultures 93; political structures 139; 138; behavioural spill-over effect
socio-demographics 137 136; cognitive mobilization 100;
Philippines: environmental activism 111; environmental awareness and
environmental concern 71, 79, 190, 199; knowledge 113–114; environmental
human development 162, 167; risk/ cultures 93; individual-level variables
threat perceptions 20–22 135; political structures 139; socio-
Poferl, A. 28 demographics 137
Poland 79 public-sphere environmental behaviours
policies 223, 235–239 118–120, 123–125, 126, 129, 133–151
political culture 92
political freedom 138–139, 142, 146, 148 Qualitative Comparative Analysis (QCA)
political ideology xvii 115–123
political intervention 15
political participation 93, 101, 107, race xvii
138–139; see also socio-political context rainforest depletion 78
Index 251
rational choice xix, 94, 96–99, 100, 101, sampling issues 186–187
102 Scandinavia 3; see also Nordic countries
recycling 6, 119; personal interests 62; Ščasný, M. 189
public-sphere environmental behaviours Schaffrin, André 7–8, 56–88
142, 147; youth environmental Schmidt, Luísa 8, 221–242
consciousness 212, 214, 216 Schmidt-Catran, Alexander 7–8, 56–88
Redclift, Michael 239 Schultz, P.W. 61, 189
reflexive modernization 5 Schwartz, Shalom 92
religion 94, 102–106, 107, 108, 115, science, trust in 33, 37, 44–45, 46
122–123, 126–127 sea-level rises 63
renewable energy 62–63, 65, 70, 158; Seifert, F. 28
awareness, efficacy, and willingness self-interest 57, 99, 106; see also personal
to pay 73, 76, 80; education levels 78; interests
Environmental Performance Index 142 SEM see structural equation
Renn, O. 32 modelling
residence: awareness, efficacy, and semi-peripheral regions 163, 169–170,
willingness to pay 72, 74, 79–80; 171–173, 174–175, 176–177
environmental activism 137; Shue, H. 176
environmental concern 67; public- Sim, Stuart 240
sphere environmental behaviours 144, Slimak, M. 16
145, 148, 149; risk/threat perceptions Slovak Republic: degrowth potential
18, 24–26, 32–33, 37, 42–43, 46 171–174; environment and economy
Reyes, J.A.L. 111 dilemmas 232–234; environmental
risk perception 7, 13–30, 31–55, 207, activism 95–97, 104, 106, 116–117,
210; affluence and EU membership 119, 121, 124, 126–128; environmental
23; CEEs 4; country-level contextual concern 71, 79, 190, 199; environmental
determinants 33–35, 38–39, 45–51, 52; knowledge 224, 225–230; policies 238;
data and methods 17–18; definition of public-sphere environmental behaviours
31; degrowth potential 176; individual 140; risk/threat perceptions 20–22, 36,
differences 15–17, 24–26, 32–33, 40, 47, 49–51; salience of environmental
37–38, 41–44, 51–52; public-sphere issues 118
environmental behaviours 139–142, Slovenia: degrowth potential 171–174;
144, 145, 146, 147, 149, 150; youth environment and economy dilemmas
environmental consciousness 212, 213, 232–234; environmental activism
214, 215, 216 95–97, 104, 106, 116–117, 119, 121,
risk society xxi, 5, 33 124, 126–128; environmental concern
Romano, O. 160 71, 79, 190, 198; environmental
routines 133, 217 knowledge 225–230; policies 238;
Russia: degrowth potential 171–174; public-sphere environmental behaviours
environment and economy dilemmas 140; risk/threat perceptions 20–22, 36,
232–234; environmental activism 40, 47, 49–51; salience of environmental
95–97, 104, 116–117, 119–121, 124, issues 118
126–128; environmental concern 71, Slovic, Paul 15, 22, 31, 41
79, 190, 199, 200; environmental Smelser, N.J. 34
knowledge 224, 225–230; human social capital 101–102, 106, 108
development 162; policies 238; public- social change 34–35
sphere environmental behaviours 141; social class: environmental activism 94,
religion 94; risk/threat perceptions 99; habitus 107; risk/threat perceptions
20–22, 36, 39, 40, 47, 49–51 17, 24–26, 32–33, 37, 42–43, 46, 51–52;
see also socio-economic status
salience of environmental issues 117–118, social context 32, 207, 210
120, 124, 129, 130 social networks 136
252 Index
social trust 37, 44–45, 46; degrowth 221–222, 231, 233, 235, 239–240
potential 177; environmental activism Sweden: degrowth potential 171–174;
94, 101–102, 104, 108, 136; public- democratic institutions 122; environment
sphere environmental behaviours 142, and economy dilemmas 232–234;
143, 149 environmental activism 95, 98, 104,
socialization 92, 99–100, 101, 102, 105, 116–117, 119, 121, 124, 126–128;
107, 108 environmental concern 71, 79, 190, 198;
societal transformation 34–35, 38–39, environmental knowledge 225–230;
48–51, 52, 158, 159 policies 238; public-sphere environmental

socio-demographics: degrowth potential behaviours 140; risk/threat perceptions
177; environmental activism 113, 20–22, 36, 40, 47, 49–51; salience of
137; environmental concern 160–161; environmental issues 118; trust 102
public-sphere environmental behaviours Switzerland: degrowth potential
133; risk/threat perceptions 16, 17, 28, 171–174; democratic institutions 122;
32–33, 37, 41–44 environment and economy dilemmas
socio-economic status 16, 183; see also 232–234; environmental activism
social class 95–96, 98, 104, 106, 108, 116–117,
socio-political context 34, 39, 48–51, 52 119–121, 126–128; environmental
South Africa: degrowth potential 168; concern 71, 79, 190, 198; environmental
environmental concern 71, 79, 190, 199; knowledge 225–230; policies 238;
human development 162, 167; risk/ public-sphere environmental behaviours
threat perceptions 18, 20–22, 23, 25, 141; religion 107; risk/threat perceptions
26, 27 20–22, 36, 40, 47, 49–51; salience of
South America 78, 79, 222; see also Latin environmental issues 118
America
South Korea: degrowth potential 175; Taiwan: environmental concern 79, 190,
environmental concern 71, 79, 190, 199, 191, 197; risk/threat perceptions 20–22
200; risk/threat perceptions 19–22 tax breaks 235–236
Southern Europe 222, 239; awareness, technology: degrowth potential 168; risk/
efficacy, and willingness to pay 79; threat perceptions 43
environment and economy dilemmas Telešienė, Audronė 1–10, 31–55, 111
232–234; environmental activism Theory of Planned Behaviour (TPB) xix,
96, 129; environmental knowledge 205, 207–208, 209, 216, 217
225–230; policies 235–238; risk/threat Thiessen, V. 185
perceptions 51 threat perceptions see risk perception
Spain: degrowth potential 171–174; tourism 63, 70
environment and economy dilemmas TPB see Theory of Planned Behaviour
232–234; environmental activism trust 33, 224, 230; degrowth potential 177;
95–96, 97, 104, 116–117, 119, 121, 124, environmental activism 94, 101–102,
126–128, 129; environmental concern 104, 108, 136; environmental concern
71, 79, 190, 198, 200; environmental 160–161; Portugal 239, 240; public-
knowledge 225–230; environmental sphere environmental behaviours 142,
sociology 6; policies 238; public-sphere 143, 149; risk/threat perceptions 32, 37,
environmental behaviours 141; risk/threat 44–45, 46, 52
perceptions 20–22, 36, 40, 47, 49–51; Turkey: environmental activism 95–96, 98,
salience of environmental issues 118 104, 107; environmental concern 71, 79,
spill-over effect 135–136 190, 199, 200; religion 94; risk/threat
Steinberger, J.K. 161 perceptions 20–22
Stern, Paul C. xix, 16, 113, 134, 135
structural equation modelling (SEM) 213, 215 UNDP see United Nations Development
survey design 185–186 Programme
sustainable development 4, 8, 159, unemployment 72–73, 74, 80, 222
Index 253
United Kingdom: degrowth potential Wallerstein, I. 162, 163

171–174; environment and economy Wang, J. 114
dilemmas 232–234; environmental Ward, M.P. 190
activism 95, 98, 104, 116–117, 119, waste disposal 17–23,
121, 124, 126–128; environmental 159, 160
concern 71, 79, 190, 191, 199, 200; water 23–26, 27; drinking 35, 67,
environmental knowledge 225–230; 73, 75, 80; Ecosystem Vitality 35;
lifestyles and consumption 3; policies environmental behaviour 119; water
238; public-sphere environmental pollution 17–23, 36, 41, 142, 212, 214;
behaviours 140; religion 103; risk/threat water shortages 17–23, 28
perceptions 20–22, 36, 40, 47, 49–51; wealth see affluence
salience of environmental issues 118 Weber, Max 91, 99
United Nations Development Programme Western Europe 222; awareness,
(UNDP) 167 efficacy, and willingness to pay 79;
United States xvii, xviii, 2–3; degrowth environment and economy dilemmas
potential 168, 175; environment 231, 232–234; environmental activism
and economy dilemmas 232–234; 96, 129; environmental knowledge
environmental concern 71, 79, 190, 225–230; policies 235–238; risk/
199, 200; environmental knowledge threat perceptions 51; salience of
225–230; policies 238; risk/threat environmental issues 118
perceptions 20–22 willingness to pay 7–8, 65–66, 72–82,
Urban, D. 33, 40 115; affluence and 138, 161; Attitudes
Urban, J. 189 of Material Sacrifice index 164, 165,
167–170, 173–174, 175; confirmatory
Valente, Susana 8, 221–242 factor analysis 66–67, 69; degrowth
validity 8, 193, 194–196, 198, 199–200 potential 176; environmental
Value-Belief-Norm (VBN) theory xix, 16 taxes 133; membership of pro-
values xix, 133, 206; cognitive mobilization environmental groups 136–137; youth
101; cultural 92, 209, 216; environmental environmental consciousness 210,
concern 60–64, 65, 66; objective 212, 213, 215, 216
problems-subjective values approach women: awareness, efficacy,
60, 138; post-materialist xix–xx, 56–57, and willingness to pay 72, 74;
59–60, 81–82, 99–100, 184, 208–209, environmental activism 113, 137;
210, 211, 215; Protestant 105; public- risk/threat perceptions 24–26, 42, 43,
sphere environmental behaviours 135; 51–52; see also gender
religious 107, 108; risk/threat perceptions World Values Survey (WVS) xix, xx, 183,
16, 28; sampling issues 186, 187; social 185
change 34–35; social context 207 world-system theory 162–163
Van Griethuysen, Pascal 221
Van Liere, Kent D. 2–3 Xiao, C. 207
Van Schuur, W. 135, 138
VBN see Value-Belief-Norm theory York, R. 56–57, 60
Vogl, D. 59, 67, 134, 160–161, 191 youth environmental consciousness
vulnerability 35, 52–53, 62 205–220
Yuchtman-Ya’ar, E. 56–57
Wachinger, G. 31
Wakefield, S. 135, 137 Zelezny, L. 61

Green European 2016

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Green European 2016

Uploaded by

Copyright:

Available Formats

Downloaded by [University of California, San Diego] at 18:58 11 February 2017

Empirical, and methodologically sophisticated, this excellent volume is the first

This Page is Intentionally Left Blank

Green European addresses the quest for a better understanding of European

•• The first addresses perceptions of environmental and technological threats

Audronė Telešienė is Associate Professor of Sociology at the Institute of Public

Matthias Gross is Professor of Environmental Sociology at the Helmholtz

13 Internationalisation of Social 18 Religion, Identity and

Environmental behaviour and

Edited by Audronė Telešienė

The right of the editors to be identified as the authors of the

ISBN: 978-1-138-12395-3 (hbk)

Typeset in Times New Roman

List of common abbreviations xxvii

Introduction: how green are Green Europeans? 1

1 The perception of environmental threats in a global and

2 Vulnerable and insecure? Environmental and technological

3 Changing concern about environmental threats and risks:

4 Explaining environmental activism by national cultures:

5 Explaining membership in pro-environmental groups:

6 Petitioner, contributor, protester: the profile of Europeans

7 How far for the money? Affluence and democratic

8 Environmental concern in cross-national comparison:

9 Youth environmental consciousness in Europe:

10 Dilemmas of sustainability in turbulent times:

1.1 Dimensions 1 and 2 of environmental threat perception and

5.2 Thresholds of fuzzification for outcome variable: proportion

public involvement in pro-environmental groups:

10.7 Importance of environment concern and other priorities

10.10 Country needs economic growth to protect environment

1.1 Perception of environmental threats across countries (ISSP 2010) 19

preserve or protect environment): impact of pro-environmental

preserve or protect environment): impact of societal institutions 127

Scientific Research) – Sciences Po, Grenoble University, France. His main

Controverses climatiques, sciences et politique (2012), and also published as

and experimental practices in society. Together with Linsey McGoey he has

for Science and Technology (FCT), carrying out research on participation

Statistical Research Analysis and Documentation (ISRAD), Athens University

the Department of Social Sciences, University of Kaiserslautern, Germany. He

Alexander Schmidt-Catran is a research assistant at the Institute of Sociology

the Institute of Public Policy and Administration, Kaunas University of

In the USA environmental issues began to receive increasing attention in the

Studies of public opinion toward environmental issues have continued to be

the cross-national level (Nawrotzki and Pampel 2014). As anthropogenic climate

to explain environmental behaviours (Groot and Steg 2008). Second, a series

between proponents of various versions of an ‘affluence hypothesis’, that sees

individuals’ levels of environmental concern, given the great variation in envi-

In sum, sociological work on environmental concern has come a long way

Brechin, Steven R. and Willet Kempton. 1994. ‘Global Environmentalism: A Challenge to

Heberlein, Thomas A. 1977. ‘Norm Activation and Environmental Action: A Rejoinder to

Environment and Behavior 38:462–483.

ISSP: International Social Survey Programme (www.issp.org)

European environmental behaviour

•• What are the multiple cultural contexts underlying environmental world

Environmental sociology in Europe

so-called human exemptionalist paradigm (HEP) to a new environmental para-

sociologically important in Luhmann’s view. In the case of the environment this

complemented by the more generalized ones of industrial risks, sustainable con-

ized by a deliberate proximity to environmental groups (Lange 2011), and by

The contributions to this book

more conceptually diverse topics related to environmental world views, behav-

threats and risks

This Page is Intentionally Left Blank

The perception of environmental

List of common abbreviations xxvii

Introduction: how green are Green Europeans? 1

1.1 Dimensions 1 and 2 of environmental threat perception and

5.2 Thresholds of fuzzification for outcome variable: proportion

10.7 Importance of environment concern and other priorities

10.10 Country needs economic growth to protect environment

1.1 Perception of environmental threats across countries (ISSP 2010) 19

preserve or protect environment): impact of societal institutions 127

Figure 1.1 Dimensions 1 and 2 of environmental threat perception and position of

Figure 1.2 Dimensions 1 and 3 of environmental threat perception and position of

Table 1.2 Correlations among societal affluence, EU membership and threat perceptions*

Figure 2.3 Risk perception across socio-demographic characteristics. Mean scores