Professional Documents
Culture Documents
Published in Canada by
Oxford University Press
8 Sampson Mews, Suite 204,
Don Mills, Ontario M3C 0H5 Canada
www.oupcanada.com
Every effort has been made to determine and contact copyright holders.
In the case of any omissions, the publisher will be pleased to make
suitable acknowledgement in future editions.
Glossary 403
References 410
Index 426
Strategies for observing behaviour 137 Questions for Review and Creative Application 214
Questions for Review and Creative Application 268 Quantitative research and constructionism 328
Interactive Classroom Activities 269 Research methods and epistemological and
Relevant Websites 270 ontological considerations 328
Problems with the quantitative/qualitative contrast 329
12 Content Analysis 271 Mutual analysis 331
Chapter overview 271 Quantification in qualitative research 332
Introduction 272 Multi-strategy research 332
Personal documents 273 Two positions in the debate over quantitative and
Government documents 277 qualitative research 334
Official documents from private sources 279 Approaches to multi-strategy research 334
PART II consists of five chapters concerned with semi-structured or unstructured) and focus
quantitative research. groups, in which groups of individuals are
interviewed on a specific topic.
• Chapter 4 presents the fundamentals of quan- • Chapter 12 applies qualitative approaches to
titative research and provides the context for content analysis, a method used in the study
later chapters. of “documents” ranging from books, letters,
• Chapter 5 focuses on structured interviewing and newspapers to movies, chat lines, and
and the design of questionnaires. It delves television shows. It also examines two ways to
into how to write questions for both question- analyze language: conversation analysis and
naires and interviews. It also discusses how discourse analysis.
to compose a self-completion questionnaire, • Chapter 13 explores some approaches to
using data from already-completed question- the analysis of qualitative data, including
naires and interviews. grounded theory and coding.
• Chapter 6 covers structured observation, a
method developed for the systematic observa- PART IV moves beyond the quantitative/quali-
tion of behaviour. tative division to explore what the two approaches
• Chapter 7 deals with quantitative sampling: have in common, how they may complement each
how to select a sample and the considerations other, and how they may be combined in the same
involved in assessing what can be inferred research project.
from different kinds of samples.
• Chapter 8 presents a range of basic non- • Chapter 14 proposes that the distinction be-
technical tools for quantitative data analysis. tween quantitative and qualitative research
The emphasis is on how to choose a method of may be less fixed than is sometimes supposed,
analysis and how to interpret findings. In order and presents some ways in which they can be
to keep the focus on methodological concepts combined to produce multi-strategy research.
and interpretations, formulae are not discussed. • Chapter 15 provides guidance on writing up
research, an often-neglected area in the teach-
PART III presents five chapters on aspects of ing of the research process.
qualitative research. • Chapter 16 offers advice on conducting a re-
search project, taking readers through the
• Chapter 9 plays the same role for Part III that main steps involved.
Chapter 3 does for Part II. It provides an over-
view of the nature of qualitative research and Finally, the Appendix presents an easy-to-access
hence the context for the other chapters in this resource for successful research.
part.
• Chapter 10 discusses ethnography and partici- • The appendix explains how to use IBM
pant observation. The two terms are often used SPSS Statistics Software (SPSS) and NVivo
interchangeably to refer to the immersion of software to perform, respectively, the quan-
the researcher in a social setting, a technique titative data analyses described in Chapter 8
that is the source of some of the best-known and the qualitative data analyses discussed
studies in social research. in Chapter 13. The SPSS material has been
• Chapter 11 examines the kinds of interview updated to the latest version of IBM SPSS
that qualitative researchers conduct (typically (released in 2018).
Brief Contents
Guide to the Book
Special Features of the Book
vii
ix NEW! Organization. This fifth edition has
Acknowledgments xviii
Preface xix
been reorganized to better reflect how social
PART I Fundamental Issues in Social Research
1 General Research Orientations 2 research as a discipline is taught across
2 Research Designs 27
3 Research Ethics 50 Canada today.
Appendix to Part I The Ideal Stages of Research 69
outline a research story that has appeared in a MacEwan University, conducted qualitative inter-
views with 100 Indigenous and non-Indigenous
also noted that some teachers did not feel confi-
dent enough to address topics relating to Indigen-
parents and teachers in southern Ontario. The ous people, and were wary of giving offence. “The
major media outlet, illustrating how social research purpose of the study was to document the inter-
viewees’ perceptions of Ontario government
problem is that when you have people that are un-
comfortable and intimidated, the result is that we
policy directives designed to introduce Indigen- have educators that may not be doing it at all,” she
can have real impacts on our everyday lives. ous history, culture, and experiences into the cur-
riculum (Canadian Press, 2017).
said (Canadian Press, 2017). Milne recommended
that “Indigenous coaches” be used by teachers as
Milne found that the teachers she spoke to a learning resource. Some of the challenges she
were generally quite willing to incorporate In- identified included how to use appropriate, cul-
digenous perspectives into their classroom turally sensitive terminology when discussing In-
activities, and she observed that Indigenous par- digenous issues, and how to present the history of
ents were in favour of non-Indigenous teachers residential schools.
48 PARTIFundamentalIssuesinSocialResearch
KeyPoints
• Thereisanimportantdistinctionbetweenagen- • Replicability,validity(measurementandexternal),
eral research orientation (quantitative versus andtheabilitytoestablishcausationareimport-
qualitative)andaresearchdesign. antcriteriaforevaluatingthequalityofquantita-
• Thenomotheticapproachtoexplanationinvolves tivesocialresearch.
discoveringgenerallawsandprinciples. • Four key research designs are experimental,
• Nomothetic explanations must satisfy three cri- cross-sectional,longitudinal,andcasestudy.
teria of causation: correlation, time order, and • Threats to the establishment of causation are of
non-spuriousness. particularimportanceinnon-experimental,quan-
• Qualitative researchers usually take the idio-
graphic approach to explanation, which entails
titativeresearch.
• External validity is a concern with case studies
NEW! Expanded end-of-chapter
questions. Questions designed to
creating a rich description of a person or group (generalizability) and laboratory experiments
based on the perceptions and feelings of the (findings may not be applicable outside the re-
peoplestudied. searchenvironment).
10
Ethnography and Participant Observation
Chapter Overview
Ethnography and participant observation require extended involvement in the activities of the people
under study. This chapter explores:
• the problems of gaining access to different settings and ways of overcoming them;
• whether covert research is practicable and acceptable;
• the role of key informants;
• the different roles that ethnographers can assume in the course of their fieldwork;
• the function of field notes and the forms they can take;
• the role of visual materials in ethnography;
• bringing an ethnographic study to an end; and
• the issue of feminist ethnography.
Chapter-opening vignettes. At the Do you like to travel to places you’ve never been to
before? Have you ever observed a group of people
Ethnography and participant observation involve
placing yourself in a social environment that may be
you don’t know very well and wondered what it would foreign to you, and staying there for an extended period
beginning of each chapter, the topics to be be like to be a member of their group? Have you ever
witnessed profound human suffering and asked your-
of time. What kinds of groups or social settings intrigue
you? Non-governmental organizations? Political move-
addressed are introduced in an informal self how things ended up that way and how the people
suffering managed to endure? Would you like to give
ments? Sports teams? Criminal gangs? Hospital emer-
gency rooms? Women’s shelters? All of these can be
such people a voice or expose the hardships that they subjects of ethnographic and participant observation
and provocative way to help students grasp face? If so, doing or at least reading about ethnography
and participant observation should interest you.
research.
discussed in the chapter have been describe, it definitely feels different than prior to
[marriage]” (Green, 2010, p. 411). The people stud-
society and in the process lose their oppositional
tenor. Like other predictions for social change,
ied also mentioned that being married bestowed those regarding gay marriage can be tested only
used to study Canadian society. a sense of legitimacy on their relationship and with the passage of time.
1
bry29440_ch09_197-215.indd 12/17/18 06:55 PM
Chapter Overview
The aim of this chapter is to examine the fundamental assumptions upon which social research is based.
An important distinction commonly drawn by practitioners of social research—between the quantitative
and qualitative approaches—is explored in relation to those considerations. We will consider:
• the relationship between theory and research—in particular, whether theories and the hypotheses
derived from them are tested by gathering data (a deductive approach) or whether data gathering
is used as a means to create theory (an inductive approach);
• epistemological issues, such as whether a natural science model like the one used in chemistry
or biology is suitable for the study of the social world;
Chapter overviews. Each chapter opener
• ontological issues, such as whether the social world should be regarded as a reality external to
individuals over which they have little or no control, or as something that social actors may fashion
includes an overview that serves as a route
into their personal realities;
• how values and practical issues impinge on the research process; and map, alerting readers to what they can
• how these issues relate to both quantitative and qualitative research; a preliminary discussion,
followed up in Chapter 14, suggests that although the quantitative and qualitative orientations are expect to learn.
different, they complement each other.
Soon-Yi wants to find out why Indigenous people in The list of topics she could collect information on
Canada are more likely to live in poverty than other seems endless.
Canadians, but doesn’t know where to begin. Should Maybe rather than beginning her study by accumu-
she start by examining the history of colonialism and lating data, it would it be better to start out with some
conflict between Indigenous peoples and settler- hunches and then gather information to see whether
colonizers, such as disputes over land claims and treat- they are supported by evidence. For example, perhaps
ies, residential schools, or anti- Indigenous prejudice? the discrepancy in economic conditions is a manifesta-
Or how about gathering aggregate data on present tion of a centuries-old system of international domin-
conditions like residence patterns, economic activ- ance and exploitation. Similarly, it could have arisen
ities, the age structure, or educational trajectories? through a clash of civilizations and cultures. Then there
▲ LeonWang/Shutterstock
3 | Research Ethics 55
A key question asked in evaluation research is at its completion. The researchers also did a quali-
whether a new policy initiative or organizational tative analysis of the project by conducting five
change achieved its goals. Ideally, to answer that focus groups at the conclusion of the program. The
question the design would have one group that is quantitative results indicated that the participants
exposed to the treatment—the new initiative—and had higher levels of perceived overall health and
80 groupPart II Quantitative Research
a control that is not. Since it is often not feas- sense of community, and lower levels of physical
ible or ethical to randomly assign research partici- pain, when the program was over. The themes
pants to the two groups, such studies are usually that emerged from the focus groups included the
quasi-experimental. For instance, data gathered conclusions that the program provided the seniors
Boxes. Special feature boxes from people before a change may be compared
BOX 4.2 A multiple-indicator with structure and discipline, facilitated coping, re-
measure of another concept
with data acquired after; the “before” people quired hard work and effort, brought out their art-
provide in-depth examples of how become the control group, the “after” people the istic side, promoted social involvement, and made
In Hay’s (2014) study of secularization (see Box 1.3), frequency of religious attendance, was measured
experimental group. This approach has the added a positive contribution to the community.
the various research methods religious pluralism was measured using a single,
advantage that the two groups are basically the
with the question, “Do you currently attend church
Quantitative quasi-experimental designs in
five-point Likert item that formed part of an exten- temple or mosque?” Respondents who answered
discussed in the book have been same, making random assignment unnecessary. evaluation research go back a long way, but as the
sive survey of Canadians’ value systems. However, “yes” were then asked: “How often?” The response
Such a design was used to evaluate the effect Phinney et al. (2014) study indicates, evaluations
used in real research situations. secularization (the dependent variable in Hay’s
of a community arts program on the well-being of
choices were: “once a week or more” (given a code
based on qualitative research have also emerged.
analysis) was measured with several different indi- of 5 after reverse-coding), “monthly” (4), “every
older adults in the Vancouver area (Phinney et al., Although there are differences of opinion about
The boxes also list the advantages cators in order to tap into different dimensions of
2014). Over three years, four groups of participants
few months” (3), “once or twice a year” (2), and
how qualitative evaluation should be carried
the concept. One dimension, religiosity, was meas- “never” (1). A third dimension, concerned with the
and disadvantages of a particular took part in the collective creation of a physical out, there is consensus on the importance of,
ured by averaging the responses to three 10-point participants’ belief in the religion of their par-
work of art or a performance that was presented first, understanding the context in which an inter-
method, summarize important items indicating the importance respondents
to the public. Baseline quantitative measures of
ents, had the response categories “believe all of
vention occurs and, second, hearing the diverse
placed on: (a) “believing in God;” (b) “obeying God, it” (4), “believe most of it but not all” (3), “believe
well-being were taken in the first year of the pro- viewpoints of the stakeholders (Greene, 2000).
points, discuss methodological doing what he wishes;” and (c) “relating to God in
gram, with the same measures administered again
some parts but disbelieve others” (2), and “don’t
For example, Pawson and Tilley (1997) advocate
a personal way.” A second dimension, relating to believe any of it” (1).
controversies, and offer practical
advice. dimension of the concept (e.g., respecting confi- simply tags to allow the material to be stored quanti-
In quantitative studies, data are collected on to difficulties in holding down a job and thus
dentiality) may not necessarily score high on other tatively. Then it is necessary to go through the infor-
two or more variables, which are then examined poverty? Or is it a bit of both? To take another
dimensions (e.g., fiscal honesty or continuing edu- mation again to look for incidences of the theme or
to detect patterns of association. This approach example, a study of 1000 men found that those
cation), so that for each respondent one can have a category, and to record the appropriate numbers on
sometimes makes it difficult to show cause and who had two or more orgasms a week exhibited
multidimensional “profile.” a computer spreadsheet. This approach is sometimes
effect because the independent and dependent a 50 per cent lower mortality risk compared with
However, in much quantitative research, there called post-coding. Post-coding can be an unreliable
variables are measured simultaneously, making men who had on average less than one orgasm
is a tendency to rely on a single indicator for each procedure because there may be inconsistencies in
any demonstration of time order (showing that per week. It may be tempting to conclude that
Special Features of the Book xiii
Checklist
Checklist of issues to consider for a structured interview schedule or questionnaire
☐☐ Is a clear and comprehensive introduction to ☐☐ Are questions relating to the research topic
the research provided for respondents? asked near the beginning of the interview or
☐☐ Are there any questions used by other re- questionnaire?
searchers that would be useful? ☐☐ Have the following been avoided?
☐☐ Will the questions provide answers to all the • ambiguous terms in questions or response
research questions? choices
☐☐ Are there any questions not strictly relevant • long questions
Checklists. Most chapters also to the research questions that could be • double-barrelled questions
• very general questions
include checklists of points to dropped?
• leading questions
☐☐ Has the questionnaire been pre-tested with
keep in mind when engaging in some appropriate respondents?
• questions that include negatives
• questions using technical terms
a particular activity, whether ☐☐ If a structured interview schedule is used, are
the instructions clear? For example, with filter ☐☐ Do respondents have the knowledge required
devising a structured interview questions, is it clear which question(s) should to answer the questions?
be omitted? ☐☐ Is there an appropriate match between ques-
schedule, conducting a focus tions and response choices?
☐☐ Are instructions about how to record re-
group, or doing a literature sponses clear (for example, whether to tick ☐☐ Are the response choices properly balanced?
or circle; whether more than one response is ☐☐ Do any of the questions depend too much on
review. Checklists reinforce allowable)? respondents’ memories?
key points and remind students ☐☐ Has the number of open questions been
If using a Likert scale approach:
limited?
of things they need to consider ☐☐ Can respondents indicate levels of intensity ☐☐ Are some items that have to be
in their replies, or are they forced into “yes or
when doing their own research. no” answers?
reverse-scored included, in order to identify
response sets?
☐☐ Have questions and their answers been kept ☐☐ Is there evidence that the items really do
on the same page? relate to the same underlying cluster of atti-
☐☐ Have socio-demographic questions been tudes, so that the items can be aggregated?
left until near the end of the interview or ☐☐ Are the response choices exhaustive and not
questionnaire? overlapping?
However, there are several possible departures from as well, and to administer a structured interview to a
this pattern. focus group would be very unusual. In most survey
research a specific individual is the object of ques-
xiv Special Features of the Book
A Online focus groups are appropriate for research Qualitative interviewing alone versus
involving sensitive issues. Identify three issues ethnography
that, because of their sensitive nature, would R Outline the advantages and disadvantages of
be better researched with online rather than in- qualitative interviewing (without immersion in a
person focus groups, then explain why the online social setting) compared to ethnography.
technique would be more appropriate. A Is one method more in tune with the research
needs of qualitative researchers than the other?
Feminist research and interviewing in Explain, using the topic of intimate partner vio-
qualitative research lence to illustrate your answer.
R Why are qualitative interviews so prominent in
feminist research?
A Explain why focus groups may be superior to other
methods of inquiry for giving a voice to highly mar-
ginalized women.
the classroom.
general discussion of:
they developed;
b. whether the moderator’s control of the discus- a. difficulties in getting the interview to flow
sion was excessive, about right, or too weak, smoothly, and how those difficulties may be
and what the consequences of that were; resolved;
c. to explain the advantages of the focus group b. illustrations of how topics that were not on the
method compared to one-on-one qualitative interview guide made their way into the inter-
interviews for researching the topic chosen; view anyway, and how that helped or hindered
d. to explain the disadvantages of the focus group the investigation of the topic; and
method compared to one-on-one qualitative c. the sorts of topics that could be usefully re-
interviews for researching the topic chosen; and searched using semi-structured interviews,
e. to explain how the focus group method may be and which topics would be better pursued
better than structured interviews for exploring using ethnography or structured interviews.
Key Points
• It’s important not to exaggerate the differences methods can be used to analyze the rhetoric of
between quantitative and qualitative research. quantitative researchers.
• Connections between358 epistemology and ontology
PART IV Transcending • Some qualitative
the Quantitative/Qualitative Divideresearchers employ
and Some Practical quantifica-
Advice Key points. Each chapter concludes
on the one hand, and research methods on the tion in their work.
other, are not fixed or absolute.
• Qualitative research can exhibit featuresHypothetical
approached.
• Although the practice of multi-strategy research
normally results hasandincreased, not all writers
conclusions support
café. it. is shown the picture or clip, and each
The class
with a summary of its most
associated with a natural science model. • The view that there are epistemological and onto-
are given for points “f” and “g.”
• Quantitative research can incorporate an interpre-
student then writes up a description of the physi-
logical impediments to the combination of quan-
Once that is done, students leave their groups cal setting that is depicted (maximum 250 words).
significant points.
tivist stance. titative and qualitative research is a barrier to
and the class is reconvened. Each person in the Three volunteers then read their descriptions to the
• The artificial/natural contrast used to distinguish multi-strategy research.
class is given 15–20 minutes to write a brief Intro- rest of the class. Next, the instructor facilitates a
quantitative and qualitative research is often • There are several different ways of combining
exaggerated. duction to their group’s article. The Introduction
quantitative and qualitative discussion based
research; someon the
canfollowing questions:
should not
• A quantitative research approach canexceed
be used250to words and should be
be planned writ-
in advance, others cannot.
tenqualitative
in such a manner a. How are the three descriptions similar? How
analyze qualitative data, and researchthat it attracts the reader’s
attention, gives a clear indication of the article’s are they different?
b. Why are the three descriptions not identical?
Questions for Review ( ) and Creative Application ( )
focus, and R highlights the significance of the find- A
How would you account for the differences?
ings. When completed, the instructor asks for three
The natural science model and qualitative
volunteers who are willing Research
to give a brief methods
verbal and epistemological
c. What is “interpretiveand omnipotence”? Do
research synopsis of their group’s study ontological
and have considerations
their any of the descriptions assume interpretive
R Under what circumstancesIntroduction
can some qualitative
read by the rest How
R of theclosely tied are research
class. Volun- omnipotence?
methods to epis-
research use a natural science
teersmodel?
with good keyboard skillstemological
are then asked and ontological
to d. positions?
Which of the descriptions is most realistic?
Explain.
A A qualitative researcher finds that many of the A You decide to do a secondary e. Can analysis
one everof quanti-
determine whether a particu-
transcribe the three Introductions into electronic
homeless people she encounters in her fieldwork tative data taken from the General lar Social Survey,
description is more realistic than another?
format so they can be shown on the classroom
have addiction issues. How might she use that find- Victimization Study. You have Does nothe
preconceived
term “realistic” even have a definite
screen. The first of the three volunteers then gives
ing to launch a quantitative study? theoretical position when youmeaning? begin, but decide to
a synopsis of their group’s study, and their Introduc-
see if there is any association between gender and
Quantitative research and tion is shown on the screen. Thefear
interpretivism instructor
of crime. fa- does f.your
thenHow Would the person
approach deviate
or persons depicted in the
cilitates a discussion of the Introduction. The other picture or clip describe the setting in a way
R Under what circumstances can some quantitative from a strictly positivist orientation to research?
two Introductions are presented in the same way. that differs from how the three volunteers de-
research exhibit characteristics of interpretivism? scribed it? How so? Why would the person or
A A quantitative researcher finds Inthat
an alternative
25 per cent version
of Problems withthat
of this exercise the quantitative/qualitative
persons depicted have a different description?
the people aged 18–25 in awouldnationalbe sample
appropriate contrast
have for small classes, the in-
g. How could you learn about how the people in
no intention of voting in thestructor collects
next federal all the Introductions
election, R Outline some thatofthethe ways in which the quantita-
class produced and provides written comments on the picture or clip view the setting?
while the figure for people aged 65 and over is tive/qualitative contrast is not as hard and fast as
only 10 per cent. Explain this
eachdifference, making
one, returning is often
them in a later supposed.
class. h. If multiple interpretations of a physical setting
reference to how people at different ages may per- A Explain how grounded theory are possible,could
methods wouldbeinterpretations of people’s
ceive the political process differently.
2. Prior If you
to class, were
the instructor used to
selects develop
a picture or aa theory of actions and experiences
the relationship be- in the setting be even
to write up your answer in a brief
research
videoreport,
clip ofwould tween body
a person or persons shaming and the more
in a physical use ofnumerous and variegated? Explain.
social media,
it be appropriate to describe it asthat
setting qualitative
would beinappropriate
then fordescribe how this sort
qualitative i. What are the can
of research implications
be of the responses to
nature? Explain. research—for example, a market, thought
a sportsof facility,
as a forma of theorythe testing.
previous question for qualitative research?
bry29440_ch14_323-345.indd
2013_
11/20/18 11:57 AM
each chapter.
Glossary
Terms in italic type are defined elsewhere in the Glossary. are broken down into component parts, which are then
Online resources
Textbooks today do not stand on their own: they are only the central elements in a complete learning and
teaching package. Social Research Methods is no exception. This fifth Canadian edition is supported by an
outstanding array of ancillary materials for both students and instructors.
Additional Materials
In addition to the above materials, OUP Canada offers these resources free to everyone using the textbook:
www.oupcanada.com/SocialResearch5e
Special Features of the Book xvii
For Students
Student Study Guide
A comprehensive online study guide provides detailed chapter summaries, learning objectives, lists of key
terms and concepts, self-assessment quizzes, and links to useful media resources.
For Instructors
Online Instructor’s Manual
This revised online resource includes comprehensive outlines of the text’s chapters, additional assignments,
classroom activities designed to encourage student engagement, and teaching aids that will enhance the
learning experience.
PowerPoint Slides
Hundreds of slides for classroom presentation—newly updated and enhanced for this edition—summarize key
points from each chapter and can be edited to suit individual instructors’ needs.
• Wider philosophical and methodological exposure to them will reinforce the textual
issues are discussed at some length in Chapter 1. material and may be useful for later work.
Instructors who do not wish to use this contextual • The quantitative/qualitative distinction is
material can largely ignore the chapter, except used in two ways: to organize the discussion of
for the section on formulating a research ques- research methods and data analysis, and to intro-
tion. Those who do want to emphasize issues of duce some wider philosophical issues that have
context should also see Chapter 14. a bearing on social research. Chapter 1 reviews
• Practical issues involved in doing quantitative the main areas of difference between quantita-
research are the subject of Part II. Chapter 2 is tive and qualitative research, while Chapter 14
a useful introduction to this topic because it explores ways of integrating the two. If time is a
maps out the main research designs used in both concern, the latter chapter can be skimmed.
quantitative and qualitative research. • Writing up research is as much a part of the
• Practical issues involved in doing qualitative research process as data collection and analysis.
research are the subject of Part III. Again, Chapter 15 discusses a variety of issues related
Chapter 2 outlines the most common designs. to writing and should be drawn to students’
• Data analysis is covered in Chapters 8 (quanti- attention even if it is not discussed in class.
tative) and 13 (qualitative), and a guide to the • Specific advice on doing a research project.
use of computer software for these purposes can As we have already noted, the whole book is
be found in the Appendix. There is an additional relevant to student projects, but Chapter 16
guide to using Excel for data analysis on the addresses this subject directly. If time is a factor,
accompanying Dashboard. Even if the module the Appendix to Part I briefly overviews the
is taught without actual computer applications, stages of research for students.
Acknowledgments
Normally, writing the Acknowledgments for a book I’ll miss Alan Bryman, but at the same time I’m
is a rather pleasant task. One can look back on all the sustained by knowing that working with him has
hard work that was done by a network of people, and changed my outlook on the world for the better.
reflect on one’s good fortune that the immense task And of course he touched many others as well. Alan
of producing a book was made possible by the kind- was given a very fitting tribute in the International
ness of others. I feel that gratitude now in writing Journal of Social Research Methodology (2018, 21[3],
this, but it is intermingled with a sense of sadness 267–274), which attested to the tremendous impact
because while this edition of the book was being pre- his work had on a broad range of people, in many
pared, Alan Bryman passed away. fields. As a further tribute, I’d like to dedicate this
Alan died on 20 July 2017. I was in Europe at the edition of the book to his memory.
time, biding my time between conferences. Even Others who contributed to this new edition in-
though we had never met in person, his death was still clude developmental editor Amy Gordon, who got
quite touching to me. We had exchanged emails and the ball rolling and always took an optimistic view
other electronic materials over the years, and he had of challenging circumstances. Her patience, intelli-
given me a standing invitation to visit him in England gence, organizational skills, and attention to detail
whenever I had the opportunity. In the communica- were invaluable. Leslie Saffrey did an amazing job as
tions we shared Alan was always friendly and collegial, copy editor. Her diligence and professionalism shine
and never hesitated to devote his considerable talents through on every page. Jayne Baker, University of
to whatever I asked of him. It is with considerable Toronto Mississauga; Meridith Burles, U niversity
regret that I never made my way to England to see him. of Saskatchewan; Alicia Horton, University of the
Still, as co-authors of this book a special kind of Fraser Valley; Neda Maghbouleh, University of
bond developed between us. Alan greatly expanded my Toronto; Heather Mair, University of Waterloo; Oral
knowledge of research methodology and of the social Robinson, University of British Columbia; and sev-
sciences in general. One thing about him that I greatly eral anonymous reviewers provided detailed feed-
admired was how he dealt with the tensions that some- back on this edition, and in so doing vastly improved
time arise between people taking different epistemo- the quality of the book and saved me from serious
logical, ontological, and methodological approaches errors and omissions. Thank you. To be sure, the
to social research. For Alan there was no blind loyalty shortcomings that remain are mine alone.
to a particular way of doing research; he simply took My family was a constant source of comfort and
the position that what really mattered was finding the inspiration as I worked on this manuscript, and they
best method or methods to address a particular re- continue to nurture me. Much is owed to my wife
search question. And he was a master at showing how Jennifer, to my children Ted and Angelica, and to
seemingly incompatible approaches actually had many Brooke.
things in common and could complement each other. Edward Bell
Preface
There are some answers to the “Why are you
Why Study Research Methods? here?” question that I seldom get from students.
At this point in your life you probably haven’t devoted Consider the following. The acquisition of know-
much thought to social research methods. In all like- ledge through research always goes beyond coming
lihood you are reading this book in conjunction with up with new ideas or getting more information or
the first methods course you have ever taken. That developing novel ways of looking at society, how-
means you are about to experience something new. ever important those things may be. To produce or
Since you are going to expend considerable time and acquire knowledge is a political act. It is tied in with
effort doing something you’ve never done before, it the exercise of power. Social research always has pol-
would make sense to pause for a moment and ask itical ramifications, because it always implies, subtly
yourself the following existential questions: Why am or not so subtly, that some ways of organizing society
I here? Why should I read this book? Why should I are better than others. Some researchers are happy
study social research methods? What’s the point of it to leave their political involvement at that level: an
all? Pondering those questions will make what is to implication in their work that certain social struc-
come much more meaningful for you. tures or practices should change or be preserved.
So why are you here? When I ask my students Others go so far as to engage in advocacy, taking part
that question (usually on the first day of classes) in public campaigns to persuade governments and
they often tell me that a solid background in social the public at large that some sort of action should be
research methods is indispensable if they are to de- taken. Whichever position the researcher takes, the
velop a sophisticated understanding of the topics politics of research cannot be avoided. What varies is
they are passionate about. how loudly the trumpet is sounded, which is a matter
What are you passionate about? Does it matter to for the individual researcher to decide. Debates about
you that literally billions of human beings struggle to the wisdom of social researchers becoming advocates
survive on less than $2.50 per day, while others are are part of the storied history of the social sciences,
so wealthy that they can’t relate to ordinary people? some of which will be told in this book.
Do you have an interest in gender equity, environ- We often hear that “knowledge is power,” and in
mental sustainability, crime, single motherhood, many ways that is true. Dictators know this best;
ethnic tension, labour relations, racism, the living they all try to limit access to knowledge in order to
conditions of Indigenous people, changing notions preserve their control over others. In fact, the power-
of the family, sexual mores? In order to come up with ful in any society may try to inhibit free inquiry
informed, thoughtful analyses of these and other and the flow of ideas. But the “knowledge is power”
social issues, and to be capable of evaluating the maxim leaves several questions unanswered. What
claims made by others on these topics, it is crucial is knowledge? How can it be acquired? How are we
to be familiar with the various research methodol- to tell the difference between a sound idea and one
ogies used in the social sciences to investigate these that should be ignored? How can we gather informa-
subjects. Just as learning how clothes are made can tion that will help us understand our subject matter,
help you tell the difference between a good pair of and what sorts of information should we seek? How
jeans and a shoddy pair, understanding where social should that information be analyzed and evaluated?
knowledge comes from will help you distinguish be- Are there some things that we will never fully under-
tween valid claims and fatuous ones. stand, regardless of how hard we try?
xx Preface
This book explores a variety of answers to those takes in ideas and evaluates them, but also as some-
questions, which will make it somewhat different one who projects ideas onto the world. Start think-
from your other readings and textbooks. In most ing of yourself as a researcher: someone who has the
courses in the social sciences, students spend their potential to examine the human condition and con-
time absorbing, and sometimes challenging, claims tribute to our understanding of it. That’s far more
made about the social world. For example, you may interesting and impactful than just reading other
learn that men are more likely to commit violent people’s work, although reading is an essential part
crimes than women. You may come to understand of the process.
what it feels like to be homeless. Or perhaps you’ll When can you start being a researcher? At vari-
encounter a theory explaining why some countries ous points in your university career you will prob-
are rich and powerful while others are not. That ably be called on to do some small research projects.
sort of endeavour—learning about and reflecting The knowledge you will gain by learning about social
on claims to knowledge—is clearly worthwhile, but research methods can put you at an advantage with
it will not be our central concern. Instead, we will that sort of assignment, and it could even be an
focus on how those types of claims are generated entrée to more sophisticated work if you decide to go
and how they gain acceptance as knowledge. In on and develop your research capabilities in gradu-
other words, instead of exploring the body of facts, ate school or in some other way.
concepts, theories, and interpretations that social Although it is important to acknowledge that
scientists use to make sense of the world, we’ll look social research and its attendant methodologies
at how all those things are acquired or created. One have political ramifications, we shouldn’t lose sight
of the most profound questions that can be asked of of the fact that it is the height of folly to pursue one’s
someone making a claim to knowledge is, “How do political goals with bad social science. To reduce
you know that?” That simple question is at the heart suffering and promote human flourishing we need
of this book. And like so many simple questions, it rigorous social research, which is dependent on
has no simple answer. sound research methods. Learning how to take
To produce knowledge, we need a m ethodology— good field notes, getting instruction on how to con-
a way to get it. And if you accept the premise that struct effective survey questions, using appropriate
knowledge is power, then learning about research sampling procedures, and a host of other methodo-
methods means learning about how to become logical issues are not trivial matters. In fact, doing
powerful, or at least more powerful. You could even those sorts of things properly are prerequisites to
say that “methodology is power,” and you wouldn’t acquiring the knowledge needed to achieve one’s
be far off the mark. broader social goals. Without good methodolo-
How could methodology be power? For one thing, gies we have very little to offer that is not available
if you know how knowledge is generated, you won’t elsewhere.
be fooled as easily as someone who doesn’t know. If you want to learn about how knowledge is cre-
Another way to put this is that learning about re- ated in the social sciences, or if you aspire to make an
search methodologies is an excellent way to enhance impact on the world by doing social research your-
your critical thinking skills—skills that are essential self, reading this book is a good way to start. I hope
in any situation where power is exercised, which is it will at least pique your interest in how social re-
everywhere. searchers do what they do, and what the implications
Another way in which methodology can be power of social research are. To paraphrase Gore Vidal: the
is that it can help to make you a producer of ideas. more one learns, the more interesting consciousness
Start thinking of yourself not only as someone who becomes.
The first two chapters of the text will acquaint
you with a number of fundamental concepts
that feature prominently throughout the
book, in particular the notion of a general re-
search orientation and the idea of a research
design. In Chapter 1, two general orientations
are identified—the quantitative and qualita-
tive approaches—along with a variety of con-
siderations that affect the practice of social
research. Chapter 2 presents various kinds of
research design and identifies some criteria
used to evaluate social research. Together
these chapters provide the basic conceptual
building blocks that we will return to later in
the book. Chapter 3 provides a discussion of
research ethics, which are a primary concern
for every kind of social research and affect
every stage of the inquiry. Part I concludes
with an appendix that provides an overview
of the research process.
PART I
Fundamental Issues in Social Research
© kendo_OK/Shutterstock.com
1
General Research Orientations
Chapter Overview
The aim of this chapter is to examine the fundamental assumptions upon which social research is based.
An important distinction commonly drawn by practitioners of social research—between the quantitative
and qualitative approaches—is explored in relation to those considerations. We will consider:
• the relationship between theory and research—in particular, whether theories and the hypotheses
derived from them are tested by gathering data (a deductive approach) or whether data gathering
is used as a means to create theory (an inductive approach);
• epistemological issues, such as whether a natural science model like the one used in chemistry
or biology is suitable for the study of the social world;
• ontological issues, such as whether the social world should be regarded as a reality external to
individuals over which they have little or no control, or as something that social actors may fashion
into their personal realities;
• how values and practical issues impinge on the research process; and
• how these issues relate to both quantitative and qualitative research; a preliminary discussion,
followed up in Chapter 14, suggests that although the quantitative and qualitative orientations are
different, they complement each other.
Soon-Yi wants to find out why Indigenous people in The list of topics she could collect information on
Canada are more likely to live in poverty than other seems endless.
Canadians, but doesn’t know where to begin. Should Maybe rather than beginning her study by accumu-
she start by examining the history of colonialism and lating data, it would it be better to start out with some
conflict between Indigenous peoples and settler- hunches and then gather information to see whether
colonizers, such as disputes over land claims and treat- they are supported by evidence. For example, perhaps
ies, residential schools, or anti-Indigenous prejudice? the discrepancy in economic conditions is a manifesta-
Or how about gathering aggregate data on present tion of a centuries-old system of international domin-
conditions like residence patterns, e conomic activ- ance and exploitation. Similarly, it could have arisen
ities, the age structure, or educational trajectories? through a clash of civilizations and cultures. Then there
▲ LeonWang/Shutterstock
1 | General Research Orientations 3
is the question of how to gather the information to many social researchers put it, is reality “out there,” in-
address those theories. Should she consult mainly his- dependent of the person who perceives it, or is it so-
torical sources? How about living in an Indigenous com- cially constructed?
munity for a while and recording her interactions and Then there are the broader questions: What, if any-
observations? Or would it be better to stay where she is thing, can be done about the issue of poverty among
and select a few individuals for interviews in which they Indigenous Canadians? Can society be changed so that
could talk about their life stories? Or maybe she should Indigenous poverty is eliminated, or at least reduced?
design a detailed questionnaire on prejudice and send Or are the social forces that affect our lives so strong
it out to large random samples of Indigenous and non- that we have no choice but to conform to them, like
Indigenous people? leaves in a windstorm?
Soon-Yi wonders about the ethical implications of A great many social thinkers and researchers have
each of these approaches. She also puzzles over what confronted the same issues that Soon-Yi is pondering.
she should do if it turns out that Indigenous people In fact there are a number of rich intellectual traditions
take positions on her research topic that differ from that can be drawn on to make sense of these matters
her own. Should she try to determine whose perspec- and to assist people in making the kinds of methodo-
tive better reflects reality, and if so, how would she do logical and practical decisions that all researchers have
that? Can those sorts of issues ever be resolved? Are to make. In this chapter you will be introduced to that
there “right” and “wrong” answers on such matters? As body of ideas and the debates surrounding them.
hearing about the personal histories of her family 15–25 decreases, the crime rate also decreases.”
and friends. Relational statements come in two forms:
Regardless of the motivation for doing research, a. deterministic, which means the two vari-
the data gathered are usually viewed in relation to ables always go together in a particular way;
theories of some kind. That’s because theories are an for example, “as the number of people aged
attempt to “make sense of it all,” to find order and 15–25 decreases by 1 per cent, the crime
meaning in a seemingly infinite mass of informa- rate also shrinks by 1 per cent.” If research
tion. How is that done? uncovers an instance in which the variables
are not related in this way, the relational
statement must be modified.
Theory and research b. probabilistic, which means the two vari-
The connection between theory and research is ables go together with some degree of regu-
not straightforward. There are several issues at larity, but the relationship is not inevitable;
stake here, but two stand out: first, the form of the for example, “regions of the country where
theory; and second, the relationship between data the number of people aged 15–25 is decreas-
and theory. ing are more likely to experience a decline
in the crime rate than regions in which the
Degree of abstraction number of people in that age category is
The term “theory” is used in a variety of ways, but its increasing.” Here, finding a case that does
most common meaning is an explanation of observed not fit the pattern does not mean that the
regularities or patterns. For example, one could try to theory must be modified; this could simply
come up with a theory to explain why schizophre- be one of the times when the variables are
nia is more common in the working class than in not related in the usual way.
the middle class, or why more men than women are
alcoholics. Theories are composed of interrelated There are different types of theories. One distinc-
and usually verifiable statements or propositions. tion that is sometimes made is between theories of
The statements and propositions come in varying the middle range (Merton, 1967) and grand theories.
forms, and different types may be combined in the The former are more limited in scope, and can be
same theory. Here are three common components of tested directly by gathering empirical evidence. For
a theory: instance, Durkheim’s (1897/1952) theory of suicide,
which maintains that suicide is a function of the level
1. Definitions specify what the key terms in the of social integration, is a theory of the middle range.
theory mean; for example, “Attachments are One way to test it would be to compare suicide rates
stable bonds between people who are fond of for married people with those for single, divorced,
each other.” or widowed individuals. Merton’s (1938) anomie
2. Descriptions outline the characteristics of the theory, which suggests that crime is more common
phenomena of interest; for example, “Attach- when a society instills a desire for wealth in everyone
ments to parents and other family members but provides insufficient means for all to achieve it,
tend to decline in adolescence. Later, often is another theory of the middle range. Such theories
after age 25, new attachments develop as a represent attempts to understand and explain a lim-
person marries or lives with a romantic part- ited aspect of social life.
ner, becomes a parent, or gets a steady job.” Grand theories, by contrast, are general and
3. Relational statements connect two or more abstract. They include theories such as structural-
variables, so that knowing the value of one functionalism, symbolic interactionism, critical
variable conveys information about the other; theory, post-structuralism, feminism, and so on.
for example, “As the proportion of people aged Grand theories generally offer few direct indications
1 | General Research Orientations 5
of how to collect evidence to test them, but they pro- how an abstract theory like Giddens’s structuration
vide ways of looking at the world that can be the theory (1984) can be applied to a specific situa
inspiration for a wide variety of research programs. tion and yield some important insights.
For example, standpoint theory was developed by Although theory plays a crucial role in the social
Dorothy Smith (2004; 2005) and others from a gen- sciences, not all studies make reference to it. For ex-
eral feminist perspective. This theory maintains that ample, some qualitative writers focus on providing
the way we view the world and make our way in it is a rich description of the experiences of a group of
largely determined by our placement in various hier- people without trying to come up with a compre-
archies of status and power. One offshoot of stand- hensive theory that would explain those experiences.
point theory has been an increased willingness to Nonetheless, as Box 1.2 shows, social scientists are
examine the views and perceptions of poor or mar- often under pressure from their peers to relate their
ginalized groups in society. Similarly, Box 1.1 shows work to theories of some kind.
As Box 1.2 indicates, some social scientists will
reject research that has no direct connection to
theory in either the grand or the middle-range
sense of the term. However, non-theoretical work
BOX 1.1 Grand theory and social can provide insights that are useful or revealing in
research their own right. McKeganey and Barnard’s (1996)
research on British sex workers and their clients is
Giddens’s (1984) structuration theory attempts a case in point. The authors related their research
to bridge the gulf between notions of structure findings to investigations of people in the sex
and agency in social life. This theoretical issue industry in several other countries, and what they
is explored in empirical research by Dinovitzer describe offers good illustrations of ideas that form
et al. (2003) on the educational attainment of an important part of the sociologist’s conceptual
immigrant youth. The specific focus of their re- toolkit. Although it is not possible to tell whether
search was suburban Toronto immigrants in the the authors had the concept in mind when they col-
years 1976 to 1995. The data were quantitative, lected their data, their book offers real-life examples
generated through structured interviews. The of Goffman’s (1963) notion of “stigma” and the way
goal of the researchers was to tease out the stigmatized individuals, in this case sex workers and
relative influence on educational attainment their clients, manage a spoiled identity. Their analy-
of structural variables (such as class, gender, sis also sheds light on Hochschild’s (1983) concept
and the youths’ family background) and indi- of “emotional labour,” a term she coined to refer to
vidual variables over which the young people what flight attendants do when they feign friendli-
had more control (such as studying and cutting ness in order to deal with difficult passengers. Sim-
classes). ilarly, several other non-theoretical studies provide
The authors found that educational achieve- data that could eventually be used to evaluate or
ment is affected by two structural variables devise a theory.
(gender and father’s occupation) and one indi- Our discussion of what theory is and its im-
vidual factor (intellectual investment), and that portance invites consideration of another question:
bilingual ESL students do better than young im- What is the relationship between theory and re-
migrants for whom English is a first language. search? Up to this point we have focused primarily
They are not brighter, nor do they work harder, on how theory can guide research, in particular on
but they have greater parental supervision, how the collection and analysis of data can be used
and—perhaps partly as a consequence—they to test theories. But this notion of research as essen-
plan better. tially “theory testing” does not provide a complete
picture of what social scientists do. Theory may also
6 Part I Fundamental Issues in Social Research
sunflowerey/123RF
it art. What social theories might you invoke to ex-
plain what this person is doing? How might the young
person explain it? Do you think his explanation would
have anything in common with your social theories?
follow upon or arise from the collection and analysis Note that the researcher starts by coming up with
of data. Here we begin to see two different ways to go a theory that seeks to explain a particular phenom-
about acquiring knowledge: the deductive and induc- enon, and then deduces specific hypotheses from it
tive approaches. that are tested with empirical data (see Box 1.3 for
a concrete example) and then are either confirmed
or rejected. In the case of the research described in
Deductive and inductive Box 1.3, which examined several theories simul-
approaches taneously, some hypotheses were supported while
The deductive method is the most common ap- others were not. If the data gathered do not support
proach to social research. The sequence of steps the researchers’ hypotheses, the theory may have to
taken in deductive research is depicted in Figure 1.1. be revised or rejected.
1 | General Research Orientations 7
Guenter Guni/iStockphoto
has occurred in which fewer and fewer people
embrace religious beliefs and practices, a pro-
cess that became especially marked after the
middle of the twentieth century. To be sure,
religion has not disappeared in Western coun-
tries, including Canada. In fact about 26 per cent Veiled women in Iran, a society that has experienced
of Canadians attend religious services once a far less secularization than many Western countries.
month or more frequently (Statistics Canada, Do you think that some of the theories examined in
2013). Nonetheless, several indicators suggest Hay’s (2014) research can help to explain the high
that the proportion of Canadians who are firmly levels of religiosity found in Iran? For example, could
the acceptance of religious beliefs and practices
religious is considerably lower than it was sev-
there be related to cultural factors that favour so-
eral decades ago. What can account for this pro-
cietal cohesion over personal autonomy? Could low
found social change?
levels of religious pluralism be a factor?
Hay (2014) has taken a deductive approach
to this issue. As in all deductive research, his
goal was “theory testing.” He started by propos- indicators of factors that were relevant to each
ing some theories, derived testable hypotheses of the theories.
from them, and then set out to determine which, There was limited support for deprivation-
if any, of the hypotheses were supported by the compensation theory, but there was strong sup-
data. He reviewed seven well-known theories of port for three other theoretical positions. One
secularization, and deduced specific hypotheses theory that gained support involves the idea that
from each one. For instance, the deprivation- a growing acceptance of personal autonomy in
compensation theory holds that the existential spiritual and moral matters has led to increased
security that affluent people in developed soci- levels of secularization. A second position that
eties enjoy through health care and education, was consistent with the results holds that the re-
long life expectancy, and the absence of military ligious pluralism found in wealthy countries cre-
conflict shields many of us from the traumas that ates a situation in which it is very difficult for a
draw people toward religion. The hypothesis he particular faith group to make the claim that it is
derived from this theory was that people who the “one true religion,” which casts doubt on all re-
have experienced real tragedy in their lives are ligious doctrines and so increases levels of secu-
more likely to be religious than people who have larization. A third theory that was substantiated
no direct exposure to such things. Hay then ana- by the data maintains that increasing acceptance
lyzed contemporary survey data that included of human evolution and the belief that observable
measures of religious beliefs and practices, phenomena are purely material or physical in
as well as a number of variables that provided nature has reduced levels of religiosity in Canada.
1 | General Research Orientations 9
number of people in same-sex marriages in this knowledge. A fundamental debate in this context
country increased from 42,030 in 2011 to 48,740 in is whether the social sciences should follow the
2016 (Statistics Canada, 2017), which a researcher same principles and procedures as the natural sci-
could use in the context of a discussion of the ences. Three broad positions on these matters have
changing forms of the family in Canada. emerged: the positivist, interpretivist, and critical
The next section examines some epistemological approaches to social science.
issues that affect the conduct of social research.
Positivism
One epistemological position that affirms the im-
Epistemological portance of following the natural sciences is pos-
considerations itivism. Although definitions of the term vary,
Those who do social research base their work on a positivism is generally taken to entail the following:
number of epistemological assumptions—notions
of what can be known and how knowledge can be 1. Only phenomena confirmed by the senses
acquired. A related epistemological issue pertains to (sight, hearing, etc.) can be accepted as know-
the question of what should be regarded as acceptable ledge: this is the principle of empiricism.
10 Part I Fundamental Issues in Social Research
Ideas must be subjected to the rigours of em- philosophers of science and social science differ
pirical testing before they can be considered quite sharply over how best to characterize scientific
knowledge. practice, and since the early 1960s positivism has ac-
2. A key purpose of theory is to generate hypoth- quired some negative connotations. One reason for
eses that can be tested and thereby allow ex- dissatisfaction was the fact that certain researchers
planations of observed laws and principles to in the positivist tradition ignored some fundamen-
be assessed (deduction). tal differences between human beings and the often
3. Knowledge can also be arrived at through the inanimate or not fully conscious entities studied
gathering of facts that provide the basis for by natural scientists. Unlike subatomic particles or
generalizations or laws (induction). plants, for example, we humans have thoughts, feel-
4. Science must (and presumably can) be “value- ings, and values—perhaps even some capacity for
free.” That is, it must be conducted in such a volition. Those aspects of human behaviour were
way that different researchers, given the same often not addressed in the leading positivistic theor-
data, will always reach the same conclusions, ies of the day. The positivist idea that science can or
no matter how different their values might be. should be value-free was also a source of dissatisfac-
In the past this “value-free” quality was called tion. Critics pointed out that “neutral” social scien-
“objectivity”; today it is more likely to be tists often took moral positions on social issues, at
called intersubjectivity. least implicitly. For example, theories implying that
5. There is a clear distinction between scientific social equilibrium or harmony is normal seemed to
statements, which describe how and why cer- suggest that social change is not needed or not de-
tain social phenomena operate the way they sirable. Critical social science, discussed below, even
do, and normative statements, which outline went so far as to claim that it is the duty of the re-
whether certain acts or social conditions are searcher to help bring about social change in order
morally acceptable. Only scientific statements to create a better world.
have a place in the domain of science; norma-
tive statements belong in the realm of philoso- Interpretivism
phy or religion. This idea is implied by point Interpretivism to some extent grew out of the epis-
1 above, because the truth of moral claims temological critique of positivism, and provides an
cannot be confirmed by the senses. alternative to the sort of social science typically done
by positivists. Interpretive researchers maintain that
It is possible to see in these five points a link with it is the role of social scientists to grasp the subjective
some of the issues already raised about the relation- meanings of people’s actions. They make the point
ship between theory and research. Positivism as- that people act on the basis of the meanings that they
sumes a fairly sharp distinction between theory and attribute to their acts and to the acts of others. Indi-
research and includes elements of both deduction viduals use their own common-sense constructs to
and induction. One role of research is to test theor- interpret the reality of their daily lives, and it is those
ies and gather the information necessary for the de- interpretations that motivate their behaviour.
velopment of scientific laws. Positivism also implies Interpretivists claim that it is the job of the social
that it is possible to collect observations without any scientist to gain access to the “common-sense think-
reference to pre-existing theories, and to develop ing” of the people they study and hence to understand
new theories purely on the basis of those observa- people’s actions and their social world from the point
tions. Finally, theories and propositions not directly of view of the actors. Thus any thoughts constructed
testable through empirical observation are often not by the social scientist to grasp this social reality must
considered to be genuinely scientific. be founded on the common-sense interpretations
A common mistake is to treat positivism as syn- of those they study: people living their daily lives
onymous with science and the scientific. In fact, within their own social world (Schutz, 1962, p. 59).
1 | General Research Orientations 11
In order to gain access to other people’s perspectives, 1994). In research terms, according to Blumer (1962,
interpretivists commonly immerse themselves in the p. 188), “the position of symbolic interaction re-
social environments frequented by the people they quires the student to catch the process of interpreta-
study, or at least conduct lengthy interviews with tion through which [actors] construct their actions.”
them. The insights gained by the researcher com- Taking an interpretative stance can result in
monly follow from the information derived in this surprising findings, or at least findings that appear
way, so the process used tends to be inductive. surprising if the researcher’s position is outside
Many interpretive social scientists argue that the the particular social context being studied. Box 1.4
subject matter of the social sciences—people, groups, provides an interesting example.
and institutions—is fundamentally different from
that of the natural sciences. For them it follows that
the study of the social world requires a different logic
and research procedure, one that reflects what they
see as the distinctiveness of humans as against other BOX 1.4 Interpretivism in practice
living things or inanimate objects. This clash reflects
a division between the positivist emphasis on the ex- Foster (1995) conducted ethnographic re-
planation of human behaviour and society, and the search using participant observation and semi-
interpretivist preference for an empathetic under- structured interviews in a housing estate in East
standing and interpretation of human existence. This London referred to as Riverside, a residential
division, which precedes the emergence of modern complex experiencing a high level of crime ac-
social science, finds expression in Max Weber’s cording to official statistics. However, she found
(1864–1920) notion of Verstehen (which means “em- that residents did not perceive the estate to be
pathetic understanding”). Weber described soci- a high-crime area; nor were they overly anxious
ology as a “science which attempts the interpretive about becoming victims of crime. Those percep-
understanding of social action in order to arrive at tions could be attributed to a number of factors,
a causal explanation of its course and effects” (1947, but a particularly important one was “informal
p. 88). Weber’s definition seems to embrace both ex- social control,” which was used in conjunction
planation and understanding, but the crucial point with more formal methods such as policing.
is that the task of “causal explanation” is undertaken People expected a certain level of crime, but
with reference to the “interpretive understanding felt fairly secure because informal social control
of social action.” This is a different emphasis from a worked to keep the level of crime contained.
more Durkheimian view in which the external forces Informal social control has several aspects.
that affect behaviour may not be perceived by those One is that neighbours often look out for each
involved, or at least may have no meaning for them. other. In the words of one of Foster’s inter-
Symbolic interactionism is an example of a viewees: “If I hear a bang or shouting I go out. If
sociological perspective that falls under the heading there’s aggravation I come in and ring the police.
of interpretivism. The ideas of the founders of sym- I don’t stand for it” (Foster, 1995, p. 575). Another
bolic interactionism—in particular George Herbert aspect of informal social control is that people
Mead (1863–1931), who maintained that the individ- often feel secure because they know each other.
ual’s self-concept emerges through an appreciation of A second respondent said: “I don’t feel nervous
the perceptions of others—have been hotly debated. . . . because people do generally know each
Symbolic interactionists argue that interaction takes other. We keep an eye on each other’s proper-
place in such a way that individuals are continually ties. . . . I feel quite safe because you know your
interpreting the symbolic meaning of their environ- neighbours and you know they’re there . . . they
ment (including the actions of others) and acting look out for you” (Foster, 1995, p. 575).
on the basis of that imputed meaning (cf. Collins,
12 Part I Fundamental Issues in Social Research
Of course, as the example in Box 1.4 suggests, reality) into a Klasse für sich (a class for itself, one
when social scientists adopt an interpretive stance, with an awareness of its exploitation).
they are not simply revealing how members of a Critical social scientists also believe that research
social group interpret the world around them. The should be action oriented. It should involve praxis:
social scientist almost certainly aims to place those putting one’s theoretical and academic positions into
interpretations into a social scientific framework. practice. The idea of praxis is contained in Marx’s
Thus there is a double interpretation going on: the famous dictum that “philosophers have only inter-
researcher is interpreting others’ interpretations. preted the world in various ways; the point, how-
Indeed, there is even a third level of interpretation, ever, is to change it [emphasis in original]” (Marx,
because the researcher’s interpretations have to be 1845/1998, p. 574). This exhortation applies not only
further interpreted in terms of the concepts, theor- to positivism, but also to interpretive perspectives on
ies, and literature of the social sciences. Thus, in social life.
Box 1.4 the idea that Riverside was not perceived as A critical perspective has been adopted by a wide
a high-crime area by residents was Foster’s interpret- variety of scholars, including those engaged in fem-
ation of her subjects’ interpretations. She then had inist, post-colonial, anti-poverty, anti-racism, and
the additional job of placing her findings into a social queer studies. Another example is participatory
scientific framework, which she accomplished by re- action research, which is discussed in Box 1.5 and
lating them to existing concepts and discussions in examined again in later chapters. Institutional eth-
criminology: concepts such as informal social con- nography and critical discourse analysis, examined
trol, neighbourhood watch schemes, and the role of in Chapters 10 and 12, respectively, are further ex-
housing style as a possible cause of criminal activity. amples of critical social science.
The aim of this section has been to outline how
Critical approaches to social science epistemological considerations are related to re-
Like interpretivism, critical social science developed search practice. As noted, a key concern is whether
in part as a reaction to positivism. Social scientists a natural science or positivist approach can supply
who adopt a critical approach use a diversity of re- legitimate knowledge of the social world. We’ve seen
search methods, including those used by positiv- that positivists often use the deductive method, al-
ists and interpretivists, and may use a deductive though they concede that it is possible to generate
or inductive approach. But they disagree with the theories using induction. By contrast, we’ve seen that
positivist notion that researchers should take a interpretive researchers typically adopt an inductive
value-neutral stance regarding their subject matter. strategy, and that critical social scientists may use
In fact they maintain that research and knowledge either one. Another important issue involves the
should not be considered as ends in themselves, but purpose of social research, in particular whether
as means to be used to rid the world of suffering and social scientists, in both their academic work and
oppression (Neuman, 2003). their roles as citizens, should be actively engaged in
Marxists, for example, argue that those who own issues of social justice.
the means of production deceive, constrain, and It is important not to overstate the connections
exploit the weak. The masses could be free if social between epistemological issues and research prac-
scientists, by asking embarrassing questions and tice: they represent tendencies only. Particular epis-
making pointed arguments, would uncover exploit- temological principles and research practices do not
ation, expose hypocrisy, and reveal to the general necessarily go hand in hand in a neat, unambiguous
populace the nature and extent of their oppression. manner. Often hybrid approaches are taken, which
This would transform the masses from what Marx combine different positions and approaches. This
called a Klasse an sich (a class in itself, an objective point will be made again on several occasions.
1 | General Research Orientations 13
The origins of participatory action research (some- Participants first reflect on a social problem and
times referred to as “action research,” “emancipa- how it might be studied and rectified, and then use
tory research,” or “participatory research”) are the results of that process as the basis for social
found in the immediate post–Second World War action. Once action has been taken, participants
era in the work of Lewin (1946) and others who discuss the consequences of the action and how
sought to bring social science to bear directly on it might be used to generate more knowledge and
specific social problems. In the early 1970s, PAR inform further action. In this way a social situa-
began to take on a more explicitly activist orien- tion is improved by the people directly affected
tation when people in Latin America, Africa, Asia, by it, who work in collaboration with others. The
and elsewhere became disillusioned with the abil- process has the potential to transform the partici-
ity of conventional social science to improve social pants personally, intellectually, and even spirit-
conditions in their countries. These individuals ex- ually. Debbink and Ornelas (1997), for instance,
panded the methodology of PAR such that local describe how some Alberta dairy farmers worked
people affected by the issue at hand acted as part- together with Mexican campesinos (poor farm-
ners and co-equals with professional social sci- ers) and an activist intellectual to bring donated
entists and government officials. All three groups Alberta cattle to a poor, rural Mexican commun-
worked together to devise research questions, ity. Similarly, Frisby et al. (1997) outline how some
choose research methods, gather and analyze low-income women in British Columbia teamed
data, and undertake a course of action to amelior- up with social scientists and local recreation dir-
ate the problem (Smith, 1997; Frisby et al., 1997). ectors to improve poor women’s access to physical
Of all the approaches to social science, PAR is activity services. In both cases, people directly af-
the one most closely associated with social and fected by a problem and researchers from outside
political activism. In fact working for social jus- the community interacted as equals to produce
tice is built into the methodology of PAR itself. knowledge and address a social issue.
Lantz and Booth (1998) examined media treat- women’s lifestyles cause breast cancer was con-
ment of the apparent rise in the incidence of breast structed as a social fact by popular-magazine
cancer that began in the early 1980s and found writers.
that its depiction as epidemic can be treated as Similarly, Hallgrimsdottir et al. (2006, p. 266)
a social construction. They analyzed a variety of argue that the media “contributes to constructing,
popular magazines and noted that many of the reproducing and deepening the social stigmas
articles drew attention to the lifestyles of modern associated with working in the sex industry.” The
women, such as delaying first births and having authors compared media depictions of sex-trade
careers. The authors also argued that the articles workers in Victoria, British Columbia, from 1980 to
ascribe blame: “Women are portrayed as victims 2004 with accounts provided by street sex workers,
of an insidious disease, but also as victims of their escorts, and others in this sector. Whereas the media
own behaviours, many of which are related to the portrayed such people as wicked and blameworthy
control of their own fertility” (1998, p. 915). in the earlier years, and as exploited, trapped, and
This article concludes that, as a social category, innocent girls more recently, the sex workers them-
the breast cancer epidemic was represented in selves interpreted their work lives very differently.
popular magazines in a particular way—one that Although there was considerable heterogeneity in
blamed the victims and their lifestyles, particu- their experiences and attitudes toward their trade,
larly in the case of young women. Yet in fact fewer many saw sex work as a largely mundane matter
than 20 per cent of cases of breast cancer were of earning a living. Because they are marginalized
in women under the age of 50. Lantz and Booth’s and stigmatized, however, their voices are seldom
study is fairly representative of a constructionist heard; most people have their ideas about the sex
ontology in suggesting that the idea that young trade constructed for them by the media.
Box 1.6 provides two illustrations of the less strin- organization has rules and regulations and adopts
gent form of constructionism. standardized procedures for getting things done.
A second debate revolves around these questions: A division of labour assigns people to different jobs.
Is social reality akin to the physical world as most There is a hierarchy of authority, a mission state-
people see it—largely fixed and “out there,” some- ment, and so on. Objectivists see any organization
thing that individuals and groups have to confront as possessing a reality external to any of the specific
but over which they have little or no control, like individuals who inhabit it; they may leave, but it will
a snowstorm? Or is social reality not necessarily stay. Moreover, the organization represents a social
pre-existing and fixed, but rather created through order in that it exerts pressure on individuals to con-
our actions? A “yes” to the first question indicates form to organizational requirements. People learn
support for a variant of objectivism, while a “yes” to and apply the rules and regulations and follow the
the second affirms a kind of constructionism. In this standardized procedures. They do the jobs to which
chapter we’ll focus mainly on this second debate: they are appointed. If they do not do these things,
whether social reality can be created. they may be reprimanded or even fired. The organ-
Some social scientists suggest that social phe- ization is therefore a constraining force that acts on
nomena confront individuals as external facts and inhibits its members. To a large extent, this is
beyond their reach or influence. For example, an the “classic” way of conceptualizing an organization.
1 | General Research Orientations 15
An alternative ontological position challenges can negotiate their roles to the point where they are
the suggestion that things such as organizations are allowed to operate on patients. But not all writers
external realities confronting social actors who have adopting a constructionist position are similarly pre-
limited power to influence or change them. Strauss pared to acknowledge the existence or importance of
and colleagues (1973), for example, carried out an objective reality. It is precisely this split between
research in a psychiatric hospital and proposed that seeing the social world as an objective reality and
its organization was best conceptualized as one of seeing it as a subjective reality in a continuous state
“negotiated order.” Instead of viewing order as a pre- of flux that Giddens sought to straddle in formulat-
existing characteristic, they argued that it is worked ing his idea of structuration (recall Box 1.1).
at and created to some extent, and that the rules are The constructionist perspective that maintains
far less extensive and less rigorously imposed than that social reality can be negotiated also suggests
might be supposed from an objectivist account of that the concepts people employ to help them
organizations. understand the natural and social world are social
Indeed, Strauss and colleagues saw rules more as products whose meaning is constructed in and
general understandings than as commands (1973, through social interaction. For example, a concept
p. 308). Precisely because relatively little of the ac- such as “masculinity” is treated as a social construc-
tivity of doctors, nurses, and other personnel is spe- tion. This implies that masculinity is not a distinct,
cifically set down or prescribed, the social order of timeless, and universal entity, but something whose
a hospital is an outcome of agreed-upon patterns of meaning is built up through interaction. That mean-
action that are themselves the products of negotia- ing is likely to be ephemeral, in that it will vary over
tions among the different parties involved. For in- time and place. The construction of one’s masculin-
stance, the official rules may say that only a doctor ity may even have to be “recuperated” from time to
can increase medication; however, some nurses are time, such as when Canadian politicians endeavour
routinely given this power, even though this is never to restore their public gender identities (Maiolino,
actually stated in the regulations. The social order is 2015). This sort of social construction can be seen
in a constant state of change because the hospital is particularly well in discourse analysis, examined in
“a place where numerous agreements are continually Chapter 12. As Potter (1996, p. 98) observed: “The
being terminated or forgotten, but also are continu- world . . . is constituted [emphasis in original] in
ally being established, renewed, reviewed, revoked, one way or another as people talk it, write it, and
[and] revised. . . . In any pragmatic sense, this is the argue it”. This sense of constructionism frequently
hospital at the moment [emphasis added]: this is its sparks an interest in how social phenomena are
social order” (Strauss et al., 1973, pp. 316–317). The represented.
authors argued that a preoccupation with the formal
properties of organizations (rules, organizational Relationship to social research
charts, regulations, and roles) makes it hard to rec- Questions of social ontology cannot be divorced from
ognize the degree to which order in organizations issues concerning the conduct of social research.
has to be established in everyday interaction. This Ontological assumptions and commitments affect
informal organization arises because there cannot be both the way research questions are formulated and
rules for every possible contingency, and because the the way the research is carried out. For example, a
existing rules are sometimes problematic. However, researcher who sees organizations as objective social
this is not to say that the formal properties of organ- entities that act on individuals is likely to emphasize
izations have no effect on individual action. their formal properties. Alternatively, a researcher
Although Strauss and colleagues stressed the who is interested in the negotiated, changing nature
active role of individuals in the social construction of organizations is likely to focus on the active in-
of reality, they did not push the argument to an ex- volvement of people in reality construction. In each
treme. For example, they did not claim that nurses case, a different research design is required.
16 Part I Fundamental Issues in Social Research
TABLE 1.1 | F
undamental differences between quantitative and qualitative
research strategies
Quantitative Qualitative
Role of theory in research Mainly deductive; testing of theory Mainly inductive; generation of theory
Epistemological orientation Natural science model; positivism Interpretivism
Ontological orientation Objectivism Constructionism
all points in the process of social research, includ- Researchers are increasingly prepared to forewarn
ing the following: readers of their biases and assumptions and to ex-
plain how these may have influenced their findings.
• choice of research area Since the mid-1970s many researchers have pub-
• formulation of the research question lished “insider” accounts of what doing research is
• choice of method really like, as opposed to the generalities presented
• formulation of the research design and data in social research methods textbooks (like this one).
collection techniques These accounts frequently function as “confessions”
• data collection of personal biases and reveal the pride that research-
• analysis of data ers take in telling readers how open they are in re-
• interpretation of data vealing them.
• conclusions Still another approach is to argue for con-
sciously value-laden research. Some writers on social
There are, therefore, numerous points at which research celebrate what Mies (1993, p. 68) called
bias and the intrusion of values can occur during the a “conscious partiality.” For example, Tastsoglou
course of research. For example, researchers may de- and Miedema (2003) clearly adopted a feminist,
velop affection or sympathy for the people they are anti-racist a pproach in studying immigrant women
studying. It is quite common for researchers who in the Maritimes. A similar perspective allowed Pratt
spend a great deal of time with the people they study and Valverde (2002) to describe a large Canadian
to become so close to their subjects that they find it newspaper as a “notorious tabloid,” “obsessed” with
difficult to disentangle their role as social scientists what it called bogus refugees. It is also exemplified in
from their concern for the participants’ well-being. Hallgrimsdottir et al.’s (2006) condemnation of the
Equally, social scientists may be repelled by the media’s role in stigmatizing sex workers in British
people they are studying. In his research into an Columbia. In fact, some feminist researchers would
African society known as the Ik, a social anthropol- consider it inappropriate (as well as difficult) to do
ogist was appalled by what he saw as their cruelty research on women in an objective, value-neutral
toward each other (Turnbull, 1973). Although he way because that would be incompatible with their
was able to identify the desperate social and political values. Instead, many feminist researchers argue for
conditions that they faced, it was clear that he had research that exposes the conditions of women’s dis-
misgivings about what he witnessed, particularly advantage in a male-dominated society, as Demaiter
during his early time with them. and Adams (2009) did in their study of women in
One way of dealing with the problem of values and the IT sector, featured in the Methods in Motion box
bias is to recognize that research cannot be value- on page 19. Some feminist writers argue that only
free, and to try to ensure that values in the r esearch research on women intended for women is consistent
process are acknowledged and made explicit. This with women’s wider political needs.
is part of a larger process of reflexivity or self- The significance of feminism in relation to values
reflection that researchers are encouraged to carry goes further than this, however. Several feminist
out. As Turnbull (1973, p. 13) put it at the beginning social researchers in the early 1980s proposed that
of his book on the Ik: quantitative research is incompatible with feminist
ideals. For writers such as Oakley (1981), quantitative
The reader is entitled to know something of research is bound up with the male value of control,
the aims, expectations, hopes, and attitudes as seen in the researcher’s control both of the research
that the writer brought to the field [in his case, participant and of the research context and situation.
Western values about the family], for these will Moreover, the research process is seen as a one-way
surely influence not only how he sees things affair in which researchers extract information from
but even what he sees. their subjects and give little if anything in return.
1 | General Research Orientations 19
For many feminists, such a strategy borders on in society; some social scientists may favour
exploitation and is incompatible with the value
increased government intervention in eco-
that feminism attributes to sisterhood and non- nomic affairs, while others defend the free
hierarchical relationships. market; sociologists and political scientists in
The antipathy toward quantitative research re- Quebec may be split between sovereigntists
sulted in a preference for qualitative research among and federalists.
certain feminists. Qualitative research was seen not • A related issue involves research funding.
only as more consistent with the values of femin- Much social research is funded by organiz-
ism, but as more adaptable to those values. Feminist ations such as private firms or government
qualitative research came to be associated with re- departments that may have a vested interest
fusing to take a value-neutral approach and relating in the outcomes of the research. The very fact
to the people under study as human beings rather that these organizations fund some research
than research instruments. This stance demonstrates projects but not others opens the door to pol-
how strongly values can affect the process of social itical influence. Such organizations may seek
investigation. In more recent years, however, femin- to invest in studies that will be useful to them
ist attitudes toward quantitative research have been or supportive of their operations and world
softening, especially when it is employed in conjunc- views. They will often call for researchers to
tion with qualitative research (Oakley, 1998). This tender bids for an investigation in a certain
issue will be revisited in Chapters 5 and 14. area. When social researchers take part in
There are, then, different positions that can be such exercises, they enter a political arena,
taken in relation to values and value-free research. since their research may be designed to please
Few writers today believe it is possible to be truly the funding body. As a result, as Hughes
objective. There is a greater awareness of the limits (2000) observed in relation to research in the
of objectivity, and some of the more categorical pro- field of crime, an investigation of gun crimes
nouncements on the subject, such as Durkheim’s, among the “underclass” is more likely to re-
have fallen into disfavour. At the same time, giving ceive funding than one concerned with mis-
free rein to one’s political beliefs and value positions demeanours committed by agents of the state.
can be problematic. Researchers today still have to Morgan (2000) pointed out that research
fight the all-too-human propensity to demonize funded by government is typically empirical
those whose values are different from their own, and and quantitative; it tends to be concerned with
they still struggle with the temptation to summarily the short-term costs and benefits of a particu-
reject research findings when the researcher’s ideo- lar policy or innovation; and it is generally
logical or moral positions are not compatible with uncritical in the sense that the underlying
their own. government policies are not questioned: all
the government wants to know about is the ef-
Politics in social research fectiveness of their implementation. Political
At various places in this book we take the position issues often arise when the funding agency
that social research has political implications. Here itself is trying to secure a continuous stream
are some examples of the ways in which social re- of government funding.
search may be political: • Gaining access to research subjects can also be
a political process, especially in the case of or-
• Social researchers often take sides. There are ganizations. Access to organizations is usually
many ways this can happen. To consider just mediated by gatekeepers concerned not only
a few: feminist researchers may focus on the about the researcher’s motives but also about
disadvantages that women face and the possi- what the organization stands to gain from the
bilities for improving the position of women investigation, what it will lose by participating
1 | General Research Orientations 21
in terms of staff time and other costs, and the participate nevertheless reported annoyance
potential risks to its image. Often, gatekeepers with being asked about sexual orientation.
seek to influence how the investigation will • There may be pressure to restrict the publica-
take place: what kinds of questions can be tion of findings. Hughes (2000) cited a study
asked; who can and cannot be a focus of study; of plea bargaining in the British criminal jus-
the amount of time to be spent with each re- tice system as a case in point. The researchers
search participant; the interpretation of the had uncovered what were deemed at the time
findings; and the form the reports will take, to be disconcerting levels of informal plea bar-
even to the point of asking to approve drafts. gaining, and they concluded that the formal
• Public institutions, such as police departments, judicial process was being weakened. The
schools, and hospitals, as well as most com- English legal establishment sought to thwart
mercial firms, are concerned with how they the dissemination of the findings and was
are going to be represented in publications. persuaded to allow publication only when a
Consequently, gaining access is almost always panel of academics confirmed the validity of
a matter of negotiation and as such inevitably the findings. Similarly, the editors of academic
turns into a political process. The product of journals may refuse to publish pieces that do
such negotiations is often referred to as “the not conform to their own ideological or polit-
research bargain,” and in many cases there is ical preferences.
more than one bargain that has to be struck.
Once in the organization, researchers often This is just a small handful of the ways in which pol-
discover layers of gatekeepers. For example, itics intrudes in the research process.
let’s say a provincial government grants a re-
search team permission to talk to the boys in Issues related to the research
a group home. Before the research can begin, question
the head of the home has to be brought onside, Some important determinants of how research is
then the staff, and then the actual adolescents. conducted follow from the research question one is
Frequently, one of the staff is then given the trying to answer. Specifically, the choice of research
responsibility of dealing with the fieldwork- orientation, design, or method has to match the
ers. A suspicion that the researchers are really specific research question being investigated. For in-
working for management then has to be over- stance, if a researcher is interested in measuring the
come. And it is unwise to assume that simply impact of various possible causes of a social phenom-
because gatekeepers have granted access, enon, a quantitative strategy is probably appropriate.
a smooth passage will ensue in subsequent Alternatively, if the focus is on the world views of
dealings with the people to be studied. Per- members of a certain social group and how those
haps the most powerful of the boys will turn views develop, a qualitative research strategy—one
out to be the key gatekeeper. Researchers may sensitive to the way participants interpret their social
find themselves used as pawns if subgroups at- world—may be the way to go. If a researcher is in-
tempt to enlist them in advancing a particular terested in a topic on which little or no research has
goal. Research participants who doubt the util- been done, a quantitative strategy may be difficult to
ity of social research may even try to obstruct employ because there is little prior literature from
the research process. For example, in Beagan’s which to draw leads about possible causes. A more
(2001) study of “everyday inequalities” among qualitative, exploratory approach may be preferable
Canadian medical students, some of the
because that type of investigation is typically associ-
students refused to complete a survey when ated with the generation of theory rather than theory
they heard she was asking about the treatment testing (see Table 1.1) and with a relatively unstruc-
of gays and lesbians, and one student who did tured approach to the research process.
22 Part I Fundamental Issues in Social Research
A related dimension that has important im- takes may differ depending on whether the study is
plications for how the research will be conducted quantitative or qualitative in orientation. Quanti-
involves the nature of the topic and of the people tative questions usually ask whether a particular
being investigated. A researcher who wants to study variable or set of variables has an influence on the
individuals involved in illicit activities—for ex- phenomenon of interest. Duncan’s (2010) question,
ample, price fixing, shoplifting, or drug dealing— for example, reveals that her study was designed to
may find it difficult to develop the rapport with inquire into whether personal income affects the
them that is needed to conduct a social survey. It is likelihood that one will join a voluntary association,
not surprising, therefore, that researchers in these and if so, whether that relationship varies depending
areas tend to use a qualitative strategy. On the other on the neighbourhood in which one lives. Implied
hand, it’s unlikely that the hypotheses in Box 1.3 on in her questions is a certain causal model that may
secularization could have been tested using a quali- apply to some larger population, one that extends
tative approach. beyond the people studied. This is the deductive
method at work.
Formulating a research question Qualitative research questions tend to be less
Creating a research question is like picking a destin- specific, and are generally not designed to evaluate
ation for a hike: where you will start out, the route causal models of this kind. They tend to be “open-
you will take, and what you will experience along ended, evolving, and non-directional” (Creswell,
the way are largely determined by the endpoint you 2013, p. 138). The reason for this is that qualitative
are trying to reach. What you stand to accomplish researchers usually want to allow multiple interpret-
with a particular research study and how you will ations and perspectives to emerge from the people
accomplish it are profoundly affected by the goal you and settings they are examining. Also, since qualita-
have in mind, which is expressed in your research tive social scientists often use the inductive method,
question. they don’t have a fully preconceived notion of what
A research question states the purpose of the they expect to find, making it pointless to ask a
study in the form of a question. This is useful because specific question at the beginning of a study. Flex-
a question can often be more evocative and stimulat- ibility in exploring whatever one encounters in the
ing than a simple declarative statement. A question course of doing the research is essential if one is to
arouses curiosity and challenges the researcher to uncover the meanings and interpretations held by
find ways to answer it. Here are some examples of the people studied, and the processes by which they
research questions taken from a quantitative study of develop. For instance, Charpentier and Quéniart’s
voluntary associations in Canada: “Does individual (2017) research question, “What does it mean to
income level affect whether an individual is a vol- older women to grow old in a land of immigration?”
untary association member? Does this effect differ was purposely open-ended because their goal was to
between neighbourhoods?” (Duncan, 2010, p. 578). give the women they studied a voice, not to march
Here are two research questions taken from a quali- them through specific topics and issues that they as
tative study of older immigrant women in Quebec: researchers may have considered to be important.
“What does it mean to older women to grow old in a The data they collected allowed them to discern cer-
land of immigration? What are the effects of immi- tain themes and patterns of response offered up by
gration on their living conditions, especially when the participants in their study, which is the goal of
they have immigrated after 50?” (Charpentier & the inductive method.
Quéniart, 2017, p. 437). Quantitative research may start with the choice
The process of formulating and assessing research of a general area of interest: for example, male
questions is something of an art, but here are some homosexuality. At this stage, a very general research
general thoughts. As you may have noticed in the re- question might be “How do people in general feel
search questions quoted above, the form the question about gays?” This broad research area would have to
1 | General Research Orientations 23
be narrowed down: for example, to “How does the do the people under study deal with those who vio-
Canadian adult population react to the portrayal late group norms? In cases where people are pun-
of gays in television dramas?” But even that is too ished for their violations, how do the offenders make
broad, so the next level of specification might be sense of the punishment they receive, and how does
something like “Do young straight men react dif- it affect how they feel about the other members of
ferently than young straight women to the portrayal the group?”
of gay male romance on television program X? If No single study can answer all the research ques-
so, do those differences reflect a more fluid sexual tions that will occur to the researcher. Only a small
identity on the part of young women as compared number of them can be selected. The need to narrow
to young men?” Such questions could be linked to the topic is not only a matter of the time available
larger theories of sex roles, sexual orientation acqui- and the cost of doing research. It is also a reflection
sition, socialization, and the tolerance of difference of the need for a clear focus.
in society. As suggested above, the research questions may
Research questions in qualitative inquiries may change as the study progresses, for a number of rea-
become more specific over time too, but that usually sons. The discovery of a new data source may change
occurs primarily after the researcher has started the focus a bit, as might some of the initial findings.
to gather data. A general research question may be For instance, if in the study of male homosexual-
retained throughout the study, but more specific ity the researchers find that having a gay relative
sub-questions normally evolve as the research un- in the immediate family makes a large difference
folds. For example, ethnographic research may begin in people’s attitudes, the research question and the
with a general question such as “What is going on attendant methodology and theoretical orientation
here?” (Wolcott, 2008, p. 73), but as the researcher may be revised. The research question may also
learns more about the people to be analyzed, more change because of limitations in time and other
specific questions usually come to mind. These ques- resources available to the researcher. Box 1.7 offers
tions could be something like the following: “How some tips on developing research questions.
A good research question will: can make a contribution to the existing know-
• be as clear as possible so that it will be under- ledge on the topic;
standable to others; • be linked to the other research questions in the
• be researchable: it must allow for the develop- study, so that you can develop a single argument
ment of a research design and the collection of or at least a set of related arguments; this is hard
data; this means that extremely abstract terms to do with unrelated research questions; and
may not be suitable; • be neither too broad (no research project can
• relate in some way to existing studies that sug- do justice to all aspects of a topic) nor too
gest how your question may be approached. narrow (unable to make a meaningful contribu-
Even with a topic that has not been widely re- tion to an area of study).
searched, there will probably be some relevant If you are unsure about how to formulate re-
literature (for example, on related or parallel search questions (or about other aspects of research),
topics). Establishing connections with existing look at journal articles or research monographs to
studies will help to show how your research see how other researchers have handled them.
24 Part I Fundamental Issues in Social Research
Research questions set realistic boundaries for re- • the literature search;
search. A poorly formulated research question can • decisions about the kind of research design to
result in unfocused and substandard research. No employ;
matter how well designed a questionnaire is or how • decisions about what data to collect and from
skilled qualitative interviewers are, clear research whom;
questions are required to avoid going off in unneces- • the analysis of the data; and
sary directions and tangents. Research questions are • the writing up of the findings.
crucial because they guide:
Key Points
• Quantitative and qualitative research constitute • Theory can either precede research and data
different approaches to social investigation and gathering (the deductive method) or emerge out
carry with them important epistemological and of it (induction).
ontological assumptions. • Feminist researchers in the past have tended to
• Epistemological considerations loom large in the prefer a qualitative approach, although this situa-
choice of a research strategy. To a great extent, tion is changing now.
the issues revolve around the advantages and • Values can impinge on the research process in
disadvantages associated with the natural science various ways; as a consequence, research often
(in particular positivist), interpretivist, and critical has political dimensions.
approaches to science. • The political dimensions of research relate to the ex-
• Ontological considerations, such as objectivism ercise of power at different stages of an investigation.
versus constructionism, also constitute import- • Issues related to the research question can also
ant dimensions of the quantitative/qualitative affect decisions about research methods. Clear
contrast. research questions improve the chances of success.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.