Human Dimensions of Wildlife

An International Journal

ISSN: 1087-1209 (Print) 1533-158X (Online) Journal homepage: http://www.tandfonline.com/loi/uhdw20

Informing Community Engagement in Sea Turtle

Conservation by Examining Non-Conservation-
Related Participation in Northeast Brazil

Marisa A. Rinkus, Dan Kramer & Tracy Dobson

To cite this article: Marisa A. Rinkus, Dan Kramer & Tracy Dobson (2015): Informing
Community Engagement in Sea Turtle Conservation by Examining Non-Conservation-
Related Participation in Northeast Brazil, Human Dimensions of Wildlife, DOI:
10.1080/10871209.2016.1098752

To link to this article: http://dx.doi.org/10.1080/10871209.2016.1098752

Published online: 11 Dec 2015.

Submit your article to this journal

Article views: 2

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=uhdw20

Download by: [Michigan State University] Date: 18 December 2015, At: 08:10
 HUMAN DIMENSIONS OF WILDLIFE
http://dx.doi.org/10.1080/10871209.2016.1098752

RESEARCH ARTICLE

Informing Community Engagement in Sea Turtle Conservation

by Examining Non-Conservation-Related Participation in
Northeast Brazil
Marisa A. Rinkusa, Dan Kramera,b, and Tracy Dobsona
a
Department of Fisheries and Wildlife, Michigan State University, East Lansing, Michigan, USA; bJames
Madison College, Michigan State University, East Lansing, Michigan, USA
Downloaded by [Michigan State University] at 08:10 18 December 2015

ABSTRACT KEYWORDS
Participation in conservation programs is often viewed as a necessity, Collective action;
but understanding of this participation is limited. This research conservation and
takes a dierent approach by also examining participation in non- development; participation;
perceptions; social capital
conservation-related activities. Questionnaires (N = 339) were admi-
nistered in four Brazilian coastal communities where the national sea
turtle conservation project conducts monitoring and outreach.
Employing social capital as a conceptual framework, measurements
included participation in civic, community, and social activities along
with group participation and collective action. Conrmatory factor
analysis was used for validating measures of social capital and con-
servation perceptions. Structural equation modeling indicated that
social capital has a positive relationship with participation in sea
turtle conservation, suggesting that general levels of participation
and community norms of participation could be just as inuential
as conservation perceptions to engage communities in conservation.
Information gained by deconstructing participation may be used to
design and target conservation programs, improving overall
participation.

Introduction
Community engagement remains an important aspect of many conservation programs,
often serving as a vehicle for education, policy formation and implementation, legal
enforcement, and scientic monitoring. Indeed, participation has come to be synonymous
with anything that involves people (Cornwall, 2011). Although the literature oers insight
for co-management situations (Blaikie, 2006; Brosius, Tsing Lowenhaupt, & Zerner, 2005;
Ostrom et al., 2002), many conservation issues do not lend themselves to a co-manage-
ment arrangement where decision-making will be decentralized and shared among local
institutions and actors. This is often the case for endangered species conservation where
planning and decision-making is made at national or international levels (Campbell &
Godfrey, 2002).
Conservation eorts rely heavily on programs that aim to change individual con-
servation attitudes to increase participation either formally in decision-making pro-
cesses or informally through behavior change. However, the reliance on changing

CONTACT Marisa A. Rinkus rinkusma@msu.edu Department of Fisheries and Wildlife, Michigan State University,
480 Wilson Road, Room 13 Natural Resources Bldg., East Lansing, MI 48824, USA.
2015 Taylor & Francis
 2 M. A. RINKUS ET AL.

attitudes through awareness and education is not sucient, and more attention to how
attitudes and other social factors may inuence a behavior, such as participation, is
needed (Heberlein, 2012). Social capital (commonly dened as networks, norms, and
trust) oers a framework for understanding participation at the community level by
examining individual involvement in various aspects of community life, as well as the
norms, attitudes, and beliefs toward collective action through community participation
(Coleman, 1988; Putnam, 1993, 2000). According to Putnam (2000), a lack of participa-
tion in society through social networks and community organizations limits social
interaction that can encourage engagement. Understanding individual and community
level social capital may oer insight on the social factors that inuence the decision to
participate in conservation, aside from conservation attitudes. This study focuses on sea
turtle conservation.
Downloaded by [Michigan State University] at 08:10 18 December 2015

Participation
Participation, or the act of taking part in something whether it be conservation related
or otherwise, can be categorized in several dierent ways based on who is involved
(stakeholders or general public), how they are involved (education, consultation,
planning, monitoring, decision-making, research, management, enforcement), and
why they are involved (normative or pragmatic). Across disciplines, Arnsteins
(1969) Ladder of Citizen Participation has provided the backbone for the bulk of the
literature on participation and has led to this concept being viewed on a continuum
from passive education imparted by an expert authority to the complete devolution of
decision-making power to communities (Chase, Schusler, & Decker, 2000). In the
literature on conservation, the sharing of power and decision-making over a resource
(e.g., sheries, forests) is often referred to as the co-management of natural resources
(Berkes, 2009). Although community-based natural resource management and
community-based conservation are two other common forms of participation, these
types of participation are not applicable within this study and therefore outside of the
scope of this article.
Pretty (1995) presented six forms of participation: passive, consultative, bought, func-
tional, interactive, and self-mobilization. These ranged from pragmatic to normative
based on how people were involved, and addressed some of the rationale for participation
on the side of both participants and the implementing agency. In this study, the three
main types of participation employed in sea turtle conservation are passive, consultative,
and bought. Passive participation can include education or outreach activities where
individuals are provided information or informed of a recent decision or regulation
regarding species protection. Consultative is similar, but can also involve the implement-
ing agency asking questions and acquiring information from participants, as well as
informing. Bought participation, which also involves passive and consultative forms of
participation, occurs when individuals participate in return for food, cash, or other
material goods. In Brazil, the focus of this study, this is often referred to as t-shirt
conservation and is quite common in sea turtle conservation. This study examines
participation in the community, civic, and social aspects of community life, or what is
referred to here as non-conservation-related participation, through the lens of social
capital to understand how it may inuence participation in conservation.
 HUMAN DIMENSIONS OF WILDLIFE 3

Social capital

Social capital is commonly dened as networks, norms of reciprocity, and relationships of

trust that facilitate collective action (Coleman, 1988; Putnam, 1993, 2000). These three
concepts can be organized into two dimensions, structural (e.g., families, social networks,
membership in voluntary associations, churches) and cognitive (e.g., shared norms, values,
attitudes, trust, reciprocity; Grootaert & Van Bastelear, 2001). The structural dimension
represents the formal and informal networks that can develop through membership in a
community group or participating in social, civic, or community activities. Participation at
the community level in political activism (civic participation), community service (com-
munity participation), or even socializing with friends and family (social participation)
can form and strengthen social relationships that are necessary for building trust and
exposing individuals to new ideas. Therefore, the structural and cognitive aspects repre-
Downloaded by [Michigan State University] at 08:10 18 December 2015

sent a reciprocal relationship where the shared norms, attitudes, and values (cognitive
aspects) aect participation in social networks and community organizations (structural
aspects) that can build or reinforce trust and reciprocity.
A fundamental tenet of social capital is that group involvement can positively aect
individuals, communities, and resources by facilitating access to benets and the dissemina-
tion of information (Portes, 1998; Pretty & Smith, 2004). Disciplines outside conservation,
such as public health and development, have examined the inuence of general measures of
social capital at the individual and community levels. For example, social capital has been
shown to positively inuence health outcomes and behaviors (Berkman, Glass, Brissette, &
Seeman, 2000; Hawe & Shiell, 2000). Brune and Bossert (2009) found that participation in
community groups and social networks improved individual health behaviors and favorable
attitudes toward collective action and cooperation, as well as being associated with increased
involvement in community health projects. In the development literature, studies con-
ducted by the World Bank have indicated a positive relationship between social capital
and various development outcomes including watershed conservation (Krishna & Upho,
1999), voluntary solid waste management systems (Pargal, Mainul, & Gilligan, 1999), and
rural irrigation and urban sanitation (Khknen, 1999). These studies established further
the roles of networks, norms of reciprocity, and trust in facilitating cooperation and the
spread of information. Following this approach, this study illuminates further how social
capital measured at the community level (e.g., participation in existing non-conservation
organizations) may relate to participation in conservation.
In the conservation literature, social capital has often been used in the study of
participation in conservation (Ballet, Sirven, & Requiers-Desjardins, 2007; Wagner &
Fernandez-Gimenez, 2009). However, the majority of these studies focused only on social
capital measured within conservation-focused co-management groups (Cramb, 2005;
Sekhar, 2007). Macias and Nelson (2011) presented one exception examining the link
between social capital and environmental concern by measuring the structural aspects of
community social capital using a network-based approach. Social capital has also been
shown to be a determinant of environmental or conservation policy formation and
implementation (Jones, 2010; Jones, Halvadakis, & Sophoulis, 2011), attitudes (Thuy,
Dwivedi, Rossi, Alavalapati, & Thapa, 2011), performance/quality (Dulal, Foa, &
Knowles, 2011; Kramer, 2007), behavior (Jin, 2013), and activism (Marquart-Pyatt,
2012). Social capital may also be inuential in shaping conservation attitudes through
 4 M. A. RINKUS ET AL.

the exchange of information, lowering of transaction costs, and reinforcing social bonds
and norms (Pretty & Smith, 2004). Social capital can also have a negative eect on natural
resources, such as facilitating the subversion of resource management restrictions
(Lansing, 2009) or have no eect at all (Duit, Hall, Mikusinski, & Angelstam, 2009;
Grafton & Knowles, 2004).
Despite the variety of studies related to conservation and social capital, limited research
has examined social capital in relation to wildlife conservation by measuring the structural
and cognitive aspects of non-conservationrelated participation. This article posits that
this omission overlooks the signicant inuence of non-conservation-related social life, as
demonstrated by social capital analysis, on participation in conservation. This study
measures non-conservation participation as represented by the cognitive and structural
dimensions of social capital, and explores relationships among social capital, conservation
perceptions, and conservation participation. The conceptual model used in this study
Downloaded by [Michigan State University] at 08:10 18 December 2015

(Figure 1) was developed based on the World Banks Social Capital Assessment Tool
(SOCAT; Krishna & Shrader, 1999) where the structural components of social capital are
represented as group membership and social, civic, and community participation; and the
cognitive components as collective action and social trust. Conservation perceptions are a
separate measure in the model aside from the higher order factor for social capital.
Although this conceptual model of social capital can be viewed as similar to Ajzen and
Fishbeins (1980) theory of reasoned action (TRA) and theory of planned behavior (TPB)
(Ajzen, 1985), there are two main dierences. First, TRA and TPB place emphasis on the
individual, focusing on the inuence of individual beliefs on individual action, whereas
social capital addresses the relationship of the individual to society examining the reci-
procal relationships between general attitudes, norms, and beliefs regarding collective
action and actual participation (past and present) at both the individual and community

Figure 1. Conceptual model of the relationships among social capital, sea turtle conservation percep-
tions, and participation in sea turtle conservation.
Note. Ovals represent latent concepts and squares represent observed variables.
 HUMAN DIMENSIONS OF WILDLIFE 5

levels. Second, social capital explicitly addresses the role of social relationships in facil-
itating participation.
The purpose of this study is to enhance understanding of participation in conservation
by examining the social factors that inuence participation at the community level. The
hypothesis is that community level non-conservation participation (i.e., social, civic,
community) along with norms and attitudes that predispose people toward collective
action are as inuential as conservation attitudes on participation in sea turtle
conservation.

Methods
Study area
Downloaded by [Michigan State University] at 08:10 18 December 2015

This study was conducted in four coastal towns located in the state of Bahia (BA) on
Brazils northeastern coast: Stio do Conde, Poas, Siribinha, and Barra do Itarir. Stio do
Conde is the largest of the four communities with ~2,500 residents, and Siribinha the
smallest with ~500 year-round residents (Instituto Brasileiro de Geograa e Estatstica
[IBGE], 2010). Located within the municipality of Conde (population ~23,600), these
communities are characterized by a historical dependence on coconut and sugar planta-
tions, as well as shing, and more recently an inux of seasonal and weekend beach
tourism. According to the Colnia de Pescadores Z-31 (shers association) in Stio do
Conde, there are approximately 450 commercial shers from the surrounding commu-
nities, but this does not include those who sh for recreation or in times of need to feed
their families.
The national Brazilian sea turtle conservation program, Projeto TAMAR (hereafter
referred to as TAMAR), manages 22 research stations along the entire coast of Brazil, a
few visitor centers, merchandise stores, and a foundation that manages eorts related to
community engagement. Since its inception in 1980, TAMAR has released more than 15
million turtle hatchlings into the ocean or approximately one million per year with
4050% hatching in the state of Bahia (Fundao Centro Brasileiro de Proteo e
Pesquisa das Tartarugas Marinhas [Fundao Pr TAMAR], 2010, 2011). TAMAR attrib-
uted this success to increased monitoring, support from communities, and heightened
awareness of sea turtle conservation by shers and beachgoers (Marcovaldi, Patiri, &
Thom, 2005). Established in October 1991, the TAMAR research station in Stio do
Conde monitors 85 km of coastline running north and south, encompassing the four
communities in this study where TAMAR concentrates the majority of conservation
eorts. This stretch of coastline contains prime nesting beaches for the olive ridley
(lepidochelys olivacea), hawksbill (eretmochelys imbricata), and loggerhead turtle species
(caretta caretta). Seasonally, TAMAR employs one person in each of the communities of
Barra do Itarir, Poas, and Siribinha for monitoring, and two people in Stio do Conde to
conduct year-round monitoring for the beaches.
At the community level, TAMAR engages in sensibilizao (awareness raising) and
environmental education, presentations and expositions, income generation, preservation,
and valorizao (recovery) of local culture, and nancial or other support to community
associations or activities. From 2008 to 2012 in Stio do Conde, sta conducted or
organized 36 school presentations for 2,511 primary and secondary local school children
 6 M. A. RINKUS ET AL.

and more than 10,000 people participated in hatchling or rehabilitated sea turtle releases
on the beach. These releases occur primarily throughout the reproductive season, which is
also the tourist season, meaning that participants in this type of activity are more likely to
be tourists than local residents. The TAMAR sea turtle conservation project does not fall
under the category of co-management where community members are involved in con-
servation decision-making; instead it focuses community engagement around increasing
awareness of sea turtle conservation.

Data collection
Data collection in this study used a sequential mixed-methods approach involving semi-
structured interviews and household questionnaires (Creswell & Plano, 2010).
Information from the semi-structured interviews served to inform the sampling plan
Downloaded by [Michigan State University] at 08:10 18 December 2015

and content of the questionnaire. Interview schedules and household questionnaires

were adapted from the World Bank SOCAT (Krishna & Shrader, 1999). Interviews
included the following thematic sections: community characteristics, community govern-
ance and decision-making, community institutions, and community involvement in sea
turtle conservation. Fifteen semi-structured interviews were conducted across the four
communities until saturation was reached and no new information was found, as evi-
denced by identical or similar responses to open- and closed-ended questions (Rubin &
Rubin, 2005). Interview participants were selected to represent various community groups,
including: local TAMAR sta, religious groups, leaders of local shing associations, the
health clinic, a non-prot youth organization, and other individuals identied by respon-
dents as highly involved in the community through snowball sampling.
To determine the sampling frame for the questionnaire, information was collected from
local public health ocials on the number of families within each community (Stio do
Conde: 678 families, Poas: 117, Siribinha: 95, Barra do Itariri: 294) and an informal count of
houses. Maps of each community were drawn with estimated counts of the number of houses
per street (both seasonal and permanent). In the larger communities of Stio do Conde and
Barra do Itariri, a stratied sampling frame based on locally identied neighborhoods was
devised with 20% of houses (1 in every 5 houses) per street within each neighborhood being
approached. For the smaller communities of Poas and Siribinha, every house was
approached except for unoccupied seasonal dwellings. To ensure gender and age diversity,
any person 18 years of older found to be at home and consenting to take part in the study was
asked to complete the questionnaire. Days and times were also varied, taking into account
varying working schedules and culturally appropriate times for visitors. Questionnaires were
administered orally in Portuguese in or near the respondents home with the average time for
completion being approximately 40 minutes. Questionnaires were completed over a 2-month
time period between August 1 and September 30, 2011. All questionnaires were administered
by one of three local research assistants or the rst author.

Sample of respondents
In total, 341 questionnaires were administered across Stio do Conde (n = 174), Poas
(n = 57), Siribinha (n = 39), and Barra do Itariri (n = 71). Two incomplete questionnaires
were removed leaving a total sample size of n = 339. Of these respondents, 197 were
 HUMAN DIMENSIONS OF WILDLIFE 7

women and 143 were men. Income and education level was low among all communities
with 61% of the overall sample indicating a household monthly income of two minimum
salaries or less (approximately R$1080 or US$540) and 57% having only completed some
primary schooling. Although comparable to state averages, the northeast region of Brazil
is considered to be poorest and literacy rates the lowest in the country. Demographic
characteristics of the sample are representative of the population according to the 2010
census for the municipality of Conde (IBGE, 2010).

Measures
The questionnaire consisted of ve sections: (a) demographic information about house-
hold members; (b) housing characteristics; (c) perceptions of the environment and sea
Downloaded by [Michigan State University] at 08:10 18 December 2015

turtle conservation; (d) group membership in community organizations, associational

networks, (civic, community, and social participation) and collective action; and (e)
community trust and cooperation. Group membership and associational networks repre-
sent the structural dimension of social capital, whereas collective action and social trust
represent the cognitive dimension. Each section was constructed of a mix of open- and
close-ended questions, including scale response and rating type questions. All research
instruments were pre-tested in the eld and revised as appropriate. The measures were
selected based on their reliability and validity in international contexts (Krishna &
Shrader, 1999; Narayan & Cassidy, 2001).

Associational networks (civic, community, social participation)

This measure was developed following the SOCAT household questionnaire (Krishna
& Shrader, 1999), using denitions and categorizations of social, civic, and community
participation based on Hodgkin (2008). Participants were asked how often they had
participated in various civic, community, and social activities over the past 3 years.
Civic participation included those activities that related to political or community
activism, either on an individual or collective basis (six items). Community participa-
tion was a mix of civic and social activities such as volunteering for a local charity,
service club, or school group where group participation was not required or implied (6
items). For example, donating money or materials to a local charity can be an activity
regarded as community participation that does not require ongoing participation or
membership. Social participation included activities such as visiting with family and
friends, or attending a social event in a public space (4 items).

Group membership
Belonging to a community group or organization was measured by asking respondents to
list up to three groups they consider themselves members of and in which they regularly
participate. A dichotomous variable was then created to indicate participation or no
participation in a group. The number of people who participated in more than one
group was too small to consider as a separate category.
 8 M. A. RINKUS ET AL.

Trust
Ten items measuring generalized social trust (e.g., Most people here are basically honest
and can be trusted) and cooperation (e.g., If you have a problem, there is always someone to
help) using a ve-point scale were adapted from the SOCAT household questionnaire
(Krishna & Shrader, 1999).

Collective action
Collective action was measured using 6 items, a combination of 2 items on a 4-point scale
and 4 items on a 5-point scale. These items assessed respondent participation in collective
action (e.g., In the last year, how many times did you unite with other people from this town
in order to resolve a problem?), perception of others participation in collective action,
Downloaded by [Michigan State University] at 08:10 18 December 2015

perception of collective spirit among community members, perception of inuence in

community-level decision-making, and leadership involvement in decision-making (e.g.,
The leaders here bring people together to discuss issues before making any decisions about
the progress of the town).

Participation in sea turtle conservation

Participation in sea turtle conservation was measured through a binary variable where
respondents were asked whether or not they had ever participated in sea turtle conserva-
tion through a TAMAR sponsored activity (example activities as dened by the respon-
dents and TAMAR included: attending a release of hatchlings, educational presentation,
notifying TAMAR of a nest or sick animal, visiting the program oce, employment, and
other). A dichotomous dummy variable (0 = has not participated, 1 = has participated)
was used for analysis purposes.

Perceptions of sea turtle conservation and TAMAR

Ten statements (ve-point scale from strongly disagree to strongly agree) were created to
evaluate respondent perceptions of sea turtle conservation, TAMAR, and nature conserva-
tion in general. General questions regarding sea turtle conservation focused on perceived
importance (e.g., The conservation of sea turtles is important for your family) and desire to
learn more or participate in sea turtle conservation (e.g., You would like to participate in
sea turtle conservation or know more about conservation in general). Questions were also
based upon TAMARs educational messaging regarding sea turtle conservation (e.g., Upon
nding a sick animal on the beach you would notify TAMAR) and TAMARs involvement
with the community (e.g., TAMAR should talk with people here before making decisions
about sea turtle conservation).

Demographics
Socioeconomic information was collected for each household member, including com-
munity of residence, income, age, gender (coded as woman = 1 and man = 0), marital
status, education, profession, and whether or not they received any government benets.
 HUMAN DIMENSIONS OF WILDLIFE 9

Household income was calculated based on reported incomes for each individual house-
hold member recorded. Income and education were fairly homogeneous, so only gender,
age, and community of residence were tested as covariates in the model.

Results
Data analysis was conducted in a 2-step process analyzing the measurement models rst
and then the structural model. Given the focus of this article is the structural model,
discussion of the measurement models is limited, but additional information is available
on request.

Measurement models
Downloaded by [Michigan State University] at 08:10 18 December 2015

As in other social capital studies, factor analysis was used to create latent variables
representing the components of social capital as dened above (Jones, 2010; Mitchell &
Bossert, 2007; Narayan & Cassidy, 2001), in addition to a measure representing sea turtle
conservation perceptions. Conrmatory factor analysis (CFA) allowed for observed items
to be tested as associated with each factor, ultimately reducing the number of variables to
factors representing unobserved or latent constructs. Given the use of these scales in
previous research (with the exception of sea turtle conservation perceptions), CFA was
used instead of exploratory factor analysis (EFA) to test previous theory related to the
dimensions of social capital, not generate theory as is common with EFA. In addition, the
use of CFA as the rst step in structural equation modeling (SEM) allows the researcher to
use the latent factors directly in analysis without computing factor scores and allows for
more modeling exibility for the inclusion of additional variables (Brown, 2006). In some
cases, the measurement models were modied by removing items based on theoretical and
methodological issues, including model improvement. For example, some items were
weakly correlated and exhibited weak factor loadings. According to Brown (2006) factor
loadings greater than .3 or .4 are considered salient for applied research.
Of the social capital components, social participation was not found to be a salient
latent variable due to correlation between indicators leading us to use one observed
variable for social participation in the higher order factor for social capital. Negatively
or similarly worded items among the latent factors appeared to be inuenced by method
eects based on highly correlated measurement errors, as was evident in the latent factor
for trust where 7 of the 10 items were removed. An EFA was run separately for the trust
items, but did not reveal multiple dimensions. A second order factor for social capital was
constructed using the following latent variables: community participation, civic participa-
tion, collective action, trust, and observed variables for social participation and group
participation. Trust was not signicant in the second order factor for social capital and
was left as a separate latent variable.
The latent factor for sea turtle conservation perceptions was separated into two factors,
one for beliefs and one for behaviors with 5 of the total 10 indicators being removed due
to poor factor loadings and model t. However, only one factor representing sea turtle
conservation perceptions was used in the SEM given the high residual correlation of one
item in the second factor to all other items in both factors. Although the 2 factors were
highly correlated with each other, a combined factor of all the items did not meet the t
 10 M. A. RINKUS ET AL.

Table 1. Standardized parameter estimates for each observed variable.

Std. factor S. t- p-
Latent variable Observed variable loading E. value value
Civic participation In the last three years, how often have you:1
Actively participated in an informational campaign? .73 .05 15.78 <.001
Participated in an election campaign? .51 .06 8.51 <.001
Participated in a protest or strike? .50 .07 7.18 <.001
Spoken with the mayor or town representative to the mayors .52 .06 9.50 <.001
oce?
Notied the police or courts about a problem in town? .60 .06 11.07 <.001
Participated in a community development meeting? .73 .04 16.55 <.001
Community In the last three years, how often have you:1
participation Actively participated in an association or group/organization? .57 .05 11.83 <.001
Spoken with an inuential person? .75 .04 18.17 <.001
Participated in a workshop? .50 .06 8.39 <.001
Spoken with others about a problem in town? .62 .05 12.34 <.001
Donated money or materials to an association or group? .50 .06 8.98 <.001
Volunteered for a charity organization? .60 .06 10.63 <.001
Downloaded by [Michigan State University] at 08:10 18 December 2015

Social In the last three years, how often have you:1

participation Visited family or friends outside of the house? .49 .04 12.21 <.001
Participated in a sporting event?2
Attended a party?2
Entertained family or friends at home?2
Group Do you participate in a community group or organization? .41 .07 5.79 <.001
participation
Collective action In the last year, how many times did people here organize to .62 .07 8.42 <.001
register a complaint or make a request of the mayor or the towns
representative to the mayor?1
In the last year, how many times did you unite with others from .79 .08 10.50 <.001
this town in order to resolve a problem?1
Do you think that people here would contribute money toward a .37 .07 5.23 <.001
project to better this town?3
The leaders here bring people together to discuss decisions before .37 .06 5.77 <.001
making any decision about the progress of the town.4
In general, how would you classify the spirit of participation in this .33 .06 5.11 <.001
town?5
What level of inuence do residents have on the development of .39 .06 6.20 <.001
the town?6
Trust People here are only interested in their own well-being.4 .60 .05 11.61 <.001
Some people here are more trustworthy than others.4 .59 .03 19.07 <.001
Here people have to be alert or someone is likely to take .59 .03 19.07 <.001
advantage of you.4
Most people here are basically honest and can be trusted.7 <.001
If you have a problem, there is always someone to help.7
You dont pay attention to the opinions of others.7
Most people here are willing to help if you need it.7
You feel accepted as a member of this town.7
If you drop your purse or wallet in the street and someone sees it,
they will return it to you.7
If you leave something outside your house while you are away or
overnight, it will still be there when you get back.7
Sea turtle The conservation of sea turtles is important for you family.4 .65 .04 17.38 <.001
conservation You would like to participate in sea turtle conservation.4 .78 .03 22.50 <.001
perceptions You trust TAMAR to make decisions that benet this town.4 .71 .03 19.73 <.001
The protection of sea turtles is important for shers.4 .59 .03 15.57 <.001
You would like to know more about nature conservation.4 .69 .04 20.10 <.001
(Continued )
 HUMAN DIMENSIONS OF WILDLIFE 11

Table 1. (Continued).
Std. factor S. t- p-
Latent variable Observed variable loading E. value value
Upon nding a sick animal on the beach you would notify TAMAR.8
Cars, motorcycles and four-wheelers or dune buggies should not
travel on the beach at any time.8
TAMAR should talk with people here before making decisions
about sea turtle conservation.8
Lights from houses and hotels on the beach can impede newly
hatched sea turtles from reaching the ocean.8
Note. The parameter estimates provided here represent the factor loadings for the indicators in the measurement model
from the CFA run within the SEM. Trust was not used in the nal SEM model, therefore the parameter estimates
presented correspond to the factor loadings from the CFA for the latent factor trust run independently from the SEM.
1
Scale ranged from 1 (never) to 5 (always).
2
Items not used in the measurement model for social participation.
3
Scale ranged from 1 (would not contribute at all) to 4 (would contribute a lot).
4
Scale ranged from 1 (strongly disagree) to 5 (strongly agree).
Downloaded by [Michigan State University] at 08:10 18 December 2015

5
Scale ranged from 1 (very low) to 5 (very high).
6
Scale ranged from 1 (no inuence) to 4 (much inuence).
7
Items not used in the measurement model for trust.
8
Items not used in the measurement model for sea turtle conservation perceptions.

criteria. The residuals for two sets of items within the nal factor were allowed to correlate
based on modication indices and a review of the questions. Standardized parameter
estimates for each observed item variable are reported in Table 1.
All CFA models and the SEM were analyzed through the use of Mplus 7.0 software
(Muthn & Muthn, 2011). Mplus can appropriately model categorical variables in a
generalized latent variable framework and defaults to using probit regression with a
robust weighted least squares estimator based on a diagonal weight matrix (WLSMV)
(Muthn & Muthn, 2011). All the indicator variables used to measure the latent
variables are explicitly treated as ordinal data, using the WLSMV estimator, which
does not assume multivariate normality of those observed indicators. Model t was
tested using several criteria, including likelihood ratio chi-square statistic, root mean
square error of approximation (RMSEA .06), CFI and TLI (seeking .951.00), and
weighted root mean residual (WRMR) (WRMR .95, closer to 1 indicating good t;
Brown, 2006). Although WRMR is still considered an experimental model t criteria
lacking signicant research regarding cut o points (Yu & Muthn, 2002), it is included
here for the knowledge of the reader. Reliability scores for each scale were calculated
with SAS 9.3 software using the method by Green and Yang (2009), an alternative to the
traditionally reported coecient alpha, that allows for nonlinearity among factors and
item scores. Although the reliability coecient for collective action is low (>.65 is
considered acceptable; Vaske, 2008), the inclusion of this latent factor improved the
model t for the second order factor of social capital. With only one missing response
for social participation, all available data were used to estimate the model, the default
for dealing with missing data in Mplus. Model t measures and construct reliability for
all latent variables are reported in Table 2.
Structural equation model
Structural equation models estimate direct and indirect eects between latent and
observed variables simultaneously, incorporating the measurement models in the overall
structural model. Based on the latent factors above, a structural equation model was
 12 M. A. RINKUS ET AL.

Table 2. Model t and construct reliability for measurement models (n = 339).

Latent variable 2 df p-value RMSEA CFI TLI WRMR
Social capital 264.704 165 .0000 .042 .947 .939 .905
Community participation 27.294 9 .0013 .077 .951 .918 .694 .67
Civic participation 23.542 9 .0051 .069 .961 .936 .603 .69
Collective action 4.093 8 .8486 .000 1.000 1.033 .279 .50
Perceptions sea turtle conservation 7.743 3 .0516 .068 .997 .989 .271 .70
Trust 2.020 1 .1553 .055 .995 .986 .310 .52
Note. CFA models with a non-signicant chi-square statistic, a p-value > .10 or where 2 /df 3, are preferred.
All models indicated acceptable or good model t.

constructed to analyze the relationship between social capital, sea turtle conservation
perceptions, and participation. Trust was not signicant in the higher order factor for
social capital and therefore was included as a separate variable in the model. Given that
Downloaded by [Michigan State University] at 08:10 18 December 2015

the observed data were treated as categorical, the standardized parameters for all models
are reported (Byrne, 2012).
Although the initial model indicated good t with the data (2 = 508.927, df = 366) with
a RMSEA .06 and a CFI of .957, the latent variable for trust was not signicant and
exhibited high residual variance. Trust, therefore, was removed from the model for clarity,
resulting in an improved model t. The nal model (Figure 2) explained 42% of the
variance in sea turtle conservation participation. Social capital predicted 75%, 86%, and
54% of the variance in community participation, civic participation and collective action
respectively, reinforcing the validity of this construct. Both social capital ( = .392, p < .01)
and sea turtle conservation perceptions ( = .409, p < .01) had a signicant positive direct
relationship with participation in sea turtle conservation. The model also indicates that
social capital and sea turtle conservation perceptions are slightly correlated. Gender and

Figure 2. Results of structural equation model examining relationships among social capital, sea turtle
conservation perceptions, and participation in sea turtle conservation. Fit indices: 2 = 403.186,
df = 290, RMSEA = .034, CFI = .964, TLI = .959, WRMR = .924.
Note. All parameters reported are standardized. Circles represent latent concepts and squares represent
observed variables.
 HUMAN DIMENSIONS OF WILDLIFE 13

age were tested as covariates and found to both be signicant, but did not improve model
t when added individually or together. Only one community (Barra), measured as a xed
eect, was found to be statistically signicant.
Furthermore, the addition of gender, age, and community of residence to the model
only slightly increased the R2 and slightly reduced the eect of social capital and sea turtle
conservation perceptions on conservation participation. Given that gender, age, and
community of residence did not improve overall model t, the second model was deemed
to be the most parsimonious and therefore only the results of the model in Figure 2 will be
discussed.

Discussion
Downloaded by [Michigan State University] at 08:10 18 December 2015

These results indicate that both social capital and conservation perceptions have a positive
relationship to participation in conservation, meaning that as social capital increases and
perceptions become more positively aligned with conservation, participation in sea turtle
conservation increases. Additionally, social capital and conservation perceptions were
positively correlated. This is important given that conservation eorts rely heavily on
programs that aim to change individual conservation perceptions to increase participation
either formally or informally through behavior change. The correlation between social
capital and conservation perceptions, as well as their equally positive inuence on participa-
tion in conservation, provide further evidence that conservation perceptions are related to
other aspects of community social structure. Although it is possible that relationships
between perceptions and participation, and social capital and participation could be inter-
dependent, it is dicult to test both directions of these relationships with cross-sectional
data and would be better suited for a longitudinal study with a larger sample size.
The ndings presented provide support for civic and community participation and
collective action as social capital constructs, and further validate the conceptualization of
social capital. However, trust was not found to be signicant in the second-order factor for
social capital, nor was it signicant as an independent factor in the structural model. Thuy
et al. (2011) also found trust not to be signicant when examining the relationship
between social capital and conservation perceptions around a protected area in
Vietnam. Social trust, often viewed as the glue that holds groups together, can take
generations to build and be inuenced by personal and institutional relationships. One
explanation for why trust was not signicant here may be that only generalized social trust
was measured, not institutional trust, which could have a dierent eect (Jones, 2010).
Given that trust is an abstract construct, common conceptions of trust in the social capital
literature center on two types: generalized trust (trust in people overall) and trust in
particular people or groups of people (e.g., family, friends, neighbors or people of the same
religion or ethnicity) or institutions (e.g., government, non-prots, police; Grootaert,
Narayan, Jones, & Woolcock, 2003; Narayan & Cassidy, 2001). Narayan and Cassidy
(2001) have recommended that questions regarding trust be tailored to the community
being studied based on local context. More time in the study communities would have
allowed for this examination. Furthermore, trust might need to be measured dierently in
Brazil and other countries where political corruption is high and levels of trust toward
public institutions are low (Wike & Holzwart, 2008).
 14 M. A. RINKUS ET AL.

Although data were not collected to measure political trust, during the time of data
collection many people expressed their displeasure with the current mayor of the munici-
pality and lamented the level of corruption in governmentlocal, state, and federal.
Another factor possibly aecting the measurement of social trust in the communities
studied is the increase of new and temporary residents from other parts of the state. A
common theme in the interviews revolved around changes to the community, such as the
increase in crime and drugs over the years as tourism and second home development
increased. Although cultural dierences may provide an explanation for the varying
signicance of trust, it may also relate to Woolcock and Narayans (2000) assertion that
trust is not a measure of social capital, but an outcome of the creation and maintenance of
norms and networks over time. Therefore, low levels of social capital in these commu-
nities resulted in low levels of trust. Whether viewed as an outcome or predictor, it is
widely believed that trust is needed to maintain and build social capital (Berkes, 2009;
Downloaded by [Michigan State University] at 08:10 18 December 2015

Bouma, Bulte, & Van Soest, 2008; Cramb, 2005), requiring that social capital be viewed as
more of a dynamic process. Furthermore, the measurement of social capital is still under
much debate in the literature and can benet from such studies.
The main nding indicates that social capital is positively related to participation in
conservation. The interplay between individual characteristics and contextual or situa-
tional circumstances is important in predicting and understanding participation at the
community level (Wandersman & Giamartino, 1980). This signies that time needs to be
spent engaging with the community and community organizations, not just increasing
awareness of sea turtle conservation. For example, it may be more benecial (and ecient)
for conservation programs to work through and with existing organizations to increase
awareness, participation, and support since the individuals participating are more likely to
be open to calls for collective action and are accustomed to working together voluntarily
for individual, familial, and community benets. Furthermore, by understanding how,
where, and why people participate in their communities, conservation sta can tailor
conservation programs to make them more accessible and attractive to community
members.
In the case of sea turtle conservation in the four communities studied, participation is
currently limited to educational outreach events and compliance. More meaningful parti-
cipation would require a collaborative relationship with schools and shing associations
that work to understand how these groups want to participate in sea turtle conservation.
The program could also benet from more attempts to connect with religious organiza-
tions and events that are targeted at families or individuals of all ages. Such integration
with other organizations would create a larger platform for information exchange and
facilitate the creation of new networks or relationships. As noted earlier, social capital in
relation to conservation is often examined and measured at the group level. However,
social capital at the group level only deals with direct stakeholders, or those already
involved, and dismisses the larger social and institutional context. Although studies at
the group level are important for understanding and managing group dynamics, the
approach used in this study provides a broader picture of social capital at the community
level.
The positive inuence of social capital on conservation participation leads to two
conclusionsthat conservation would work best in communities with existing high levels
of social capital and that if social capital is increased, conservation participation may also
 HUMAN DIMENSIONS OF WILDLIFE 15

increase. Although increased social capital may provide a better environment for imple-
menting conservation programs, to be eective conservation eorts cannot aord to
ignore other communities. A complex task, improving social capital requires supporting
and building social structure, including formal and informal institutions (Berkes, 2009;
Brune & Bossert, 2009; Pretty & Smith, 2004). This may be challenging for conservation
managers at the community or regional scale due to local politics and power dynamics,
not to mention limited resources and expertise in capacity building. Although TAMAR
provides nancial or in-kind support to dierent groups and organizations, more eort
could be made to provide institutional support (training, facilitation, and assistance in
connecting organizations to other available resources). Thuy et al. (2011) suggested that
community protected area programs should work to provide more opportunities for
interaction and participation among community members through existing formal or
informal organizations. De Souza Briggs (2004) noted a need to understand social capital
Downloaded by [Michigan State University] at 08:10 18 December 2015

under development, which could provide a greater understanding of the creation and
maintenance of social capital (p. 155). Future research regarding the development of social
capital would benet most from in-depth qualitative studies to capture its dynamic and
nuanced process (Grootaert et al., 2003).
As development and urbanization continue, visitors to coastal Brazil looking to escape
hectic city life are venturing further to more remote communities and beaches that were
once safe nesting areas for sea turtlesfree from human activity, lights, and motorized
vehicles. The social and biological repercussions of this expansion will require enhanced
community engagement that would benet from long-term interdisciplinary study. The
separate treatment of conservation from other forms of community participation, in both
research and conservation, also perpetuates the view that conservation issues are separate
from peoples social and economic lives. The model presented can be used as a tool for
analyzing the relationship between conservation- and non-conservation-related participa-
tion at the individual and community levels. If conservation managers take account of
aspects related to social capital and incorporate this information into their planning,
conservation participation may have a better chance of becoming integrated in formal
and informal community institutions. Further examination of social capital could benet
community engagement in conservation by providing a road map of community social
structure to elucidate the cognitive and structural connections between conservation and
non-conservation participation, leading to the creation of more long-term and meaningful
engagement. This study presents a novel approach to understanding the role of social
capital in conservation participation and demonstrates the importance of non-conserva-
tion-related participation to conservation.

Acknowledgments
We thank Projeto TAMAR and sta in Stio do Conde, Bahia, Brazil, Dr. Snia Seixas, Ncleo de
Estudos e Pesquisas Ambientais, Universidade Estadual de Campinas (UNICAMP), research assis-
tants Kercia Ribeiro, Emanuel Soares, and Jordana Nascimento, and research participants in the
communities of Stio do Conde, Poas, Siribinha, and Barra de Itariri for their contributions to and
support of this research. We also acknowledge Dr. Steve Pierce, Assistant Director, Center for
Statistical Consulting (CSTAT) at Michigan State University for his statistical assistance and the
editors and anonymous reviewers for their comments and suggestions.
 16 M. A. RINKUS ET AL.

Funding
This research was supported by the U.S. Student Fulbright Program in Brazil.

References
Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckman
(Eds.), Action-control: From cognition to behavior (pp. 1139). Heidelberg, Germany: Springer.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood
Clis, NJ: Prentice-Hall.
Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Institute of
Planners, 35, 216224. doi:10.1080/01944366908977225
Ballet, J., Sirven, N., & Requiers-Desjardins, M. (2007). Social capital and natural resource manage-
ment a critical perspective. The Journal of Environment & Development, 16, 355374.
Downloaded by [Michigan State University] at 08:10 18 December 2015

doi:10.1177/1070496507310740
Berkes, F. (2009). Evolution of co-management: Role of knowledge generation, bridging organiza-
tions and social learning. Journal of Environmental Management, 90, 16921702. doi:10.1016/
j.jenvman.2008.12.001
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health:
Durkheim in the new millennium. Social Science & Medicine, 51, 843857. doi:10.1016/S0277-
9536(00)00065-4
Blaikie, P. (2006). Is small really beautiful? Community-based natural resource management in
Malawi and Botswana. World Development, 34, 19421957. doi:10.1016/j.worlddev.2005.11.023
Bouma, J., Bulte, E., & Van Soest, D. (2008). Trust and cooperation: Social capital and community
resource management. Journal of Environmental Economics and Management, 56, 155166.
doi:10.1016/j.jeem.2008.03.004
Brosius, J. P., Tsing Lowenhaupt, A., & Zerner, C. (2005). Communities and conservation: Histories
and politics of community-based natural resource management. Walnut Creek, CA: AltaMira
Press.
Brown, T. A. (2006). Conrmatory factor analysis for applied research. New York, NY: Guilford Press.
Brune, N. E., & Bossert, T. (2009). Building social capital in post-conict communities: Evidence
from Nicaragua. Social Science & Medicine, 68, 885893. doi:10.1016/j.socscimed.2008.12.024
Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts, applications, and
programming. New York, NY: Routledge.
Campbell, J., & Godfrey, M. (2002). Community-based conservation via global legislation?
Limitations of the Inter-American Convention for the protection and conservation of sea turtles.
Journal of International Wildlife Law and Policy, 5, 121143. doi:10.1080/13880290209354002
Chase, L. C., Schusler, T. M., & Decker, D. J. (2000). Innovations in stakeholder involvement:
Whats the next step? Wildlife Society Bulletin, 28, 208217.
Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology,
94, S95S120. doi:10.1086/ajs.1988.94.issue-s1
Cornwall, A. (Ed.). (2011). The participation reader. New York, NY: Zed Books.
Cramb, R. A. (2005). Social capital and soil conservation: Evidence from the Philippines. The
Australian Journal of Agricultural and Resource Economics, 49, 211226. doi:10.1111/
ajar.2005.49.issue-2
Creswell, J. W., & Plano, V. L. (2010). Designing and conducting mixed method research. Thousand
Oaks, CA: Sage.
De Souza Briggs, X. N. (2004). Social capital: Easy beauty or meaningful resource? Journal of the
American Planning Association, 70, 151158.
Duit, A., Hall, O., Mikusinski, G., & Angelstam, P. (2009). Saving the woodpeckers: Social capital,
governance, and policy performance. The Journal of Environment & Development, 18, 4261.
doi:10.1177/1070496508329302
 HUMAN DIMENSIONS OF WILDLIFE 17

Dulal, H. B., Foa, R., & Knowles, S. (2011). Social capital and cross-country environmental
performance. The Journal of Environment & Development, 20, 121144. doi:10.1177/
1070496511405153
Fundao Centro Brasileiro de Proteo e Pesquisa das Tartarugas Marinhas, (Pr TAMAR).
(2010). Relatrio de Atividades 2010. Brazil: Author.
Fundao Centro Brasileiro de Proteo e Pesquisa das Tartarugas Marinhas, (Pr TAMAR).
(2011). Relatrio de Atividades 2011. Brazil: Author.
Grafton, R. Q., & Knowles, S. (2004). Social capital and national environmental performance: A
cross-sectional analysis. The Journal of Environment & Development, 13, 336370. doi:10.1177/
1070496504271417
Green, S. B., & Yang, Y. (2009). Commentary on coecient alpha: A cautionary tale. Psychometrika,
74, 121135. doi:10.1007/s11336-008-9098-4
Grootaert, C., Narayan, D., Jones, V. N., & Woolcock, M. (2003). Integrated questionnaire for the
measurement of social capital (SC-IQ). Washington, DC: World Bank.
Grootaert, C., & Van Bastelaer, T. (2001). Understanding and measuring social capital: A synthesis of
Downloaded by [Michigan State University] at 08:10 18 December 2015

ndings and recommendations from the social capital initiative. Washington, DC: World Bank.
Hawe, P., & Shiell, A. (2000). Social capital and health promotion: A review. Social Science &
Medicine, 51, 871885. doi:10.1016/S0277-9536(00)00067-8
Heberlein, T. A. (2012). Navigating environmental attitudes. New York, NY: Oxford University
Press.
Hodgkin, S. (2008). Telling it all a story of womens social capital using a mixed methods approach.
Journal of Mixed Methods Research, 2, 296316. doi:10.1177/1558689808321641
Instituto Brasileiro de Geograa e Estatstica (IBGE). (2010). Censo Demogrco 2010. Retrieved
from http://www.ibge.gov.br/cidadesat/
Jin, M. (2013). Does social capital promote pro-environmental behaviors? Implications for colla-
borative governance. International Journal of Public Administration, 36, 397407. doi:10.1080/
01900692.2013.773038
Jones, N. (2010). Environmental activation of citizens in the context of policy agenda formation and
the inuence of social capital. The Social Science Journal, 47, 121136. doi:10.1016/
j.soscij.2009.05.008
Jones, N., Halvadakis, C. P., & Sophoulis, C. M. (2011). Social capital and household solid waste
management policies: A case study in Mytilene, Greece. Environmental Politics, 20, 264283.
doi:10.1080/09644016.2011.551032
Khknen, S. (1999). Does social capital matter in water and sanitation delivery?: A review of the
literature (Social Capital Initiative Working Paper No. 13). Washington, DC: World Bank.
Kramer, D. B. (2007). Determinants and ecacy of social capital in lake associations. Environmental
Conservation, 34, 186194. doi:10.1017/S0376892907004171
Krishna, A., & Shrader, E. (1999). Social capital assessment tool. Prepared for the Conference on
Social Capital and Poverty Bank, June 2224. Washington, DC: World Bank.
Krishna, A., & Upho, N. (1999). Mapping and measuring social capital: A conceptual and
empirical study of collective action for conserving and developing watersheds in Rajasthan,
India (Social Capital Initiative Working Paper No. 13). Washington, DC: World Bank.
Lansing, D. (2009). The spaces of social capital: Livelihood geographies and marine conservation in
the Cayos Cochinos Marine Protected Area, Honduras. Journal of Latin American Geography, 8,
2954. doi:10.1353/lag.0.0032
Macias, T., & Nelson, E. (2011). A social capital basis for environmental concern: Evidence from
Northern New England. Rural Sociology, 76, 562581. doi:10.1111/ruso.2011.76.issue-4
Marcovaldi, M. A., Patiri, V., & Thom, J. C. (2005). Projecto TAMAR-IBAMA: Twenty-ve years
protecting Brazilian sea turtles through a community-based conservation programme. MAST.
3 and 4, 3962.
Marquart-Pyatt, S. (2012). Explaining environmental activism across countries. Society and Natural
Resources, 25, 683699. doi:10.1080/08941920.2011.625073
 18 M. A. RINKUS ET AL.

Mitchell, A. D., & Bossert, T. J. (2007). Measuring dimensions of social capital: Evidence from
surveys in poor communities in Nicaragua. Social Science & Medicine, 64, 5063. doi:10.1016/j.
socscimed.2006.08.021
Muthn, L. K., & Muthn, B. O. (2011). Mplus users guide (7th ed.). Los Angeles, CA: Muthn &
Muthn.
Narayan, D., & Cassidy, M. F. (2001). A dimensional approach to measuring social capital:
Development and validation of a social capital inventory. Current Sociology, 49, 59102.
doi:10.1177/0011392101049002006
Ostrom, E., Dietz, T., Nives, D., Stern, P., Stonich, S., & Weber, E. U. (2002). The drama of the
commons. Committee on the Human Dimensions of Global Change, National Research Council
(U.S.). Washington, DC: National Academy.
Pargal, S., Mainul, H., & Gilligan, D. (1999). Social capital in solid waste management: Evidence
from Dhaka, Bangladesh (Social Capital Initiative Working Paper No. 16). Washington, DC:
World Bank.
Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual Review of
Downloaded by [Michigan State University] at 08:10 18 December 2015

Sociology, 24, 124. doi:10.1146/annurev.soc.24.1.1

Pretty, J., & Smith, D. (2004). Social capital in biodiversity conservation and management.
Conservation Biology, 18, 631638. doi:10.1111/cbi.2004.18.issue-3
Pretty, J. N. (1995). Participatory learning for sustainable agriculture. World Development, 23, 1247
1263. doi:10.1016/0305-750X(95)00046-F
Putnam, R. D. (1993). The prosperous community: Social capital and public life. The American
Prospect, 13, 3542.
Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. New York,
NY: Simon & Schuster.
Rubin, H. J., & Rubin, I. S. (2005). Qualitative interviewing: The art of hearing data (2nd ed.).
Thousand Oaks, CA: Sage.
Sekhar, N. U. (2007). Social capital and sheries management: The case of Chilika Lake in India.
Environmental Management, 39, 497505. doi:10.1007/s00267-006-0183-0
Thuy, N. N., Dwivedi, P., Rossi, F., Alavalapati, J. R. R., & Thapa, B. (2011). Role of social capital in
determining conservation attitude: A case study from Cat Tien National Park, Vietnam.
International Journal of Sustainable Development & World Ecology, 18, 143153. doi:10.1080/
13504509.2011.560455
Vaske, J. (2008). Survey research and analysis: Applications in parks, recreation and human dimen-
sions. State College, PA: Venture.
Wagner, C. L., & Fernandez-Gimenez, M. E. (2009). Eects of community-based collaborative
group characteristics on social capital. Environmental Management, 44, 632645. doi:10.1007/
s00267-009-9347-z
Wandersman, A., & Giamartino, G. A. (1980). Community and individual dierence characteristics
as inuences on initial participation. American Journal of Community Psychology, 8, 217228.
doi:10.1007/BF00912661
Wike, R., & Holzwart, K. (2008, April 15). Where trust is high, crime and corruption are low. Pew
Global Attitudes Project. Retrieved from http://www.pewglobal.org/2008/04/15/where-trust-is-
high-crime-and-corruption-are-low/
Woolcock, M., & Narayan, D. (2000, August). Social capital: Implications for development theory,
research, and policy. Washington, DC: World Bank Research Observer.
Yu, C. Y., & Mthen, B. O. (2002, April). Evaluation of model t indices for latent variable models
with categorical and continuous outcomes. Paper presented at the annual meeting of the
American Educational Research Association, New Orleans, LA.