Social

Research
Methods
Alan Bryman
third edition
OXFORD
UNIVERSITY PRESS
---_.-"


OXFORD
UNIVERSITY PRESS
,
GreatClarendonStreet,OxfordoX26DP
OxfordUniversityPressisadepartmentoftheUniversityofOxford.
Itfurthers theUniversity'sobjectiveofexcellenceinresearch,scholarship,
andeducationbypublishingworldwidein
Oxford NewYork
Auckland CapeTown DaresSalaam HongKong Karachi
KualaLumpur Madrid Melbourne MexicoCity Nairobi
NewDelhi Shanghai Taipei Toronto
Withofficesin
Argentina Austria Brazil Chile CzechRepublic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
SouthKorea Switzerland Thailand Turkey Ukraine Viemam
Oxfordisaregisteredtrade markofOxfordUniversityPress
intheUKandincertainothercountries
Published intheUnitedStates
byOxfordUniversityPressInc.,NewYork
AlanBryman2008
Themoralrightsoftheauthorhavebeenasserted
Database rightOxfordUniversityPress(maker)
Firstedition2001
Secondedition2004
Thisedition 2008
Allrightsreserved.Nopanofthispublicationmaybereproduced,
storedinaretrievalsystem,ortransmitted,inanyformorbyanymeans,
withoutthepriorpermission inwritingofOxfordUniversityPress,
orasexpresslypermittedbylaw,orundertermsagreedwith theappropriate
reprographicsrightsorganization.Enquiriesconcerningreproduction
outside thescopeofthe aboveshouldbesenttotheRightsDepartment.
OxfordUniversityPress,attheaddressabove
Youmustnotcirculate thisbookinanyotherbindingorcover
andyoumustimposethesameconditiononanyacquirer
BritishlibraryCataloguinginPublicationData
Dataavailable
libraryofCongressCataloginginPublicationData
Dataavailable
libraryofCongressCataloging-in-PublicationData
Bryman,Alan.
Socialresearchmethods/ AlanBryman.- 3rded.
p.cm.
Textaccompaniedbyacompanionwebsite.
ISBN-13:978-0-19-920295-9
1. Socialsciences-Research. 2. Socialsciences-Methodology. I. Title.
H62.B7872008
300.72-dc22
2008003361
TypesetbyGraphicraftlimited,HongKong
Printed inItaly
onacid-freepaperby
L.E.G.O.S.p.A.- Lavis (TN)
ISBN978-0-19-920295-9
1 3 5 7 9 10 8 6 4 2
____1_

Social research
strategies
Chapter outline
Introduction 4
Theoryandresearch 6
Whattype oftheory? 6
Deductiveandinductivetheory 9
Epistemological considerations 13
Anaturalscienceepistemology:positivism 14
Interpretivism 15
Ontologicalconsiderations 18
Objectivism 18
Constructionism 19
Relationshipto socialresearch 21
Researchstrategy:quantitativeandqualitativeresearch 21
Influencesonthe conductofsocialresearch 24
Values 24
Practicalconsiderations 26
Key points 27
Questions for review 28
--__ L




-
4 Social researchstrategies , ' .. :.._.:
r
o
Chapter guide
Thechiefaimofthischapteristoshowthatavarietyofconsiderationsenterintothe processofdoing
socialresearch.Thedistinctionthatiscommonlydrawn amongwritersonand practitionersofsocial
research betweenquantitativeresearchand qualitativeresearch isexplored inrelationtothese
considerations. Thischapterexplores:
the natureofthe relationshipbetweentheoryand research, inparticularwhethertheory guides
research (knownasadeductiveapproach)orwhethertheory isanoutcomeofresearch(knownas
aninductiveapproach);
epistemologicalissues-that is, ones todo with what is regarded as appropriate knowledge about
the socialworld;oneofthe mostcrucialaspectsisthe question ofwhetherornotanaturalscience
modeloftheresearch processissuitable forthe studyofthesocialworld;
ontologicalissues-thatis,onestodowithwhetherthe socialworldisregarded assomething
external tosocialactors orassomethingthatpeople are inthe processoffashioning;
the waysinwhichtheseissuesrelatetothe widelyuseddistinctioninthe socialsciencesbetween
twotypesofresearchstrategy:quantitativeand qualitativeresearch;thereisalsoapreliminary
discussion,whichwill befollowedupinChapter24,thatsuggeststhat, whilequantitativeand
qualitative research representdifferentapproachestosocialresearch, weshouldbewaryofdriving
awedge betweenthem;
the waysinwhichvaluesandpracticalissuesalsoimpingeonthe socialresearch process.

Introduction
This book isaboutsocial research. It attempts to equip
people who have some knowledgeofthe socialsciences
with an appreciation of how social research should be
conductedand whatitentails.Thelatterproject involves
situating social research in the context of sociology,
which intummeans attendingtothe questionofitsrole
in the overall enterprise of the discipline. It would be
mucheasierto'cuttothechase'andexplorethenatureof
methods of social research and provide advice on how
best to choose between and implement them. After all,
manypeoplemightexpectabookwiththetideofthepre-
sent one tobe concernedmainlywith the waysinwhich
the different methods in the social researcher's arsenal
canbeemployed.
Butthe practice ofsocialresearchdoes not exist in a
bubble, hermetically sealed off fromthesocial sciences and
thevariousintellectualallegiancesthattheirpractitioners
hold.Twopointsareofparticularrelevance here.
First,methodsofsocialresearcharecloselytied todif-
ferent visions of how social reality should be studied.
Methodsarenotsimplyneutraltools:theyarelinkedwith
the ways inwhich social scientists envision the connec-
tion between different viewpoints about the nature of
socialrealityand how itshould beexamined. However,
itispossible to overstatethis point. While methodsare
not neutral,they are not entirelysuffused with intellec-
tual inclinations either. Secondly, there isthe question
ofhow researchmethodsand practice connectwith the
wider social scientific enterprise. Research data are
invariablycollected inrelationtosomething.The'some-
thing' maybeaburningsocialproblemor,more usually,
atheory.
Thisisnot tosuggest that researchisentirelydictated
by theoretical concerns. One sometimes finds simple
'fact-finding' exercises published. Fenton et al. (1998)
conducted a quantitative content analysis of social




I
ng
1S
e
g
if-
:I.
:h
)f
- .,
e
research reported in the British mass media. They exam-
ined national and regional newspapers, television and
radio, and also magazines. They admit that one of the
main reasons for conducting the research was to establish
the amount and types of research that are represented.
sometimes, such exercises are motivated by a concern
about a pressing social problem. McKeganeyand Barnard
(1996) conducted qualitative research involving observa-
tion and interviews with prostitutes and their clients in
Glasgow. One factor that seems to have prompted this
research was the concern about the role of prostitutes in
spreading HN infection (McKeganeyand Barnard 1996:
3). Another scenario occurs when research is done on a
topic when a specific opportunity arises. The interest of
Westergaard et al. (1989) in the effects of redundancy
seems to have been profoundly motivated by the oppor-
tunity that arose when a Sheffield steel company, which
was close to their institutional base at the University of
Sheffield, made a large number of people redundant. The
finn's management approached the authors a year after
the redundancies to conduct research on what had hap-
pened to the individuals who had been made redundant.
The authors conducted social survey research using a
structured interview approach on most of those made
Student experience
redundant. Of course, the authors were influenced by
theories about and previous research on unemployment,

but the specific impetus for the research on the effects of

redundancy was not planned. Yet another stimulus for
research can arise out of personal experiences. Lofland
and Lofland (1995) note that many research publications
emerge out of the researcher's personal biography, such
as Zukin's (1982) interest in loft living arising out of her
living in a loft in New York City. Another example is
O'Reilly's (2002) investigation of British expatriates
living on the Costa del Sol in Spain, which stemmed
from her and her partner's dream of moving to the area
themselves, which in fact they eventually did. Certainly,
my own interest in Disneytheme parks can be traced back
to a visit to Disney World in Florida in 1991 (Bryrnan
1995, 1999), while my interest in the representation of
social science research in the mass media (Fenton et al.
1998) can almost certainly be attributed to a wounding
experience with the press reported in Haslam and
Bryrnan (1994).
By and large, however, research data achieve signific-
ance in sociology when viewed in relation to theoretical
concerns. This raises the issue of the nature of the rela-
tionship between theory and research.
Personal experience as a basis for
research interests
For her research, IsabellaRobbins was interested inthe ways in which mothers frame decisions regarding
vaccinationsfor their children. Thistopic had a particular significance for her. She writes:
Asthe mother ofthree childrenI haveencountered some tough decisions regarding responsibility towards
mychildren. Reading sociology, as a mature student, gave methe toolsto helpunderstand myworldand to
contextualizesomeofthe dilemmas I hadfaced. Inparticular, I had experienceda difficult decision regarding
the vaccination status of mychildren.
To readmoreaboutIsabella's research experiences, go to the Online Resource Centre that accompanies
thisbookat http://www.oxfordtextbooks.co.uk/orc/brymansrm3el.




Theoryand research

Characterizing the nature of the link between theory and

research is by no means a straightforward matter. There
are several issues at stake here, but two stand out in par-
ticular. First, there is the question of what form of theory
one is talking about. Secondly, there is the matter of
whether data are collected to test or to build theories.
Theory is important to the social researcher because it
provides a backcloth and rationale for the research that is
being conducted. It also provides a framework within
which social phenomena can be understood and the
research findings can be interpreted.
Whattypeoftheory?
The term 'theory' is used in a variety of different ways, but
its most common meaning is as an explanation of observed
regularities, for example, why sufferers of schizophrenia
are more likely to come from working-class than middle-
class backgrounds, or why work alienation varies by tech-
nology. But such theories tend not to be the stuff of
courses in sociological theory, which typically focus much
more on theories with a higher level of abstraction. Ex-
amples of such theories include structural-functionalism,
symbolic interactionism, critical theory, poststructural-
ism, srructuration theory, and so on. What we see here is
a distinction between theories of the former type, which
are often called theories of the middlerange(Merton 1967),
..

Research in focus 1.1
.
~ .
and grand theories, which operate at a more abstract and
general level. According to Merton, grand theories offer
few indications to researchers as to how they might guide
or influence the collection of empirical evidence. So, if
someone wanted to test a theory or to draw an inference
from it that could be tested, the level of abstractness is
likely to be so great that the researcher would find it
difficult to make the necessary links with the real world.
There is a paradox here, of course. Even highly abstract
ideas, such as Parsons's notions of 'pattern variables' and
'functional requisites', must have some connection with
an external reality, in that they are likely to have been
generated out of Parsons's reading of research or his
reflections upon that reality or others' writings on it.
However, the level of abstractness of the theorizing is so
great as to make it difficult for them to be deployed in
research. For research purposes, then, Merton argues
that grand theories are of limited use in connection with
social research, although, as the example in Research in
focus 1.1 suggests, an abstract concept like social capital
(Bourdieu 1984) can have some pay-off in research terms.
Instead, middle-range theories are 'intermediate to gen-
eral theories of social systems which are too remote from
particular classes of social behavior, organization and
change to account for what is observed and to those
detailed orderly descriptions of particulars that are not
generalized at all' (Merton 1967: 39).
J'
Grandtheoryandsocialresearch
Butlerand Robson(2001) used Bourdieu'sconceptof social capital as a means of understandinggentrification
of areas of London. Whilethe term 'social capital' has acquired an everyday usage, Butlerand Robson follow
Bourdieu'stheoretical use of it, whichdraws attention to the social connectedness and the interpersonal
resourcesthat those with social capital can drawon to pursue their goals. Whilethe term has attracted the
interest of social policy researchers and others concerned with social exclusion, its use inrelationto the middle
classhas been less prominent, accordingto Butlerand Robson. Bourdieu'streatment implies that those with
social capital cultivatesignificant social connections and then drawupon those connections as resources for
their goals. Butlerand Robson conducted semi-structuredinterviewswith 'gentrifiers' ineach of three inner
Londonareas. Ofthe three areas, Telegraph Hill was the strongest interms of social capital.According to the
l.




ictand
soffer
.guide
So, if
~ r e n c e
less is
ind it
vorld,
stract
" and
with
been
,r his
III it.
is so
sd in
gues
with
.h in
pital
nns.
gen-
rom
and
lose
not
n
e
authors, thisisrevealed in'its higher levels ofvoluntary co-operation and senseofgeographically focused
unity' (Butler and Robson 2001: 2159). It isthe recourse to these networks ofsociality that accounts forthe
,
successful gentrification ofTelegraph Hill. Battersea, one ofthe other twoareas,entails a contrasting
impetus for gentrification inBourdieu's terms. Here, economic capital wasmoresignificant forgentrification
than the social capital that wasimportant inTelegraph Hill. The roleofeconomic capital inBattersea canbe
seen inthe 'competitive accessto an increasingly desirable and expensive stockofhousing and an exclusive
circuit ofschooling centredon private provision' (Butler and Robson 2001: 2159). In the former, it issociality
that provides the motor forgentrification, whereasinBattersea it isthrough market forces and isonly partially
influenced bypatternsofsocial connectedness. This studyisan interesting example ofthe wayinwhich a
relatively high-level theoretical notion-social capital and its kindred conceptofeconomic capital-associated
witha social theoristcan be employed to illuminate research questions concerning the dynamics of modern
urban living.
Byand large, then, it is not grand theory that typically
guides social research. Middle-range theories are much
more likely to be the focus of empirical enquiry. In fact,
Mertonformulated the idea as a means of bridging what
he saw as a growing gulf between theory (in the sense of
grand theory) and empirical findings. This is not to say
that there were no middle-range theories before he
wrote: there definitely were, but what Merton did was
to seek to clarify what is meant by 'theory' when social
scientists write about the relationship between theory
and research.
Middle-rangetheories, unlike grand ones, operate in a
limiteddomain, whether it isjuvenile delinquency, racial
prejudice, educational attainment, or the labour process
(see Research in focus 1.2). They vary somewhat in their
range of application. For example, labelling theory repre-
sents a middle-range theory in the sociologyof deviance.
Its exponents sought to understand deviance in terms of
the causes and effectsof the societal reaction to deviation.
It was held to be applicable to a variety of different forms
of deviance, including crime and mental illness. By
contrast, Cloward and Ohlin's (1960) differential associ-
ation theory was formulated specifically in connection
with juvenile delinquency, and in subsequent years this
tended to be its focus. Middle-range theories, then, fall
somewhere between grand theories and empirical find-
ings. They represent attempts to understand and explain
a limited aspect of social life.
Research in focus 1.2
Labour process theory: a middle-range theory
In the sociology ofwork, labourprocess theorycan be regarded as a middle-range theory. The publication of
LabarandMonopolyCapital(Braverman 1974) inaugurated a streamofthinking and research aroundthe idea
ofthe labourprocess and in particular on the degreeto which there has been an inexorable trendtowards
increasing control overthe manual worker and the deskilling ofmanual labour. Aconference volume of
much ofthis work waspublished as LabourProcess Theory (Knights and Willmott 1990). P.Thompson (1989)
describes the theoryas having fourelements: the principle that the labour process entails the extraction of
surplus value; the need for capitalist enterprises constantly to transform production processes; the quest for
control overlabour: and the essential conflict betweencapital and labour. Labour process theoryhas been
the focus ofconsiderable empirical research (e.g, Knights et 01. 1985).
- - - - - - - - - - ~ - - - - - - - - - - - - -




Even the grand/middle-range distinction does not
entirely clarify the issues involved in asking the decep-
tively simple question of what is theory?'. This is because
the term 'theory' is frequently used in a manner that
means little more than the background literature in an
area of social enquiry. To a certain extent, this point can
be taken to apply to fact-finding exercises such as those
referred to above. The analysis of the representation of
social research in the media by Fenton et al. (1998) was
undertaken against a'background of similar analyses in
the USA and of studies of the representation of natural
science research in the media in several different coun-
tries, In many cases, the relevant background literature
relating to a topic fuels the focus of an article or book and
thereby acts as the equivalent of a theory, as with the
research referred to in Research in focus 1.3. The litera-
I
ture in a certain domain acts as the spur to an enquiry.
The literature acts as an impetus in a number of ways: the
researcher may seek to resolve an inconsistency between
different findings or between different interpretations of
findings; the researcher may have spotted a neglected
aspect of a topic; certain ideas may not previously have
been tested a great deal (such as the notion that recycling
may be gendered, which prompted the investigation in
Research in focus 1.3); the researcher may feel that exist-
ing approaches being used for research on a topic are
deficient, and so provides an alternative approach; and
soon.
Research in focus 1.3
Background literature as theory: recycling and
the domestic division of labour
Oates and McDonald(2006)conducted a postal questionnaire surveyof households in Sheffield that had
agreed to participate in a trial whereby they were given blue 'wheelie bins' in which paper waste could be
deposited, The survey was conducted three years after the scheme had been launched. The questionnaire
was primarily concerned with the matter of who within households is involved in recycling activities. A total
of 1,532 questionnaires was mailedout to participants inthe trial.The researchers received back 469 usable
questionnaires. which comprised 31 per cent of all those sent out. They expected that recycling activitywould
be highlygendered, in a manner similar to other domestic tasks. The authors found 'a clear gender difference
in[the] data, as the domestic division of labour literature might predict. However,there isalsoa significant
amount of joint activitywhichwas not anticipated.... [Men] are more likely to participate in recycling as part
of a joint initiative and activitythan they are to do it alone.' (Oates and McDonald(2006: 427)
People sometimes suggest that researchers find what they expect to find. Thisisan interesting example of a
set offindings that are onlypartiallyintune with the researchers' expectations! However.the main point to
register isthat it isthe literature on the domestic division of labour that forms the starting point ofthis study,
as the quotation above indicates. Interestingly, in the article in which this research is reported, much ofthis
literature is referred to ina section withthe subtitle Theories and trends inthe domestic division of labour',
Social scientists are sometimes prone to being some- not been preoccupied with theory. Such research is con-
what dismissive of research that has no obvious connec- ditioned by and directed towards research questions that
tions with theory-in either the grand or middle-range arise out of an interrogation of the literature. The data
senses of the term. Such research is often dismissed as collection and analysis are subsequently geared to the
naive empiricism (see Key concept 1.1). It would be illumination or resolution of the research issue or prob-
harsh, not to say inaccurate, to brand as naive empiricism lem that has been identified at the outset. The literature
the numerous studies in which the publications-as-theory acts as a proxy for theory. In many instances, theory is
strategy is employed, simply because their authors have latent or implicit in the literature.



,
Key concept 1.1
What is empiricism?
The term 'empiricism' is used ina number of differentways, but two stand out. First, it is usedto denote a
general approach to the studyof realitythat suggeststhat onlyknowledge gainedthrough experience and the
senses isacceptable. Inother words,this positionmeans that ideas must be subjected to the rigours of testing
beforethey can be consideredknowledge. The second meaningof the term isrelated to this and refersto a
beliefthat the accumulation of 'facts' isa legitimate goal in itsown right. It isthis second meaningthat is
I
sometimesreferredto as 'naiveempiricism'.
Indeed, research that appears to have the character-
istics of the fact-finding exercise should not be prema-
turely dismissed as naive empiricism either. McKeganey
and Barnard's (1996) research on prostitutes and their
clients is a case in point. On the face of it, even if one strips
away the concern with HN infection, the research could
be construed as naive empiricism and perhaps of a rather
prurient kind. However, this again would be a harsh and
probably inaccuratejudgement. For example, the authors
relate their research findings to the literature report-
ing other investigations of prostitutes in a number of
different countries. They also illuminate their findings
by drawing on ideas that are very much part of the socio-
logist's conceptual tool kit. One example is Goffman's
(1963) notion of 'stigma' and the way in which the stig-
matized individual seeks to manage a spoiled identity;
another is Hochschild's (1983) concept of 'emotional
labour', a term she coined to denote the way in which
airline flight attendants need to express positive emotions
as part of the requirements for their jobs. In doing so,
they contrive a demeanour of friendliness when dealing
with passengers, some of whom may be extremely
difficult.
It is not possible to tell from McKeganey and Barnard's
(1996) report whether the concepts of stigma and emo-
tionallabour influenced their data collection. However,
raising this question invites consideration of another
question: in so far as any piece of research is linked to the-
ory, what was the role of that theory? Up to this point,
I have tended to write as though theory is something that
guides and influences the collection and analysis of data.
In other words, research is done in order to answer ques-
tions posed by theoretical considerations. But an alterna-
tive position is to view theory as something that occurs
after the collection and analysis of some or all of the
data associated with a project. We begin to see here the
significance of a second factor in considering the relation-
ship between theory and research-whether we are refer-
ring to deductive or inductive theory.
Deductive and inductive theory
Deductive theory represents the commonest view of the
nature of the relationship between theory and social
research. The researcher, on the basis of what is known
about in a particular domain and of theoretical consider-
ations in relation to that domain, deduces a hypothesis
(or hypotheses) that must then be subjected to empirical
scrutiny. Embedded within the hypothesis will be con-
cepts that will need to be translated into researchable
entities. The social scientist must both skilfully deduce a
hypothesis and then translate it into operational terms.
This means that the social scientist needs to specify how
data can be collected in relation to the concepts that make
up the hypothesis.
This view of the role of theory in relation to research is
very much the kind of role that Merton had in mind in
connection with middle-range theory, which, he argued,
'is principally used in sociology to guide empirical inquiry'
(Merton 1967: 39). Theory and the hypothesis deduced
from it come first and drive the process of gathering data
(see Research in focus 1.4 for an example of a deductive
approach to the relationship between theory and data).
The sequence can be depicted as one in which the steps
outlined in Figure 1.1 take place.
The last step involves a movement that is in the
opposite direction from deduction-it involves induction,
as the researcher infers the implications of his or her
findings for the theory that prompted the whole exercise.
The findings are fed back into the stock of theory and the
research findings associated with a certain domain of
enquiry. This can be seen in the case of the final reflec-
tions of Butler and Robson's (2001) study of gentrifica-
tion in three areas of London when they write:




Each of the three groups has played on its strengths,
where it has them. Gentrification, given this, cannot in
any sense be considered to be a unitary phenomenon,
but needs to be examined in each case according to
its own logic and outcomes. The concept of social
capital, when used as an integrated part of an
extended conceptual framework for the apprehension
of all forms of middle-class capital relations, can thus
play an important part in discriminating between
differing types of social phenomena. (Butler and
Robson 2001: 2160)
In these final reflections they show how their findings and
the interpretations of those findings can be fed back into
both the stock of knowledge concerning gentrification in
cities and, in the third of the three sentences, the concept
of social capital and its uses.
However, while this element of induetiveness un-
doubtedly exists in the approach outlined, it is typically
deemed to be predominantly deductive in orientation.
Moreover, it is important to bear in mind that, when this
deductive approach, which is usually associated with quan-
titative research, is put into operation, it often does not fol-
low the sequence outlined in its pure form. As previously
noted, 'theory' may be little more than the literature on a
certain topic in the form of the accumulated knowledge
gleaned from books and articles. Also, even when theory
or theories can be discerned, explicit hypotheses are not
always deduced from them in the way that Kelley and
De Graaf (1997) did in Research in focus 1.4. A further
pointto bear in mind is thatthe deductive process appears
very linear-one step follows the other in a clear, logical
sequence. However, there are many instances where this
is not the case: a researcher's view of the theory or litera-
The process of deduction
ture may have changed as a result of the analysis of the
collected data; new theoretical ideas or findings may be
published by others before the researcher has generatec
his or her findings; or, as in the case of Layder et al
(1991), the relevance of a set of data for a theory rna:
become apparent after the data have been collected.
Research in focus 1.4
A deductive study
Kelley and De Graaf(1997) showthat a number of studies have examined the factors that have an impact
upon individuals' religiousbeliefs. such as parents, schools, and friends, but they also argue that there are
good grounds for thinkingthat the nation into which one is born will be an important cross-cultural factor.
These reflectionsconstitute what they refer to as the 'theory' that guided their research and fromwhich the
following hypothesis was derived: 'People born into religiousnations will, in proportion to the orthodoxy of



their fellow-citizens, acquiremoreorthodox beliefs than otherwise similar peoplebornintosecularnations'
(Kelley and DeGraaf 1997: 641). Thereare twocentralconceptsinthis hypothesis that would needto be
I
measured: national religiosity (whether it isreligious or secular)and individual religious orthodoxy. The authors
hypothesized further that the religious orientation ofthe individual's family (whetherdevout or secular)
would affect the natureofthe relationship between national religiosity and religious orthodoxy.
Totest the hypotheses, a secondary analysis ofdata deriving from survey research basedon large samples
from fifteen nations wasconducted. UK readerswill be interestedto know that the British and Northern Irish
(and Irish Republic) data were derived from the British Social Attitudes survey for1991 (Jowell et01.1992).
Religious orthodoxy wasmeasured byfoursurvey questionsconcernedwith religious belief. Thequestions
askedabout (1)whetherthe person believed inGod, (2) hisor her past beliefs about God, (3) howclosethe
individual feltto God, and (4) whether he or she feltthat Godcaresabout everyone. To measure national
religiosity, the fifteen nations wereclassified intoone offive categories ascending from secular to religious.
The classification wasundertaken according to 'an unweighted averageof parental church attendance ... and
religious belief inthe nation as a whole' (Kelley and DeGraaf1997: 647). Family religious orientation was
measured on a scale offive levels of parentalchurchattendance. The hypotheses were broadly confirmed and
the authorsconclude that the 'religious environment ofa nation has a major impact on the beliefs of itscitizens'
(Kelley and De Graaf 1997: 654). Someofthe implications ofthe findings fortheoriesabout international
differences in religiosity are then outlined.
Thisstudydemonstrates the process wherebyhypotheses are deduced from existing theoryand these then
guidethe process ofdata collection sothat they can be tested.
This may all seem rather surprising and confusing.
There is a certain logic to the idea of developing theories
andthen testing them. Ineverydaycontexts, wecommonly
think of theories as things that are quite illuminating but Deductive and inductive approaches to the
that need to be tested before they can be considered valid relationship between theory and research
or useful. In point of fact, however, while the process of
Deductive approach
deduction outlined in Figure 1.1 does undoubtedly occur,
it is better considered as a general orientation to the link
between theory and research. As a general orientation, its
broad contours may frequently be discernible in social
research, but it is also the case that we often find depar-
tures from it. However, in some research no attempt is
made to followthe sequence outlined in Figure 1.1. Some
researchers prefer an approach to the relationship
Inductive approach
between theory and research that is primarily inductive.
With an inductive stance, theory is the outcome of re-
search. In other words, the process of induction involves
drawing generalizable inferences out of observations.
Figure 1.2 attempts to capture the essence of the differ-
ence between inductivism and deductivism.
However,just as deduction entails an element of induc-
tion, the inductive process is likelyto entail a modicum of

deduction. Once the phase of theoretical reflection on a

set of data has been carried out, the researcher may want
L



,.
J
to collect further data in order to establish the conditions equally itis necessary to be aware that very often what
I
in which a theory will and will not hold. Such a general one ends up with can often be little more than empirical
strategy is often called iterative: it involves a weaving' generalizations of the kind Merton (1967) wrote about.
I
back and forth between data and theory. It is particularly Research in focus 1.5 is an example of research that can
evident in grounded theory, which will be examined in be classified as inductive in the sense that itdevelops a
Chapter 22, but in the meantime the basic point is to note theory out of interview data deriving from men suffering
that induction represents an alternative strategy for link- from chronic illness concerning what determines success-
ing theory and research, although it contains a deductive ful coping mechanisms for males afflicted with such a
element too. condition. In fact, the analytic strategy adopted by the
However, as with 'theory' in connection with the author (Channaz 1997) was grounded theory, and it is
deductive approach to the relationship between theory certainly the case that many of the most prominent ex-
and research, we have to be cautious about the use of amples of inductive research derive from this tradition
the term in the context of the inductive strategy too. (see the other chapters in Strauss and Corbin 1997b, from
While some researchers undoubtedly develop theories, which Channaz's example was taken).
Research in focus 1.5
An inductivestudy
Charmaz (1991, 1997) has been concerned to examine a number of aspects of the experiences of people
with chronic illness. One phase of her research has entailed the examination specifically of men with such a
condition. Inone of her reports (Charrnaz 1997), she discusses the results of her research into twenty men
sufferingfrom chronic illness. The bulk of her data derives from semi-structured interviews. In order to bring
out the distinctivenessof men's responses, she compared the findingsrelatingto men with a parallel study
of women with chronic illness. She argues that a keycomponent of men's responses isthat of a strategy of
preserving self. Althoughthe experience of chronic illnessinvariably necessitates a change of lifestyle that itself
occasions a change in personal identity, the men sought to preserve their sense of self bydrawing on 'essential
qualities, attributes, and identities of [the] past self (Charrnaz 1997: 49). Bycontrast, women were less reliant
in their strategies of preserving self on the recapturing of past identities. She relates her theoretical reflections
of her data to her male respondents' notions of masculine identity. Her emphasis on the idea of preservingself
allows her to assess the factors that lie behind whether a man with chronic illness will 'reconstruct a positive
identity or sink into depression' (Charrnaz 1997: 57). Ifthey were unable to have access to actions that would
allowtheir sense of past self to be extended into the future (for example, through work),the probabilityof their
sinkinginto depression was enhanced.
Inthis study, the inductive nature of the relationship between theory and research can be seen in the waythat
Charmaz's theoretical ideas (such as the notion of 'preserving self) derive fromher data rather than being
formed before she had collected her data.
Charmaz's (1997) research is an interesting illustration as especially strong in terms of generating theories out of
of an inductive approach. Two points are particularly data. This contrasts with the nature of many supposedly
worth noting about it. First, as previously noted, it uses a inductive studies, which generate interesting and illumin-
grounded theory approach to the analysis of data and to ating findings but whose theoretical significance is not
the generation of theory. This approach, which was first entirely clear. They provide insightful empirical general-
outlined by Glaser and Strauss (1967), is often regarded izations, but little theory. Secondly, in much the same



hat
cal
ut.
'an
; a
ng
ss-
a
Ie
is
{-
n
n
way that the deductive strategy is associated with a not generate theory, but also theory is often used at the
quantitative research approach, an inductive strategy of very least as a background to qualitative investigations.
linking data and theory is typically associated with a It is useful to think of the relationship between theory
qualitative research approach. It is not a coincidence that and research in terms of deductive and inductive strat-
Charmaz's (1997) research referred to in Research in egies. However, as the previous discussion has implied,
focUS 1.5 isbased on in-depth, semi-structured interviews the issues are not as clear-cut as they are sometimes
that produced qualitative data in the form of respondents' presented. To a large extent, deductive and inductive
detailed answers to her questions. However, as will be strategies are possibly better thought of as tendencies
I
shown below, this characterization of the inductive strat- rather than as a hard-and-fast distinction. But these are
egy as associated with qualitative research is not entirely not the only issues that impinge on the conduct of social
straightforward: not only does much qualitative research research.

;

Epistemologicalconsiderations
Anepistemological issue concerns the question of what is ethos as the natural sciences. The position that affirms the
(or should be) regarded as acceptable knowledge in a dis- importance of imitating the natural sciences is invariably
cipline. A particularly central issue in this context is the associated with an epistemological position known as
question of whether the social world can and should be positivism (see Keyconcept 1.2).
studied according to the same principles, procedures, and
Keyconcept1.2
Whatispositivism?
Positivism isan epistemological positionthat advocates the application of the methods of the natural sciences
to the study of social realityand beyond. But the term stretches beyond this principle,though the constituent
elements vary between authors. However, positivism isalso taken to entail the following principles:
1. Only phenomena and hence knowledge confirmed by the senses can genuinely be warranted as knowledge
(the principleof phenomenalism).
2. The purpose of theory isto generate hypotheses that can be tested and that will thereby allowexplanations
of lawsto be assessed (the principleof deductivism).
3. Knowledgeisarrived at through the gathering of facts that provide the basisfor laws(the principleof
inductivism).
4. Science must (and presumably can) be conducted in a waythat isvalue free (that is,objective).
s. There isa clear distinction between scientificstatements and normative statements and a beliefthat the
former are the true domain of the scientist. This last principleis implied bythe first because the truth or
otherwise of normative statements cannot be confirmed bythe senses.



A natural science epistemology:
positivism
The doctrine of positivism is extremely difficult to pin
down and therefore to outline in a precise manner, be-
cause it is used in a number of different ways by authors.
For some writers, it is a descriptive category-one that
describes a philosophical position that can be discerned
in research-though there are still disagreements about
what it comprises; for others, it is a pejorative term used
to describe crude and often superficial data collection.
It is possible to see in the five principles in Key con-
cept 1.2 a link with some of the points that have already
been raised about the relationship between theory and
research. For example, positivism entails elements of
both a deductive approach (principle 2) and an inductive
strategy (principle 3). Also, a fairly sharp distinction is
drawn between theory and research. The role of research
is to test theories and to provide material for the develop-
ment of laws. But either of these connections between
theory and research carries with it the implication that it
is 'l0ssible to collect observations in a manner that is not
influenced by pre-existing theories. Moreover, theoret-
ical terms that are not directly amenable to observation
are not considered genuinely scientific; they must be sus-
ceptible to the rigours of observation. All this carries with
it the implication of greater epistemological status being
given to observation than to theory.
It should be noted that it is a mistake to treat positivism
as synonymous with science and the scientific. In fact,
philosophers of science and of the social sciences differ
quite sharply over how best to characterize scientific
practice, and since the early 1960s there has been a drift
away from viewing it in positivist terms. Thus, when writ-
ers complain about the limitations of positivism, it is not
entirely clear whether they mean the philosophical term
or a scientific approach more generally. Realism (in par-
ticular, critical realism), for example, is another philo-
sophical position that purports to provide an account of
the nature of scientific practice (see Keyconcept 1.3),
Key concept 1.3
"".
V(d/ What is realism?
Realismshares two features with positivism: a belief that the natural and the social sciences can and should
apply the same kinds of approach to the collection of data and to explanation, and a commitment to the view
that there isan external realityto which scientists direct their attention (in other words, there isa realitythat is
separate from our descriptions of it). There are two major forms of realism:
Empiricalrealismsimplyasserts that, through the use of appropriate methods, realitycan be understood.
This version of realismissometimes referred to as naiverealismto reflect the fact that it isoften assumed by
realists that there isa perfect (or at least very close) correspondence between realityand the term used to
describe it. As such, it 'failsto recognise that there are enduring structures and generative mechanisms
underlyingand producing observable phenomena and events' and is therefore 'superficial' (Bhaskar 1989:
2). This is perhaps the most common meaning of the term. When writers employ the term 'realism' in a
general way, it is invariablythis meaning to which they are referring.
Criticalrealism isa specificformof realismwhose manifesto is to recognize the realityof the natural order
and the events and discourses of the social world and holds that 'we will only be able to understand-and
so change-the social world ifwe identifythe structures at workthat generate those events and discourses,
... These structures are not spontaneously apparent in the observable pattern of events; they can only be
identifiedthrough the practical and theoretical work of the social sciences', (Bhaskar 1989: 2)
Critical realismimplies two things. First, it implies that, whereas positiviststake the viewthat the scientist's
conceptualization of reality actually directly reflects that reality, realists argue that the scientist's
conceptualization issimplya way of knowingthat reality. As Bhaskar (1975: 250) has put it: 'Science, then, is
the systematic attempt to express in thought the structures and ways of acting of things that exist and act



m that it
.at is not
theoret-
ervation
t be sus-
ieswith
IS being
sitivism
In fact,
s differ
ientific
l a drift
mwrit-
t is not
ilterm
in par-
philo-
unt of
3).
j
ew
lat is
J.
d by
to

d
.es.
e
independently ofthought: Critical realistsacknowledgeand accept that the categoriesthey employto
understand realityare likely to be provisional. Thus, unlikenaive realists, critical realists recognizethat there is
a distinction between the objects that are the focus of their enquiries and the terms they use to describe,
account for, and understand them. Secondly, byimplication, critical realists unlikepositivists are perfectly
content to admit intotheir explanationstheoretical terms that are not directlyamenable to observation.
Asa result, hypothetical entitiesthat account for regularitiesinthe natural or social orders (the 'generative
mechanisms' to whichBhaskarrefers)are perfectlyadmissiblefor realists, but not for positivists. Forcritical
realists, it isacceptable that generative mechanismsare not directlyobservable, sincethey are admissableon
the groundsthat their effectsare observable. What makes critical realism critical isthat the identification of
generative mechanismsoffersthe prospect of introducingchanges that can transformthe status quo. Research
infocus 24.1 providesan example of research usinga critical realist approach. Thisexample can be read
profitably at this stage even though it is ina much later chapter.
The crux of the epistemological considerations that
form the central thrust of this section is the rejection by
some writers and traditions of the application of the
canons of the natural sciences to the study of social real-
ity. A difficulty here is that it is not easy to disentangle the
natural science model from positivism as the butt of their
criticisms. In other words, it is not always clear whether
they are inveighing against the application of a general
natural scientific approach or of positivism in particular.
There is a long-standing debate about the appropriate-
ness of the natural science model for the study of society,
but, since the account that is offered of that model tends
to have largely positivist overtones, it would seem that
it is positivism that is the focus of attention rather
than other accounts of scientific practice (such as critical
realism-see Keyconcept 1.3).
Interpretivism
Interpretivism is a term given to a contrasting epistemo-
logy to positivism (see Key concept 1.4). The term sub-
sumes the views of writers who have been critical of the
application of the scientific model to the study of the
social world and who have been influenced by different
intellectual traditions, which are outlined below. They
share a view that the subject matter of the social sciences
-people and their institutions-is fundamentally differ-
ent from that of the natural sciences. The study of the
social world therefore requires a different logic of
research procedure, one that reflects the distinctiveness
of humans as against the natural order. Von Wright
(1971) has depicted the epistemological clash as being
between positivism and henneneutics (a term that is
drawn from theology and that, when imported into the
social sciences, is concerned with the theory and method
of the interpretation of human action). This clash reflects
a division between an emphasis on the explanation of
human behaviour that is the chief ingredient of the posi-
tivist approach to the social sciences and the understand-
ing of human behaviour. The latter is concerned with the
empathic understanding of human action rather than
with the forces that are deemed to act on it. This contrast
reflects long-standing debates that precede the emer-
gence of the modem social sciences but find their expres-
sion in such notions as the advocacy by Max Weber
(1864-1920) of an approach referred to in his native
German as Verstehen (which means understanding).
Weber (1947: 88) described sociology as a 'science which
attempts the interpretive understanding of social action
in order to arrive at a causal explanation of its course and
effects'. Weber's definition seems to embrace both expla-
nation and understanding here, but the crucial point is
that the task of 'causal explanation' is undertaken with
reference to the 'interpretive understanding of social
action' rather than to external forces that have no mean-
ing for those involved in that social action.
One of the main intellectual traditions that has been
responsible for the anti-positivist position has been
phenomenology, a philosophy that is concerned with the
question of how individuals make sense of the world
around them and how in particular the philosopher
should bracket out preconceptions in his or her grasp of
that world. The initial application of phenomenological
ideas to the social sciences is attributed to the work of
Alfred Schutz (1899-1959), whose work did not come to
the notice of most English-speaking social scientists until
the translation from German of his major writings in the
1960s, some twenty or more years after they had been



Keyconcept1.4
Ie
d
a
Whatisinterpretivism?
p
tc
Interpretivism isa term that usuallydenotes an alternative to the positivistorthodoxy that has held swayfor
a
decades. It is predicated upon the viewthat a strategy is required that respects the differences between
n
people and the objects of the natural sciences and therefore requires the social scientist to grasp the
n
subjective meaning of social action. Its intellectual heritage includes: Weber's notion of Verstehen; the
tt
hermeneutic-phenomenological tradition; and symbolic interactionism.
o
written. His work was profoundly influenced by Weber's
concept of Verstehen, as well as by phenomenological
philosophers, like Husserl. Schutz's position is well cap-
tured in the following passage, which has been quoted on
numerous occasions:
The world of nature as explored by the natural
scientist does not 'mean' anything to molecules,
atoms and electrons. But the observational field of
the social scientist-social reality-has a specific
meaning and relevance structure for the beings
living, acting, and thinking within it. Bya series of
common-sense constructs they have pre-selected
and pre-interpreted this world which they experience
as the reality of their daily lives. It is these thought
objects of theirs which determine their behaviour
by motivating it. The thought objects constructed
by the social scientist, in order to grasp this social
reality, have to be founded upon the thought objects
constructed by the common-sense thinking of men
[and women!], livingtheir daily lifewithin the social
world. (Schutz 1962: 59)
Two points are particularly noteworthy in this quotation.
First, it asserts that there is a fundamental difference
between the subject matter of the natural sciences and
the social sciences and that an epistemology is required
that will reflect and capitalize upon that difference. The
fundamental difference resides in the fact that social real-
ity has a meaning for human beings and therefore human
action is meaningful-that is, it has a meaning for them
and they act on the basis of the meanings that they
attribute to their acts and to the acts of others. This leads
to the second point-namely, that it is thejob of the social
scientist to gain access to people's 'common-sense think-
ing' and hence to interpret their actions and their social
t1
world from their point of view. It is this particular feature a
that social scientists claiming allegiance to phenomeno- f
logy have typically emphasized. In the words of the d
authors of a research methods text whose approach is
P
described as phenomenological: 'The phenomenologist 11
views human behavior ... as a product of how people
tl
interpret the world.... In orderto grasp the meanings of v
a person's behavior, the phenomenologist attempts to see o
thingsfrom thatperson's point of view' (Bogdan and Taylor
o
1975: 13-14, emphasis in original).
In this exposition of Verstehen and phenomenology,
it has been necessary to skate over some complex issues.
c
In particular, Weber's examination of Verstehen is far
c
more complex than the above commentary suggests,
c
because the empathetic understanding that seems to be
i
implied above was not the way in which he applied it
(Bauman 1978), while the question of what is and is not
s
a genuinely phenomenological approach to the social
e
sciences is a matter of some dispute (Heap and Roth
a
1973). However, the similarity in the writings of the
c
hermeneutic-phenomenological tradition and of the
Verstehen approach, with their emphasis upon social
t
action as being meaningful to actors and therefore need-
ing to be interpreted from their point of view, coupled
E
with the rejection of positivism, contributed to a stream
E
of thought often referred to as interpretivism (e.g. J. A.
Hughes 1990).
Verstehen and the hermeneutic-phenomenological tra-
dition do not exhaust the intellectual influences on inter-
pretivism. The theoretical tradition in sociologyknown as
symbolic interaetionism has also been regarded by many
writers as a further influence. Again, the case is not clear-
cut. The implications for empirical research of the ideas of
the founders of symbolic interactionism, in particular
George Herbert Mead (1863-1931), whose discussion
of the way in which our notion of self emerges through
an appreciation of how others see us, have been hotly
..
b


J
,
I
debated. Therewas a school of research, known as the leastfindings thatappearsurprisingifalargelyexternal
Iowaschool, thathas drawnheavilyon Mead'sconcepts stanceistaken-thatis,apositionfrom outsidethe par-
I
t
i
I
f
!
I
I
I
I
!
I
!
I
r
f
i
ayfor
en
rfeature
romeno-
: of the
roach is
aologist
people
lingsof
:s to see
and ideas, but has proceeded in a direction that most
people would prefer to depict as largely positivist in
tone (Meltzeret al. 1975).Moreover,some writershave
arguedthat Mead's approach is farmore consistent with a
naturalscience approach thanhas typicallybeenrecog-
nized(McPhailandRexroat1979).However,thegeneral
tendency has been to view symbolic interactionism as
occupying similarintellectualspaceto the hermeneutic-
phenomenologicaltradition and so broadlyinterpretat-
ivein approach. This tendencyis largelytheproductof
thewritingsofHerbertBlumer,astudentofMead'swho
actedashismentor'sspokesmanandinterpreter,andhis
followers (Hammersley1989;R.Collins1994).Notonly
did Blumercoin the term symbolic interaction; he also
provided a gloss on Mead's writings thathas decidedly
interpretative overtones. Symbolic interactionistsargue
thatinteractiontakes place in such awaythatthe indi-
vidual is continuallyinterpretingthe symbolic meaning
of his or her environment (which includes the actions
ticularsocialcontextbeingstudied.Researchinfocus1.6
providesan interestingexampleofthis possibility.
Ofcourse,asthe exampleinResearchinfocus 1.6sug-
gests, when the socialscientistadoptsan interpretative
stance,he or she isnotsimplylayingbarehowmembers
of a socialgroupinterpretthe world aroundthem. The
socialscientistwillalmostcertainlybeaimingtoplacethe
f
I
!
I
I
I
interpretations thathave been elicited into a socialsci-

entificframe. Thereisadoubleinterpretationgoingon:
the researcher is providing an interpretation of others'
interpretations.Indeed,thereisathirdlevel ofinterpre-
tationgoingon,becausethe researcher'sinterpretations
haveto be furtherinterpretedin termsof theconcepts,
theories, andliterature of a discipline. Thus, takingthe
exampleinResearchinfocus1.6, Foster's(1995)sugges- f
tionthatRiversideis notperceivedasahighcrimeareaby
residentsisherinterpretationofhersubjects'interpreta-
tions.Shethenhad theadditional jobofplacingherinter-
esting findings intoa social scientific frame, which she
Taylor
ofothers)and acts on the basisofthis imputedmeaning.
Inresearchterms,accordingtoBlumer(1962: 188),'the
iology,
position of symbolic interaction requires the studentto
issues.
catchthe processofinterpretationthroughwhich[actors]
is far
construct their actions', a statement that brings out
ggests,
clearlyhisviews oftheresearchimplicationsofsymbolic
s to be
interaetionismandofMead'sthought.
lied it
Itshouldbe appreciatedthattheparallelismbetween
isnot
symbolic interactionism and the hermeneutic-phenom-
social
enologicaltraditionshouldnotbe exaggerated.The two
accomplishedby relatingthemto existingconceptsand
discussions in criminology of such things as informal
social control, neighbourhood watch schemes, and the
role ofhousingasapossiblecauseofcriminalactivity.
The aim of this section has been to outline how epi-
stemologicalconsiderations-especiallythoserelatingto
thequestionofwhetheranaturalscienceapproach,and
inparticularapositivistone, cansupplylegitimateknow-
ledgeof thesocialworld-arerelated to research prac-
tice. There is a link with the earlier section in that a

Roth
are unitedin theirantipathyfor positivism andhave in deductive approach to the relationship between theory
)f the
common an interpretative stance. However, symbolic andresearchistypicallyassociatedwithapositivistposi-
f the
interactionism is, at least in the hands of Blumer and tion. Keyconcept 1.2 doestry to suggestthat inductivism
social
the many writers and researchers who have followed isalso afeatureofpositivism(thirdprinciple),but,inthe
need-
in his wake, a type of social theorythathas distinctive working-throughofitsimplementationinthepracticeof
rpled
epistemologicalimplications;the hermeneutic-phenom- socialresearch, itisthe deductiveelement(secondprin-
ream
enological tradition, by contrast, is best thought of as
J.A.
a general epistemological approach in its own right.
Blumermay have beeninfluenced by the hermeneutic-
I tra-
phenomenological tradition, but there is no concrete
nter-
evidence of this. There are other intellectual currents
rnas
thathaveaffinitieswiththe interpretativestance,such as
iany
the working-through of the ramifications of the works
.ear-
ofthe philosopherLudwig Wittgenstein (Winch 1958),
is of
but the hermeneutic-phenomenological,Verstehen, and
ular
symbolic interactionist traditions can be considered
sion major influences.
ugh
Taking an interpretative stance can mean that the
)tly researchermaycome up with surprisingfindings, or at
ciple) that tends to be emphasized. Similarly, the third
levelofinterpretationthataresearcherengagedininter-
pretativeresearchmustbringinto operationisverymuch
part of the kind of inductive strategy described in the
previoussection.However, whilesuchinterconnections
between epistemological issues and research practice
exist, it is important not to overstate them, since they
representtendenciesratherthandefinitivepointsofcor-
respondence.Thus, particularepistemologicalprinciples
andresearchpracticesdonotnecessarilygohandinhand
in aneatunambiguousmanner.This pointwill be made
againon severaloccasionsandwill be a specialfocus of
Chapter24.
t
[
!
t
!
!
I.
1
t
,
i



Research in focus 1.6
(
Interpretivism in practice
Foster(1995)conducted ethnographic research using participant observation and semi-structured interviews
ina housingestate in East London, referred to as Riverside. The estate had a high level of crime,as indicated by
official statistics on crime. However, she found that residents did not perceive the estate to be a highcrime
area. This perception could be attributed to a number of factors, but a particularly important reason was the
existence of 'informalsocialcontrol'. People expected a certain level of crime, but felt fairly secure because
informal social control allowed levelsof crime to be contained. Informal social control compriseda number
of different aspects. One aspect wasthat neighbours often looked out for each other. Inthe words of one of
Foster's interviewees: 'If' hear a bang or shouting Igo out. Ifthere's aggravation Icome inand ringthe police.
(
Idon't stand for it.' Another aspect of informal social control was that people oftenfelt secure because they
kneweach other. Another respondent said: 'I don't feel nervous ... because people do generallyknoweach
J--
other. We keep an eye on each others properties ... Ifeel quite safe because you knowyour neighbours and
P
you knowthey're there ... they lookoutfor you'. (Foster 1995:575)
t
a
Ontological considerations

- ~
,. lj
'-
Questions of social ontology are concerned with the
nature of social entities. The central point of orientation
here is the question of whether social entities can and
should be considered objective entities that have a reality
external to social actors, or whether they can and should
be considered social constructions built up from the per-
ceptions and actions of social actors. These positions are
frequently referred to respectively as objectivism and
constructionism. Their differences can be illustrated by
reference to two of the most common and central terms in
social science----organization and culture.
Objectivism
Objectivism is an ontological position that implies that
social phenomena confront us as external facts that are
beyond our reach or influence (see Keyconcept 1.5).
We can discuss organization or an organization as a
tangible object. It has rules and regulations. It adopts
standardized procedures for getting things done. People
are appointed to different jobs within a division oflabour.
There is a hierarchy. It has a mission statement. And so
on. The degree to which these features exist from organ-
ization to organization is variable, but in thinking in these
terms we are tending to the view that an organization has
a reality that is external to the individuals who inhabit it.
Moreover, the organization represents a social order in
that it exerts pressure on individuals to conform to the
requirements of the organization. People learn and apply
the rules and regulations. They follow the standardized
procedures. They do the jobs to which they are appointed.
People tell them what to do and they tell others what to
do. They learn and apply the values in the mission state-
ment. If they do not do these things, they may be repri-
manded or even fired. The organization is therefore a
constraining force that acts on and inhibits its members.
The same can be said of culture. Cultures and subcul-
tures can be viewed as repositories of widely shared values
and customs into which people are socialized so that
they can function as good citizens or as full participants.
Cultures and subcultures constrain us because we inter-
nalize their beliefs and values. In the case of both organ-
ization and culture, the social entity in question comes
across as something external to the actor and as having an
almost tangible reality of its own. It has the characteristics
of an object and hence of having an objective reality. To a
very large extent, these are the 'classic' ways of conceptu-
alizing organization and culture.
c



interviews
licatedby
1crime
'asthe
:ause
nber
neof
police,
they
each
'sand
onhas
abitit.
derin
tothe
apply
xlized
inted.
hatto
state-
repri-
ore a
lers.
bcul-
slues
that
ants,
ater-
gan-
mes
gan
sties
Foa
xu-
.. "'t, I
t in "
Socialresearchstrategies 19
1I;l: ,,1 ' '. "
,
I
,
Keyconcept1.5
!
!
A'.
W
What is objectivism?
f
f
Objectivism isanontological position that assertsthatsocial phenomenaandtheirmeanings haveanexistence
that isindependentofsocialactors.It implies that socialphenomenaand thecategories that we use in
everyday discoursehavean existence that isindependent or separatefrom actors.
Constructionism
However, we can consider an alternative ontological
position---<onstructionism (Keyconcept 1.6). Thisposi-
tionchallenges the suggestion that categories such as
organizationandculturearepre-givenandthereforecon-
frontsocialactorsasexternal realitiesthat theyhaveno
roleinfashioning.
Letus take organization first. Strauss et al. (1973),
drawingon insights from symbolicinteractionism, car-
riedout researchinapsychiatrichospitaland proposed
that itwas best conceptualized as a 'negotiated order'.
Insteadoftakingtheviewthatorderinorganizationsisa
6' :. ...
Vt::!I
Keyconcept1.6
i
pre-existing characteristic,they argue thatit is worked at.
I
Ruleswerefarless extensive andlessrigorously imposed
thanmightbesupposedfromtheclassicaccountoforgan-
ization. Indeed, Strauss et al. preferto refer tothem as
'much lesslikecommands, and muchmorelikegeneral
I
understandings'(1973: 308).Preciselybecauserelatively
I
littleofthespheresofactionofdoctors,nurses,andother
personnelwasprescribed,thesocialorderofthehospital
wasan outcome ofagreed-uponpatterns ofactionthat
were themselves the products of negotiations between
the different parties involved. The social order is in I
r
,:
a constant state of change because the hospital is 'a
placewherenumerous agreementsarecontinuallybeing
I
i
t
!
What is constructionism? !
I
Constructionism isanontological position(oftenalsoreferredto as constructivism) that assertsthatsocial
phenomena andtheirmeaningsare continuallybeingaccomplishedbysocialactors. Itimplies that social
phenomenaandcategories are not onlyproducedthroughsocial interactionbut that theyare ina constant
state of revision. In recent years, the termhasalsocometoinclude the notion that researchers' ownaccounts
ofthesocial world are constructions. In other words, the researcheralways presentsaspecific version ofsocial
reality, ratherthan onethat can be regarded asdefinitive. Knowledge isviewed as indeterminate. The
discussion of postmodernisminChapter27furtherexaminesthisviewpoint.Thissenseofconstructionismis
usually allied to theontological version ofthe term.In other words. these arelinked meanings. Both meanings
areantitheticaltoobjectivism (see Key concept 1.5),but the secondmeaning isalsoantitheticaltorealism
(see Key concept 1.3). Thefirst meaningmightbe thought of usefully as constructionisminrelation to the
socialworld;the secondasconstructionisminrelationto the natureofknowledgeofthesocialworld(and
indeedthenaturalworld).
Increasingly,the notionofconstructionisminrelationto the natureofknowledgeof thesocialworldisbeing
incorporated intonotions ofconstructionism. but inthisbookIwill be using the terminrelation to thefirst
meaning,wherebyconstructionismis presentedas anontological position inrelating to socialobjectsand
categories-thatis,onethatviewsthemassociallyconstructed.
..-- -_._._--- ---'---



r
f
terminated or forgotten, but alsoas continually being estab-
lished, renewed, reviewed, revoked, revised.... In any
pragmatic sense, this is the hospital at the moment: this is
its social order' (Straussetal. 1973: 316-17). The authors
argue that a preoccupation with the formal properties of
organizations (rules, organizational charts, regulations,
roles) tends to neglect the degree to which order in organ-
izations has to be accomplished in everyday interaction,
though this isnot to say that the formal properties have no
element of constraint on individual action.
Much the same kind of point can be made about the
idea of culture. Instead of seeing culture as an exter-
nal reality that acts on and constrains people, it can be
taken to be an emergent reality in a continuous state of
construction and reconstruction. Becker (1982: 521),
for example, has suggested that 'people create culture
continuously.... No set of cultural understandings ...
provides a perfectly applicable solution to any problem
people have to solve in the course of their day, and they
therefore must remake those solutions, adapt their
understandings to the new situation in the light of what is
different about it.' Like Strauss et al., Becker recognizes
that the constructionist position cannot be pushed to the
extreme: it is necessary to appreciate that culture has a
reality that 'persists and antedates the participation of
particular people' and shapes their perspectives, but it
is not an inert objective reality that possesses only a sense
of constraint: it acts as a point of reference but is always in
the process of being formed.
Neither the work of Strauss et al. nor that of Becker
pushes the constructionist argument to the extreme. Each
admits to the pre-existence of their objects of interest
(organization and culture respectively). However, in
each case we see an intellectual predilection for stressing
the active role of individuals in the social construction of
social reality. Not all writers adopting a constructionist
Research infocus 1.7
position are similarly prepared to acknowledge the exis-
tence or at least importance of an objective reality. Walsh
(197?: 19), for example, has written that 'we cannot take
for granted, as the natural scientist does, the availability
of a preconstituted world of phenomena for investigation'
and must instead 'examine the processes by which the
social world is constructed.' Constructionism essentially
invites the researcher to consider the ways in which social
reality is an ongoing accomplishment of social actors
rather than something external to them and that totally
constrains them.
Constructionism also suggests that the categories that
people employ in helping them to understand the natural
and social world are in fact social products. The cat-
egories do not have built-in essences; instead, their mean-
ing is constructed in and through interaction. Thus, a
category like 'masculinity' might be treated as a social
construction. This notion implies that, rather than being
treated as a distinct inert entity, masculinity is construed
C
as something whose meaning is built up during interac-
refl-
tion. That meaning is likely to be a highly ephemeral one,
WOl
in that it will vary by both time and place. This kind of
ism
stance frequently displays a concern with the language
WOJ
that is employed to present categories in particular ways.
con
It suggests that the social world and its categories are not
cat'
external to us, but are built up and constituted in and
through interaction. This tendency can be seen particu-
larly in discourse analysis, which is examined in Chapter
Re
20. N;Potter (1996: 98) observes: 'The world ... is con-
Qu
stituted in one way or another as people talk it, write
issi
it and argue it.' This sense of constructionism is highly
log
antithetical to realism (see Key concept 1.3). Construc-
wa
tionism frequently results in an interest in the representa-
tion of social phenomena. Research in focus 1.7 provides
an illustration of this idea in relation to the representation
of the breast cancer epidemic in the USA.
M;
di:
Constructionisminaction
n
Lantzand Booth(1998) haveshown that breastcancercanbetreatedas a social construction. Theynotethat fu
US datashowa risein the incidenceof thediseasesincethe early1980s.which hasledto the depictionof the
e"
trend asanepidemic.Theauthorsexamineda varietyof popularmagazinesusingqualitativecontent
is
analysis(see Keyconcept 12.1 for a brief descriptionof this method). They notethat manyof the articlesdraw
is
attention to thelifestyles of modernwomen. suchasdelayingfirst births. diet andalcohol consumption. and
til
havingcareers.Theauthorsarguethatthearticles:
tl:




ge the exis.
ility, Walsh
:annottake
availabilhy
'estigation'
which the
essentially
hichsocial
.ial actors
iat totally
ories that
renatural
The cat-
eirmean,
Thus, a
a social
an being
mstrued
interac-
iral one,
kind of
inguage
ir ways.
are not
in and
iarticu-
.hapter
is con-
, write
highly
istruc-
senta-
ovides
tation
element in its social construction.
Constructionism is also frequently used as a term that
reflects the indeterminacy of our knowledge of the social
world (see Key concept 1.6 and the idea of construction-
ism in relation to the nature of knowledge of the social
world). However, in this book, I will be using the term in
connection with the notion that social phenomena and
categories are social constructions.
Relationship to social research
Questions of social ontology cannot be divorced from
issues concerning the conduct of social research. Onto-
logical assumptions and commitments will feed into the
ways in which research questions are formulated and
ascribe blame to individualbehaviors by listinga wide array of individual riskfactors (many of which are not
behaviors of 'traditional' women), and then offeringprudent prescriptions for prevention. Women are

portrayed as victimsof an insidiousdisease, but also as victimsof their own behaviors, many of which are
related to the control of their own fertility.... These articles suggest that nontraditional women experience
pathological repercussions within their bodies and, in turn, may be responsible for our current epidemic of
breast cancer. (Lantz and Booth 1998:915-16)
This article suggests that. as a social category, the breast cancer epidemic is being represented in popular
magazines in a particular way-<>ne that blames the victimsand the lifestylesof modern women in particular.
This is inspite of the fact that fewer than 20per cent of cases of breast cancer are in women under the age of
50. Lantzand Booth's study isfairly representative of a constructionist ontology insuggestingthat the epidemic
is not simplybeing construed as a social fact but is being ascribed a particular meaning (one that blames the
victimsof the disease). Inthis way, the representation of the disease in popular magazines forms an important
research is carried out. If a research question is formu-
lated in such a way as to suggest that organizations and
cultures are objective social entities that act on indi-
viduals, the researcher is likely to emphasize the formal
properties of organizations or the beliefs and values of
members of the culture. Alternatively, if the researcher
formulates a research problem so that the tenuousness
of organization and culture as objective categories is
stressed, it is likely that an emphasis will be placed on
the active involvement of people in reality construction.
In either case, it might be supposed that different
approaches to the design of research and the collection of
data will be required.
.'

Research strategy: quantitative and
, .-.
.
qualitative research
Many writers on methodological issues find it helpful to
distinguish between quantitative and qualitative research.
The status of the distinction is ambiguous, because it is
almost simultaneously regarded by some writers as a
fundamental contrast and by others as no longer useful or
even simplyas 'false' (Layder 1993: 110). However, there
is little evidence to suggest that the use of the distinction
is abating and even considerable evidence of its con-
tinued, even growing, currency. The quantitative/qualita-
tive distinction will be employed a great deal in this book,
because it represents a useful means of classifying differ-
ent methods of social research and because it is a helpful
umbrella for a range of issues concerned with the practice
of social research.
On the face of it, there would seem to be little to the
quantitative/qualitative distinction other than the fact
that quantitative researchers employ measurement and
qualitative researchers do not. It is certainly the case
that there is a predisposition among researchers along
these lines, but many writers have suggested that the





r
liter
use
Fundamentaldifferencesbetweenquantitativeand qualitativeresearchstrategies
aki
Quantitative Qualitative
ant
Principalorientationtotheroleof
theoryinrelationtoresearch
Epistemologicalorientation
Ontologicalorientation
Deductive;testingof theory
Naturalsciencemodel,inparticularpositivism
Objectivism
Inductive;generationoftheory
Interpretivism
Constructionism
a u
tap'
leal
to I
bee
differences are deeper than the superficial issue of the
presenceor absence ofquantification.Formany writers,
quantitativeand qualitative researchdifferwith respect
totheirepistemologicalfoundationsandinotherrespects
too.Indeed,ifwetaketheareas thathavebeenthefocus
ofthelastthreesections-theconnectionbetweentheory
andresearch,epistemologicalconsiderations,andontolo-
gical considerations-quantitativeandqualitativeresearch
canbe takento form two distinctive clusters ofresearch
strategy. Byaresearchstrategy,Isimplymeanageneral
orientation to the conduct ofsocial research. Table 1.1
outlinesthe differences betweenquantitativeand quali-
tativeresearchintermsofthethree areas.
Thus, quantitative research can be construed as a
research strategythatemphasizes quantification in the
collectionand analysisofdataand that:
entails a deductive approach to the relationship
betweentheoryand research, in which the accentis
placedonthetestingoftheories;
has incorporatedthe practices and norms ofthe nat-
uralscientificmodelandof positivismin particular;and
embodiesaviewofsocialrealityasanexternal,objec-
tivereality.
Bycontrast, qualitative research can be construed as a
research strategythat usually emphasizes words rather
thanquantificationinthe collection and analysisofdata
andthat:
predominantlyemphasizes an inductiveapproach to
therelationshipbetweentheoryandresearch,inwhich
theemphasisisplacedonthegenerationoftheories;
hasrejected thepracticesand normsofthenaturalsci-
entificmodel and ofpositivisminparticularinprefer-
enceforanemphasisonthewaysinwhichindividuals
interprettheirsocialworld;and
ide
ety
fro
as
embodiesaviewofsocialrealityasaconstantlyshift- cOl
ingemergentpropertyofindividuals'creation. yOl
for
Thereis,infact,considerablymoretothequantitative/
fin'
qualitative distinction than this contrast. In Chapters 6
24
and 16 the nature of quantitative and then qualitative
sec.
research respectively will be outlined in much greater
G11
detail, while in Chapters24 and 25 the contrastingfea-
tureswillbefurtherexplored.Inparticular,anumberof
Ad
distinguishingfeaturesflowfromthecommitmentofthe
on
quantitativeresearchstrategytoapositivistepistemology
vie
and from the rejection of that epistemology by practi-
rat
tioners of the qualitative research strategy. In other
wi
words, the threecontrastsinTable1.1arebasic,though
an
fundamental,ones.
an
However, the interconnections between the differ-
co
entfeaturesofquantitativeand qualitativeresearchare
ar
not as straightforward as Table 1.1 and the previous
fo
paragraph imply. While it isuseful to contrastthe two
re
research strategies, it is necessary to be careful about
Vt
hammering a wedge between them too deeply. Itmay
0\
seemperversetointroduceabasicsetofdistinctionsand
til
thensuggestthattheyareproblematic.Arecurringtheme
re
ofthisbookisthatdiscussingthenatureofsocialresearch
is just as complex as conducting research in the real
world. You may discover general tendencies, but they
are precisely that-tendencies. In reality, the picture
becomesmore complicatedthemoreyoudelve.
..
For example, it is common to describe qualitative
research as concerned with the generation ratherthan
the testingoftheories. However, thereare examples of
studiesinwhich qualitativeresearch hasbeen employed
to test rather than to generate theories. For example,
Adler and Adler (1985) were concerned to explore the
issueofwhetherparticipationinathleticsinhigheredu-
cationinthe USAisassociatedwithhigherorlowerlevels
ofacademicachievement, anissueonwhich theexisting




I
!
I
l
I
I
I
concerned to measure a wide variety of concepts, but
exhibited little evidenceof aconcernto test theories of
unemployment or of a stressful life event like redund-
ancy. Instead, itsconclusionsrevolvearoundseekingto
understandhowthosemaderedundantrespondedtothe
experience in terms of such things as their job-search
methods,theirinclinationtofindjobs,andtheirpolitical
attitudes.As such, ithas interpretivistovertonesinspite
ofbeing anexerciseinquantitativeresearch.
The point that is being made in this section is that
quantitativeandqualitativeresearchrepresentdifferent
researchstrategiesand thateach carries with it striking
differencesintermsofthe roleoftheory,epistemological
issues, andontologicalconcerns. However, the distinc-
tion is not a hard-and-fast one: studies that have the
broadcharacteristicsofoneresearchstrategymayhavea
characteristicoftheother.Notonlythis,butmanywriters
argue thatthe two can be combined within an overall
researchproject,and Chapter25 examinespreciselythis
possibility. InChapter25,Iwillexamine whatisincreas-
inglyreferredtoasmixed methods research. Thistermis
widelyused nowadaystorefertoresearchthatcombines
methods associatedwithbothquantitativeand qualita-
tiveresearch. Chapter25 willexplore the differentways
in which social researcherscombine these two research
strategies.
InResearchinfocus1.8,there isanexampleofamixed
methods study. I am presenting it here partly to pro-
vide an early insightinto the possibility ofdoing mixed
methods research,butalso to show how awedgeneed
notand should not be driven betweenquantitativeand
.qualitativeresearch. Bycontrastingthe twoapproaches,
itiseasy tosee themasincompatible.As the examplein
Researchinfocus 1.8 shows, they can befruitfully com-
binedwithinasingleproject.Thispointwillbeamplified
furtherthroughoutChapter25.
noftheory
ulyshift-
n.
ttitative/
iapters 6
ialitativs
1 greater
ring fea-
rmberof
ntofthe
emology
ypracti-
ln other
,though
~ differ-
arch are
orevious
the two
11 about
.It may
onsand
gtheme
esearch
:he real
literature was inconsistent.This isan illustration ofthe
useof the existing literature on a topic being employed as
akindofproxyfortheory. Thefirstauthorwasaparticip-
antobserverforfour years ofabasketballprogrammein
a university, and both authors carried out 'intensive,
tapedinterviews'withplayers. The authors' findings do
leadthem toconcludethatathleticparticipationislikely
to result in lower academic achievement. This occurs
becausethe programmeparticipantsgraduallydriftfrom
idealisticgoalsabouttheiracademiccareers, and avari-
etyoffactors lead them tobecomeincreasinglydetached
fromacademic work. For example, onestudent is quoted
assaying: 'IfIwas a studentlike most otherstudents I
coulddo well, butwhen you play thecalibreof ball wedo,
youjust can'tbeanabove-averagestudent.What Istrive
for nowisjustto beanaverage student....Youjust can't
findthetimetodo all thereading'(Adler andAdler1985:
247). This study shows how, although qualitative re-
searchistypicallyassociatedwithgeneratingtheories,it
canalsobeemployedfortesting them.
Moreover, it isstriking that, although the Adler and
Adlerstudy isbroadly interpretivist in epistemological
orientation, with its emphasis on how college athletes
viewtheir socialsituation, the findings have objectivist,
rather than constructionist, overtones. For example,
whenthe authors describe thestudents' academic perform-
anceas'determinedlessbydemographiccharacteristics
andhighschoolexperiencesthanbythestructureoftheir
collegeexperiences' (Adler andAdler 1985: 249), they
arepositingasocialworldthatis'outthere'andthathasa
formal, objectivequality. It isan exampleof qualitative
research in the sense that there is no quantification or
verylittleofit,but itdoes not haveall the otherfeatures
outlined in Table 1.1. Similarly, the previously men-
tionedstudybyWestergaardetal. (1989)oftheeffectsof
redundancywasaquantitativestudyinthesenseofbeing
i
I
f
society' (Beck1992), whichhas attracted a good deal ofsociological attention. At theheightof the disease
i
f
t
i
'ut they
picture
tlitativs
er than
iples of
iployed
Research infocus 1.8
Mixedmethodsresearch-anexample
In2001 Britain was profoundly affectedbythe Foot and Mouth Disease(FMD), whichhad a bigimpacton
people'smovements. Poortingaet 01. (2004)were interested inhowfar thepublictrusted theinformationthe
:ample,
ore the
government wassupplying andhowit perceived therisksassociated withthedisease. Suchissueswere of
f
interest in part because theresearchers felt thatthewaysinwhichthepublicresponds to acrisis wasan
I
L
eredu-
rlevels
xisting
importanttopic, but also because the issuesconnect withtheinfluenceinrecent years ofthenotionofthe 'risk
during2-5 April 2001. the researchersconducted a surveybyadministering aself-completionquestionnaire
- ~ ~



(see Chapter9)tosamples intwocontrasting areas: BudeinCornwalland NorwichinNorfolk. Thesetwoareas
were chosen becausetheywere verydifferently ~ e t e d byFMD. Thequestionnairecoveredthe following
areas: levelofagreementwithstatementsaboutthe outbreakofthe disease(e.g. 'Mymainconcernsabout
FMDaretodowiththe possibleimpactsonhuman health');perceptionsofwhowastoblame;levelof
agreementwithstatementsaboutthe government'shandlingofFMD; degrees oftrust invarioussourcesof
informationaboutthe disease;and personal information,such asanyconnectionwiththe farmingortourist
industries.Inaddition,aqualitativeresearch method-focusgroups(see Chapter19)-wasemployed. InMay
andJune 2001, thesegroupswereconvenedand membersofthe groupswere askedaboutthe samekindsof
issuescovered inthe questionnaire.Focusgroup participants were chosen fromamongthose whohad
indicated intheirquestionnairerepliesthatthey were willingto beinvolvedinafocusgroupdiscussion.Three
focusgroupdiscussionstookplace.Whilethe questionnairedata were ableto demonstratethe variationin
suchthingsastrust invariousinformationsources,the focusgroups revealed'valuableadditionalinformation,
especially onthe reasons,rationalizationsand argumentsbehind people'sunderstandingof the FMDissue'
(Poortingaet 01. 2004: 86).Asaresult,the researcherswere abletoarriveatamorecomplete accountofthe
FMD crisisthan could have been obtained by either a quantitative or a qualitative research approach alone.
This and other possible advantages of mixed methods research will be explored further in Chapter 25.
Influences on the conduct of
social research
Wearebeginningtogetapicturenowthatsocialresearch
is influenced by a variety of factors. Figure 1.3 sum-
marizes the influences thathave been examined so far,
but has added two more-the impact of values and of
practical considerations.
Values
Valuesreflecteitherthe personalbeliefsorthefeelingsof
aresearcher.Onthefaceofit,wewouldexpectthatsocial
Influences on social research
Theory Practical considerations Epistemology
<,
/
/
Values Ontology
'"
scientists should be value free and objective in their
research.Afterall, onemightwanttoarguethatresearch
thatsimplyreflectedthe personalbiasesofits practition-
ers couldnotbeconsideredvalidandscientificbecauseit
wasboundup withthe subjectivitiesof its practitioners.
Such a viewisheldwithless andless frequencyamong
socialscientistsnowadays.EmileDurkheim(1858-1917)
wrotethatoneofthe corollariesofhis injunctionto treat
socialfactsasthingswasthatall'preconceptionsmustbe
eradicated'(Durkheirn1938:31).Sincevaluesareaform
ofpreconception,hisexhortationwasatleastimplicitlyto
dowithsuppressingthemwhenconductingresearch.His
positionisunlikelyto be regardedascrediblenowadays,
becausethereisagrowingrecognitionthatit isnotfeas-
ible to keep the valuesthata researcherholdstotallyin
check. These can intrude at any or all ofa number of
pointsinthe processofsocial research:
choiceofresearcharea;
formulationofresearchquestion;
choiceofmethod;
formulation ofresearch design and data collection
techniques;
implementationofdatacollection;


etwoareas
lowing
about
of
rcesof
tourist
ed.lnMay
'kindsof
Id
on. Three
tionin
umation,
issue'
tofthe
alone.
in their
research
ractition-
ecauseit
titioners.
y among
8-1917)
1totreat
;must be
reaform
ilicitlyto
arch. His
rwadays,
notfeas-
otallyin
mber of
illection
analysisofdata;
interpretationofdata;
conclusions.
There are, therefore,numerouspointsatwhichbias and
theintrusionofvaluescanoccur.Valuescan materialize
atanypointduringthecourseofresearch.Theresearcher
maydevelopanaffectionorsympathy,whichwasnotneces-
sarily present at the outset of an investigation, for the
peoplebeingstudied.Itisquitecommon,forexample,for
researchersworkingwithinaqualitativeresearchstrategy,
and inparticularwhentheyuse participantobservation
orveryintensiveinterviewing,todevelopaclose affinity
with the peoplewhomtheystudytotheextentthatthey
finditdifficultto disentangletheirstanceassocial scien-
tistsfrom theirsubjects'perspective.Thispossibilitymay
beexacerbatedbythe tendencythatBecker(1967)iden-
tifiedforsociologistsinparticulartobe very sympathetic
to underdog groups. Equally, social scientists may be
repelled by the peopletheystudy.The social anthropo-
logist Colin Turnbull (1973) reports the results of his
researchinto anAfrican tribeknownas the Ik.Turnbull
was appalledby whathe witnessed: a loveless (and for
him unlovable) tribe thatleft its young andvery oldto
die. While Turnbullwas able to pointto the conditions
that hadled to this stateofaffairs, he was very honestin
hisdisgustforwhathewitnessed,particularlyduringthe
period of his initialsojourn amongthe tribe. However,
thatverydisgustisaproductofWesternvaluesaboutthe
family,anditislikely,ashe acknowledged,thatthesewill
haveinfluencedhisperceptionofwhathewitnessed.
Student experience
Anotherpositionin relation to thewhole questionof
values and bias is to recognize and acknowledge that
I
i
researchcannotbe valuefree butto ensurethatthereis
I

no untrammelledincursionof valuesintheresearchpro-
cess and to be self-reflective and so exhibit reflexivity
aboutthepartplayedbysuchfactors.As Turnbull(1973:
f
13) put it at the beginning of his bookon the Ik: 'the
reader is entitled toknowsomethingofthe aims, expecta-
tions, hopesandattitudesthatthewriterbroughtto the
field with him, for these will surelyinfluencenot onlyhow
he sees things buteven whathe sees.' Researchersare
I
increasinglypreparedto forewarnreadersoftheirbiases
andassumptionsandhowthesemayhave influencedthe
subsequentfindings. Therehasbeena growthsince the
I
mid-1970sofcollectionsofinsidereportsofwhatdoinga
pieceofresearchwas reallylike, asagainstthegenerali-
ties presentedinsocialresearchmethodstextbooks(like
this one!).Thesecollectionsfrequentlyfunctionas'con-
I
r
fessions', an elementof which isoften thewriter'spre-
paredness to be open about his or herpersonal biases.
Thispointwillbe takenupfurtherinChapter27.
Stillanother approach is toargue for consciouslyvalue- I
ladenresearch.This isa positiontakenbysomefeminist
l
writerswhohave arguedthat only research on women I
t
thatisintendedlyfor womenwill be consistentwith the
wider political needs ofwomen. Mies (1993: 68) has
arguedthat in feminist research the'postulate ofvalue
free research, of neutralityand indifferencetowardsthe
research objects,hasto be replacedbyconscious partial-
ity, whichisachievedthroughpartialidentificationwith
theresearchobjects'(emphasesinoriginal).
The influence of feminism on research questions
SarahHansonis veryclearaboutthe influenceoffeminismon her research andonher researchquestions
inparticular.
Myresearch project focused on the representation of women through the front covers of fivewomen's
magazines, combiningtheapplication of feminist theory with the decoding practices of content analysis.
Throughoutthe projectI wantedtounderstand the natureofwomen'smagazines, the influences theyhave
on women's sense of selfand identity and the role the magazines play. I asked:dowomen'smagazines
supportordestroywomen'sidentityanddotheyencourage self-respect orself-scrutiny?I wantedto
combine theory with fact, focusingon the meanings behind the presentation of images and text.
Similarly, for her research on NGOs(non-governmental organizations) and sex workers in Thailand, Erin
Sanderswrotethatshe'employedafeministmethodology-andassuchattemptedtoengage withmy
researchparticipants, particularlythe sex workers, as a "friend" rather thanas a "researcher" '. She also writes:



01
Ichoseto use a feminist methodology because Iwantedto eliminate the power imbalance inthe research
UI
relationship. Asthere are a number of power issues witha White', Western' womaninterviewing 'Non-
rb
White', 'Non-Western' sex workers, I had hoped a'ferninist methodology ... would help redresssomeof the
It
power issues.
ar
tr
To read more about Sarah's and Erin's research experiences, go to the Online ResourceCentre that
re
accompanies this book at http://www.oxfordtextbooks.co.uk/ore/brymansrm3el.
The significance of feminism in relation to values goes
further than this, however. In particular, several feminist
social researchers around the early 1980s proposed that
the principles and practices associated with quantitative
research were incompatible with feminist research on
women. For writers like Oakley (1981), quantitative
research was bound up with male values of control that
can be seen in the general orientation of the research
strategy-s-control of the research subject/respondent and
control of the research context and situation. Moreover,
the research process was seen as one-way traffic, in which
researchers extract information from the people being
studied and give little, or more usually nothing, in return.
For many feminists, such a strategy bordered on exploita-
tion and was incompatible with feminism's values of
sisterhood and non-hierarchical relationships between
women. The antipathy towards quantitative research
resulted in a preference for qualitative research among
feminists. Not only was qualitative research seen as more
consistent with the values of feminism; it was seen as
more adaptable to those values. Thus, feminist qualita-
tive research came to be associated with an approach
in which the investigator eschewed a value-neutral
approach and engaged with the people being studied as
people and not simply as respondents to research instru-
ments. The stance of feminism in relation to both quanti-
tative and qualitative approaches demonstrates the ways
in which values have implications for the process of social
investigation. In more recent years, there has been a soft-
ening of the attitudes of feminists towards quantitative
research. Several writers have acknowledged a viable and
acceptable role for quantitative research, particularly
when it is employed in conjunction with qualitative
research (Jayaratne and Stewart 1991; Oakley 1998).
This issue will be picked up in Chapters 16, 24, and 25.
There are, then, different positions that can be taken up
in relation to values and value freedom. Far fewer writers
overtly subscribe to the position that the principle of
objectivity can be put into practice than in the past.
Quantitative researchers sometimes seem to be writing in
a way that suggests an aura of objectivity (Mies 1993),
b'
but we simply do not know how far they subscribe to
such a position. There is a greater awareness today of the
limits to objectivity, so that some of the highly confident,
not to say naive, pronouncements on the subject, like
Durkheim's, have fallen into disfavour. A further way
in which values are relevant to the conduct of social
research is through the following ethical principles or
standards. This issue will be followed up in Chapter 5.
Practical considerations
Nor should we neglect the importance and significance of
practical issues in decisions about how social research
should be carried out. There are a number of different
dimensions to this issue. For one thing, choices of
research strategy, design, or method have to be dove-
tailed with the specific research question being investig-
ated. If we are interested in teasing out the relative
importance of a number of different causes of a social
phenomenon, it is quite likely that a quantitative strategy
will fit our needs, because, as will be shown in Chapter 6,
the assessment of cause is one of its keynotes. Alter-
natively, if we are interested in the world views of
members of a certain social group, a qualitative research
strategy that is sensitive to how participants interpret
their social world may be the direction to choose. If a
researcher is interested in a topic on which no or virtually
no research has been done in the past, the quantitative
strategy may be difficult to employ because there is little
prior literature from which to draw leads. A more
exploratory stance may be preferable, and, in this con-
nection, qualitative research may serve the researcher's
needs better, since it is typically associated with the gen-
eration rather than the testing of theory (see Table 1.1)
and with a relatively unstructured approach to the
research process (see Chapter 16). Another dimension
may have to do with the nature of the topic and of the
people being investigated. For example, if the researcher
needs to engage with individuals or groups involved in
illicit activities, such as violence (Patrick 1973), pilferage
(Ditton 1977), or drug dealing (P. A. Adler 1985), it is



,
'esearch
'Non-
lmeofthe
'at
ibscribe to
xlayofthe
confident,
ibject,like
.rther way
t of social
nciplesor
apterS.
.ificanceof
1research
fdifferent
:hoices of
I be dove-
ginvestig-
Ierelative
ofa social
te strategy
Chapter6,
tes. Alter-
views of
eresearch
;interpret
loose.If a
ir virtually
rantitative
ereislittle
. A more
I thiscon-
'searcher's
hthegen-
fable 1.1)
:h to the
Iimension
mdofthe
'esearcher
ivolvedin
,pilferage
985), itis
unlikelythatasocialsurveywouldgaintheconfidenceof
thesubjectsinvolvedor achieve the necessary rapport.
Itisnot surprising, therefore, that researchers in these
areashavetended to usea qualitative strategy. Bycon-
trast,itdoes not seem likelythat the hypothesis in the
researchdescribed inResearchin focus1.4could have
beentested with a qualitative method like participant
observation.
Studentexperience
I
I Apracticalconsiderationinthechoiceof
research method
I
I
!
One of thefactorsthatinfluencedRebeccaBarnes'schoice of thesemi-structuredinterviewfor her studyof
violenceinwomen'ssame-sexintimate relationships wasthat she felt thatthe topic isahighly sensitive
area and thatshe therefore needed tobe able toobserve her interviewees' emotional responses.
!
I feltthat,giventhe sensitivity of the researchtopic.semi-structured,in-depthinterviewswould be most I
appropriate.This gavemethe opportunity toelicit women'saccountsof abuse ina settingwhereI wasable !
to observetheiremotionalresponses to theinterviewand endeavourtominimize anydistress or other
negativefeelingsthatmightresultfrom participatingintheresearch.
t
To read more about Rebecca'sresearch experiences, go to the Online ResourceCentre that accompanies
this book at http://www.oxfordtextbooks.co.uk/orc/brymonsrm3e/.
Key points
Quantitative and qualitative research constitute different approaches to social investigationand
carry withthem important epistemological and ontological considerations.
Theorycanbedepictedassomethingthatprecedes research(asinquantitativeresearch)oras
somethingthat emergesoutofit(asinqualitativeresearch).
Epistemologicalconsiderationsloomlargeinconsiderationsofresearchstrategy.Toalargeextent,
these revolvearound the desirabilityof employinga natural science model (and in particular
positivism)versusinterpretivism.
Ontologicalconsiderations,concerningobjectivismversusconstructionism.alsoconstitute important
dimensionsofthequantitative/qualitativecontrast.
Values may impingeon the research process at different times.
Practicalconsiderations indecisions about research methods are also important factors.
Feminist researchers have tended to prefer a qualitative approach, though there issome evidence of
achangeofviewpointinthisregard.
Whilepractical considerationsmayseem rathermun-
dane and uninteresting compared with the lofty realm
lnhabitedbythephilosophicaldebates surroundingsuch
discussions about epistemologyandontology, they areim-
portantones.Allsocialresearchisacoming-togetherof the
idealandthefeasible.Becauseofthis,therewill bemany
circumstancesinwhich the natureofthe topicorofthe
subjects ofan investigation and the constraintson are-
searcherloomlargeindecisionsabouthowbesttoproceed.
1
t
I
J._




Questionsforreview
Theory and research
Ifyouhadto conductsomesocialresearchnow, whatwouldthe topicbeandwhat factorswould
haveinfluencedyourchoice?How importantwasaddressingtheoryinyourconsideration?
Outline,usingexamplesofyour own,the differencebetweengrandandmiddle-rangetheory.
What arethe differencesbetweeninductiveanddeductivetheoryandwhyisthe distinction
important?
Epistemological considerations
What ismeantbyeachofthe followingterms: positivism;realism;andinterpretivism?Why isit
importantto understand eachofthem?
What arethe implicationsofepistemologicalconsiderationsfor researchpractice?
Ontological considerations
Whatarethe maindifferencesbetween epistemologicalandontologicalconsiderations?
What ismeant byobjectivismandconstructionism?
Whichtheoreticalideashavebeenparticularlyinstrumentalinthe growthofinterestinqualitative
research?
Research strategy: quantitative and qualitative research
Outlinethe maindifferencesbetweenquantitativeandqualitativeresearchinterms of:the
relationshipbetweentheoryanddata;epistemologicalconsiderations;andontological
considerations.
Towhatextentisquantitativeresearchsolelyconcerned withtestingtheoriesandqualitative
researchwithgeneratingtheories?
Influences on the conduct ofsocial research
What aresomeofthe maininfluencesonsocialresearch?
Online Resource Centre
http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/
Visit the Online Resource Centre that accompanies thisbookto enrich yourunderstanding of social
research strategies. Consult weblinks, test yourselfusing multiple choice questions, andgainfurther
guidance andinspiration from the Student Researcher's Toolkit.



The nature of
qualitative research
366
370
Chapter outline
Introduction
The main steps in qualitative research
Theory and research 373
Concepts in qualitative research 373
Sampling in qualitative research 375
,
Reliability and validity in qualitative research 376
I
,

Adapting reliabilityand validityfor qualitative research 376
f"
Alternative criteria for evaluating qualitative research 377
Recent discussions about quality criteria for qualitative research 380
Between quantitative and qualitative research criteria 381
Overview of the issue of criteria 383
The main preoccupations of qualitative researchers 384
Seeing through the eyes of the people being studied 385
Description and the emphasis on context 386
Emphasis on process 388
Flexibility and limited structure 389
Concepts and theory grounded in data 390
The critique of qualitative research 391
Qualitative research is too subjective 391
Difficult to replicate 391
Problems of generalization 391
Lackof transparency 392
Is it always like this? 392
Some contrasts between quantitative and qualitative research 393
Some similarities between quantitative and qualitative research 394
Feminism and qualitative research 396
Keypoints 398
Questions for review 398
r
!
f
t
I
I
!
I
,
,
I
i



---
o
~ ~
.,

l
,

Chapter guide
Qualitativeresearchisa researchstrategythatusuallyemphasizes wordsratherthan quantification
inthe collectionand analysisofdata.Asaresearch strategy itisinductivist,constructionist,and
interpretivist,but qualitativeresearchersdonotalwayssubscribetoallthree ofthesefeatures.This
chapterisconcernedwith outliningthe mainfeatures ofqualitativeresearch,whichhasbecomean
increasinglypopularapproachtosocialresearch.Thechapterexplores:
themainstepsinqualitativeresearch; delineatingthe sequenceofstagesinqualitativeresearchis
more controversialthanwith quantitativeresearch.because itexhibitssomewhatlesscodification
oftheresearchprocess;
the relationshipbetweentheoryandresearch;
thenatureofconceptsinqualitativeresearch andtheir differencesfromconcepts inquantitative
research;
howfarreliabilityand validityare appropriatecriteriaforqualitativeresearchersandwhether
alternativecriteriathataremore tailored tothe research strategyare necessary;
themainpreoccupationsofqualitativeresearchers; fiveareasare identifiedintermsofanemphasis
on:seeingthroughtheeyesofresearchparticipants;descriptionandcontext; process;flexibilityand
lackofstructure;and conceptsandtheoryasoutcomesofthe research process;
some commoncriticisms ofqualitativeresearch;
themaincontrastsbetweenqualitativeand quantitativeresearch;
thestanceoffeminist researchersonqualitative research.
Introduction
WebeganChapter6bynotingthatquantitative research
had been outlinedin Chapter1as a distinctiveresearch
strategy. Much the same kind of general point canbe
registered in relation to qualitative research. In Chap-
ter 1 it was suggested that qualitative research differs
from quantitative research in several ways. Most obvi-
ously. qualitative research tends to be concerned with
words rather than numbers, but three further features
wereparticularlynoteworthy:
1. aninductiveview ofthe relationshipbetweentheory
andresearch,wherebythe fannerisgeneratedoutofthe
latter;
2. an epistemological position described as inter-
pretivist, meaning that, in contrast to the adoption of
a natural scientific model in quantitative research, the
stressisonthe understandingofthe socialworldthrough
an examinationoftheinterpretationofthat worldbyits
participants;and
3. an ontologicalposition described asconstructionist,
which impliesthatsocialpropertiesare outcomesofthe
interactions between individuals, ratherthan phenom-
ena 'out there' and separate from those involvedinits
construction.
As Bryman and Burgess (1999) observe, although
therehas been aproliferationofwritings onqualitative
researchsincethe 1970s,stipulatingwhatitisandisnot
asadistinctresearchstrategyisbynomeans straightfor-
ward.Theyproposethreereasonsforthisstateofaffairs.
1. Asaterm 'qualitativeresearch'issometimestakento
imply an approachtosocialresearchinwhichquantita-
tive dataarenotcollectedorgenerated.Manywriterson



----
I
ication
d
This
ean
irchis
fication
3tive
er
rphasis
lityand
f
Key concept 16.1
Four traditions of qualitative research
Gubriumand Holstein(1997) suggestfourtraditions ofqualitativeresearch.
1. Naturalism: seeksto understandsocialrealityinitsownterms;'asitreallyis';providesrichdescriptions of
people andinteraction innaturalsettings.
2. Ethnomethodology: seeksto understandhowsocial orderiscreatedthrough talkand interaction;hasa
naturalisticorientation.
3. Emotionalism: exhibitsaconcern withsubjectivityand gainingaccessto'inside'experience; is concerned
withthe innerrealityofhumans.
4. Postmodernism: emphasizes 'methodtalk';issensitivetothedifferentwayssocial realitycanbeconstructed.
For more on postmodernism see Keyconcept27.2.
We encountered the term 'naturalism' in Keyconcept2.4. Theuse of the term here is more or less the same as
the second meaningreferredto inKeyconcept2.4.Thenaturalisttradition hasprobablybeen the most
common one over the years. The second tradition willbe encountered inChapter20,whenwewill belooking
atanapproach tothe collectionofqualitativedataknownasconversationanalysis. Themore recent
postmodern standpoint willbe addressedin Chapter27. The third tradition----emotionalism-has notbecome
thefocusofasignificantstream ofresearchand will notbeemphasizedinthisbook.However,the mere
presenceofthesefourcontrastingtraditions pointstothe difficultyofcreatingadefinitiveaccountofwhat
qualitativeresearch isand is not.
qualitative research are critical of such a rendition of
qualitativeresearch,because(aswe willsee)thedistinc-
tiveness ofqualitativeresearchdoesnotresidesolelyin
theabsenceofnumbers.
2. Writers like Gubrium and Holstein (1997) have
'dbyits
suggested thatseveraldifferenttraditions in qualitative
researchcanbeidentified(see Keyconcept16.1).Inaddi-
rtionist, tion, Keyconcept16.2 suggeststhattherearedistinctive
softhe stanceswithinthequalitativeresearchtradition.
renom-
3. Sometimes,qualitativeresearchisdiscussedin terms
d in its
ofthe waysinwhichitdiffersfromquantitativeresearch.
Apotentialproblemwiththis tacticisthatit meansthat
qualitativeresearchendsup beingaddressedintermsof
though
litative
whatquantitativeresearchisnot.
j isnot
ghtfor-
Silverman (1993) has been particularly critical of
accounts of qualitative research that do not acknow-
iffairs.
ledgethevarietyof forms thatthe researchstrategycan
ikento
assume. Inother words, writers like Silvermanare critical
antita-
of attempts to specifythenatureof qualitativeresearch
ters on
as a general approach (see also Key concept 16.2).
However,unlesswe can talktoacertaindegreeaboutthe
natureofqualitativeresearch,itisdifficulttoseehowitis
possible to refer to qualitative research as a distinctive
researchstrategy.Inmuchthe samewaythatinChapter6
it was recognized that quantitative researchers employ
different research designs, in writing about the char-
acteristics of qualitative research we will need to be
sensitive to the different orientations of qualitative
researchers.Withoutat leasta senseofwhatiscommon
to a set of many if not most studies that might be
described as qualitative, the very notion ofqualitative
research would be rendered problematic.Yet it is clear
i
that,for manysocialscientists, it isa helpfulandmean- !
ingful category that can be seen in a variety of ways. t
i
Examplesare: the arrival of specialistjournals, suchas
i
QualitativeSociology,QualitativeResearch,Ethnography,
and Qualitative Inquiry; texts on qualitative research
(e.g. Silverman1993,2000; Seale 1999);aHandbookof
Qualitative Research (Denzin and Lincoln 1994, 2000,
2005a); andaseriesofbooksondifferentfacetsofquali-
tativeresearch (the SageQualitativeResearchMethods
I
Series).




Key concept 16.2
The Nine Moments of Qualitative Research
Denzin and Lincoln (2005b)havesuggested that qualitative research hasprogressed througha number of
stages. Theyportraythisasa history of qualitative research in North America. It isnot clearwhythe stages
arepresented asrelatingonly to North America, but the distinctionsareworth drawingattentionto because
they relateclosely to the suggestion that thereare different traditions of qualitativeresearch.
1. Thetraditionalperiod. The earlytwentieth century up to the Second World War.This phase refers tothe
work of social anthropologists and the Chicago School. It refers to in-depth studies of 'slices of life' that
portrayedthosewho were studiedasstrange or alien. It was heavilyimbuedwith positivism.
2. Modernistphase.Post-Second World War to early 1970s. Duringthis period, qualitativeresearchers builton
the work of the traditional period but at the same time soughtto enhance the rigour of qualitativeenquiries
and began to reflect on the natureof their craft. These investigations alsoshowed a tendency towards
positivism.
3. Blurred genres.1970-B6. Thiswasa periodwhen a varietyof epistemological andontological approaches,
aswell astheoretical ideas, were beingexploredasplausiblebases for qualitativeenquiries. According to
Denzinand Lincoln, we seein this perioda continuedproclivity towardspositivism, but with the beginnings
of aninterpretivist self-consciousness, influencedby Geertz's (1973a) insistence that qualitativeresearchers
areinvolvedin interpretationsof the interpretationsof thoseon whom they conduct their investigations.
4. Crisisofrepresentation. Mid-19BOs onwards. Most of the keywritings associated with thismomentoccurred
in the 1980s. It refers to a period in which qualitative social researchers in general (thoughmuchof the
writing stemmed initially from social anthropology) developedgreater self-awareness concerning in
particular the fact that their accounts of their fieldwork arejust one wayof representing realityandthat,
moreover, their representations areheavilyinfluencedby their social locations. The 'crisis of representation'
then isthe recognition that the researcher's written work has limited scientific authority. These ideas will be
encounteredagain in Chapter 27in the section 'Writing ethnography'.
The next three phases refer to 'a triple crisis' stemmingfromthe fourth moment above.
5. Postmodernperiodof experimental ethnographicwriting. Mid-1990s. Heavilyinfluenced by postmodernism
(seeKeyconcept 27.2), work under this headingischaracterized by an awareness of the different ways of
representing research participants(often referredto as 'the other') whenwriting up findings. Qualitative
researchers havetried different ways of representing the peopleon whomtheyconduct their investigations.
6. Post-experimentalenquiry. 1995-2000. Thisperiod isassociated mainlywith the emergence of AltaMira
Press, a publisherof qualitative research that encourages experimental and interdisciplinary writing.
It describes itself as havinga 'focuson interdisciplinarywork, breakinglong-standing boundaries'
(http://www.altamirapress.com/RlA/About/ (accessed on 18March2007)).
7. Themethodologicallycontestedpresent.2000-4. Thisrefers to a period in whichthere isconsiderable
disagreement about how qualitative research shouldbe conductedandthe directionsit shouldbe heading.
It isverymuchassociated with the arrival of journalslike QualitativeInquiryand QualitativeResearch that
provideforumsfor thesedebates. While Denzinand Lincoln datethis period as 2000-4, there isa great deal
of evidence to suggest that the contested methodological differences havenot abated. Oneof the areas
that hasbeenafocusof the ongoingdebates has beenthe issue of research qualitycriteria in relationto
qualitativestudies.
8. Now.2005-. Thisperiod ischaracterized by a backlash against qualitativeresearch with a reassernon in
governmentcircles of the valueof traditional science. Some of thesepressures arereviewedin Bryman
(20070).



rof
1ges
xause
athe
hat
builton

'ds
aches,
ingto
ginnings

tions.
ccurred
:he
that,
entation'
.s will be
ernism
avsof
ative
igations.
lira
e
eading.
)that
eatdeal
'eas
nto
1 in
an
Several reasons might be proposed for theunease among
somewritersconcerningthespecificationofthenatureof
qualitative research. Two reasons mightbe regarded as
havingparticularimportance.First, qualitative research
subsumes several diverse research methods that differ
fromeachotherconsiderably.Thefollowingare the main
researchmethodsassociatedwithqualitativeresearch.
Ethnography/participant observation. While some
cautionisadvisablein treatingethnographyandpar-
ticipant observation as synonyms, in many respects
theyreferto similarifnot identicalapproachestodata
collection in which the researcher is immersed in a
social setting for some time in orderto observe and
listenwith a view to gainingan appreciation of the
culture of a social group. It has been employed in
such social researchclassics asWhyte's (1955) study
ofstreetcomerlife in a slumcommunityandGans's
(1962) research on a similargroup in the throes of
urbanredevelopment.
Qualitative interviewing. This is a very broad termto
describea wide rangeofinterviewingstyles (see Key
concept8.2 for an introduction). Moreover, qualita-
tiveresearchersemployingethnographyorparticipant
observationtypicallyengagein a substantialamount
ofqualitativeinterviewing.
Focusgroups(seeKeyconcept8.2).
Language-based approaches to the collection of
qualitative data, such as discourse and conversation
analysis.
9. The fractured future. Lincoln andDenzin(2005:1123) alsospeculate aboutwhattheimmediatefutureholds:
'Randomized fieldtrials... will occupy the timepfone group of researcherswhilethe pursuit of asocially
andculturallyresponsive, communitarian, justice-orientedsetof studieswill consumethemeaningful
workingmomentsof theother.'
Thistimelineof phasesisuseful becauseit highlightsthedifficulty of characterizing 'qualitative research'.
AsSilverman(1993) observes, the term covers a number of different researchmethods and approaches to
qualitative data thatdifferconsiderably. On theother hand, Denzin and Lincoln's 'moments' have to be treated
withsomecaution.First, ithastobeborneinmindthatworkthatcouldbedepicted intermsverysimilar
to the first two phases continues to be conducted. Indeed, many of the qualitative investigationsthat serve as
illustrationsin Part Three are of this type. Although qualitative researchersmay be more self-conscious
nowadaysabout theirinfluenceonthe researchprocessandthe significanceofhowtheywrite,many
qualitative studies arestillcharacterizedbyrealism, atleasttosome degree.Second,DenzinandLincoln's
laterphasesareassociated toomuch with particularevents-thearrival of anew publisher ornewjournals
-whichlooksstrangewhen viewed in relation totheseveral decadeswith which theearlier momentsare
associated. Third,theirninth and final moment seems to be concerned with a riftinsocial research in general
ratherthanwithinqualitativeresearchassuch.
The collection and qualitative analysis of texts and
documents.
Each of these approaches to data collection will be
examined in PartThree. The picturewithregardto the
I
verydifferentmethodsandsourcesthatcomprisequalita-
tiveresearchismadesomewhatmorecomplexbythefact
that a multi-method approach is frequently employed.
As noted above, researchersemployingethnographyor I
participant observation frequently conduct qualitative
interviews.However,theyalso oftencollectandanalyse
texts anddocumentsaswell. Thus,thereisconsiderable
I
variabilityinthecollectionofdataamongstudiesthatare
I
typicallydeemed to be qualitative. Ofcourse, quantita-
tiveresearchalso subsumesseveraldifferentmethodsof
datacollection(thesewerecoveredinPartTwo),butthe
inclusionofmethodsconcernedwiththe analysisoflan-
I
guageasaformofqualitativeresearchimpliessomewhat I
I
greatervariability.
A secondreasonwhythereissomeresistancetoadelin-
eationof the natureofqualitativeresearchisthatthecon-
nectionbetweentheoryandresearchissomewhatmore
ambiguousthaninquantitativeresearch.With the latter
researchstrategy,theoreticalissuesdrivetheformulation
ofaresearchquestion,whichintumdrives thecollection
and analysis of data. Findings then feed back into the
relevant theory. This is rather a caricature, because
whatcountsas'theory'issometimeslittle morethanthe
researchliteraturerelatingto a certainissue or area. In
qualitativeresearch,theoryissupposedtobeanoutcome
ofan investigationratherthansomethingthatprecedes




it. However, some writers, like Silverman (1993: 24),
have argued that such a depiction of qualitative research
is 'out of tune with the greater sophistication of contem-
porary field research design, born out of accumulated
knowledge of interaction and greater concern with issues
language that will be examined in Chapter 20. HOWev
qualitative research is more usually regarded as
an approach in which theory and categorization em g
erge
out of the collection and analysis of data. The more
I
. bel de i th
era point emg rna e IS at sue
of reliability and validity'. This is particularly the case qualitative research may account for the
with conversation analysis, an approach to the study of
h diff gen
a luerence with'
unease
depicting the research strategy interms of a set of stag
es.

. . : .
;1
Themainstepsinqualitativeresearch
The sequence outlined in Figure 16.1 provides a repre-
sentation of how the qualitative research process can be
visualized. In order to illustrate the steps, a published
study by Foster (1995) of crime in communities will be
used. This study was previously encountered inResearch
in focus 1.6.
Anoutlineofthemain stepsofqualitativeresearch
Step 1. General research question(s). The starring point
for Foster's (1995) studyof crime incommunities, par-
ticularlyones that contain predominandypublic hous-
ing, is the high levels of crime in poorer areas. Tothe
extent that itis a focus of attention, it is frequently
assumed that communities with high levels of crime
6_ \Nriting upfindings/conclusions
",'.,i"_," .

I
l
f
tend to have low levels of social control. But Foster

argues that we know very little about how informal
'n emerge,
social control operates in such communities and what
noregen.;
itssignificancefor crime is. She also notes that council
ce Within: estates are frequently presumed to be crime prone but
ase about> that there is little evidence on 'the diversity in experi-
ofStages. ence and attitudes of residents within individual
estates' (Foster 1995: 563). It would be easy to pre-
sume that, to the extent that council estates are prone
However
ngpoint
Thinking deeply 16.1
ties,par-
to high crime levels, they exhibit low levels of social
control. Thus Foster formulates a general set of con-
cerns revolving around council estates and their
crime-proneness and the possiblerole and dynamics of
social control in the process. She also notes that some
writers have suggested that the propensity to crime in
council estates may be in part attributed to flawsin the
design of the estates.
lichous_
Research questions in qualitative research
;. Tothe

Research questionsinqualitative research are stated withvarying degreesof explicitness, Sometimes, the
)f crime
researchquestion isembedded within a general statement ofthe orientation ofan article. Thus, the author
ofthe research covered below inResearch infocus 16.3writes at the beginning ofa longparagraph:
Themainproposition inthis article isthat different masculinities are producedthrough performances
that drawon the different cultural resources that are available ineach setting. (Swain 2004: 167)
Othersopt fora more explicit treatment of researchquestions. Ashforth et 01. (2007) wereinterested inthe
phenomenon of 'dirtywork', a termfirst introduced nearly fifty yearspreviously to referto work that istainted
'physically, socially or morally' (Hughes 195B: 122; quoted inAshforth et 01.2007: 149). The researchers
conductedsemi-structured interviews with managers ineighteensuchoccupations inorder to explorehow
the work is'normalized', that is, how they developways ofdealing with or reducing the significance ofthe taint
ofdirty work. After a discussion ofthe literature and their view of itsimplications fortheirownwork, theywrite:
In summary, our research questions were:
Research Question 1. What normalization challenges do managers in dirty workoccupations face?
Research Question 2. What tactics do managers report using to normalize dirty work? (Ashforth et 01. 2007:
151; italicized inoriginal)
Onefactorthat may affect the degreeofexplicitness withwhich research questionsare stated isthe outlet in
which the research ispublished. Ashforth et 01. (2007) published this article inthe Academy of Management
Journal, which inthe past hastended to publish mainly empirical articles deriving from quantitative research.
Itmay be that Ashforth et 01. chosethisformatfor presentingtheir research questions sothat it would exhibit
someofthe characteristics of research questionsor hypotheses inquantitative research that tend to be stated
explicitly. Another article inthe same issue alsostated the research questions very explicitly thoughtheywere
not formatted to standout inthe sameway.
Thus, tworesearch questions guided thisarticle: (1)Whatconditions trigger organizational stakeholders
and leaders to engageinsensegiving activities? and (2)Whatconditions enablesensegiving on the part of
stakeholders and leaders motivated to engagein senseglving activities? (Maitlis and Lawrence 2007: 59)
Theresearchers went on to investigate these two researchquestions by collecting qualitative data from semi-
structuredinterviews, observation and examination ofdocuments inconnection withthree British symphony
orchestras.


Step 2. Selection ofrelevant site(s) and subjects. The
research was conducted on a London council estate
(with the fictitious name 'Riverside'), which had a high
level of crime and which exhibited the kinds of hous-
ing features that are frequently associated with a
propensity to crime. Relevant research participants,
such as residents, were identified.
Step3. Collection of relevantdata. Foster describes her
research as 'ethnographic'. She spent eighteen months
'getting involved in as many aspects oflife there as pos-
sible from attending tenant meetings, the mothers and
toddlers group, and activities for young people, to
socializing with some of the residents in the local pub'
(Foster 1995: 566). Foster also tells us that 'extended
interviews' were conducted with forty-five residents of
Riverside (and another London estate, but the major-
ity were from Riverside) and twenty-five 'officials',
such as police and housing officers. Foster's account of
her research methods suggests that she is likely to have
generated two types of data: fieldwork notes based on
her ethnographic observation of life in the community
and detailed notes (and most probably transcripts) of
interviews undertaken.
Step4. Interpretationofdata. One of the key findings to
emerge from the data is the fact that, in spite of the fact
that Riverside has a high crime rate, it is not perceived
as a problem in this regard by Riverside residents. For
example, she quotes from an interview with an elderly
tenant: 'They used to say that they couldn't let the flats
[apartments] here ... but I mean as far as muggings
or anything like that you don't hear of nothing like that
even now' (Foster 1995: 568). Instead, housing prob-
lems loomed larger in the minds of residents than
crime. She also found that 'hidden economy' crimes
were prevalent on the estate and that much crime was
tolerated by residents. She also observes that, contrary
to expectations about estates like Riverside, there was
clear evidence of informal social control mechanisms
at work, such as shaming practices.
Step5. Conceptualandtheoreticalwork.Nonew concepts
seem to emerge from Foster's research, but her find-
ings enable her to tie together some of the elements
outlined above under Step 1.For example, she writes:
Crime then need not be damaging perseproviding
other factors cushion its impact. On Riversidethese
included support networks in which tenants felt that
someone was watching out for their properties and
provided linkswith people to whom they could turn
ifthey were introuble. Consequently while
'generalized fears about crime remained prevalent
familiarity and support went some way to r e u i n ~
the potential for hostileencounters. (Foster 1995:580)
Itis this step, coupled with the interpretation of data
that forms the study's findings. '
Steps Sa. Tighter specification ofthe research ques,
tion(s), and5b. Collectionoffurtherdata. There is no
specific evidence from Foster's account that she fcl.
lowed a process in which she collected further data
after she had built up early interpretations of her data.
When this occurs, as it sometimes does in research
within a grounded theory framework, there canbean
interplay between interpretation and theoriZing, on
the one hand, and data collection, on the other. Sucha
strategy is frequently referred to as an iterativeone.
She does write at one point that some residents and
officials were interviewed twice and in some cases
even three times in the course of her research. This
raises the possibility that she was re-interviewing cer-
tain individuals in the light of her emerging ideas
about her data, but this can only be a speculation.
Step6. Writing upfindings/conclusions.There is noreal
difference between the significance of writing up in
quantitative research and qualitative research, so that
exactly the same points made in relation to step 11 in
Figure 6.1 apply here. An audience has to be convinced
about the credibility and significance of the interpreta-
tions offered. Researchers are not and cannot be sim-
ply conduits for the things they see and the words they
hear. The salience of what researchers have seen and
heard has to be impressed on the audience. Foster does
this by making clear to her audience that her findings
have implications for policies regarding estates and
crime and for our understanding of the links between
housing, community, and crime. Akey point to emerge
from her work, which she emphasizes at several points
in the article and hammers home in her concluding
section, is that being an insider to Riverside allowed
her to see that a community that may be regarded by
outsiders as having a high propensity towards crime
should not be presumed to be seen in this way bymem-
bers of that community.
Two particularly distinctive aspects of the sequence of
steps in qualitative research are the highly related issues
of the links between theory and concepts with research
data. It is to these issues that we now tum.
I
Mo:
crat
thir
of '
ofl
anc
dai
rlv
tio
pa
tiv
th
gr
re
01
cc
tt
I




Theoryandresearch
I of data,
ch ques.
ere is no
she fol,
rer data
lerdata.
'esearch
anbean
~ i n g on
'.Such a
ive one.
nts and
.e cases
~ h This
ing cer-
g ideas
m,
no real
~ up in
so that
p 11 in
vinced
rpreta-
>e sim-
:Is they
en and
rrdoes
rdings
es and
tween
merge
points
uding
lowed
ed by
crime
mem-
Iceof
ssues
earch
Most qualitative researchers when writing about their
craftemphasize a preference for treating theory as some-
thing that emerges out of the collection and analysis
of data. As will be seen in Chapter 22, practitioners
of grounded theory-a frequently cited approach to the
analysis of qualitative data-especially stress the import-
anceof allowing theoretical ideas to emerge out of one's
data. But some qualitative researchers argue that qualita-
tivedata can and should have an important role in rela-
tionto the testingof theories as well. Silverman (1993), in
particular, has argued that in more recent times qualita-
tive researchers have become increasingly interested in
the testing of theories and that this is a reflection of the
growing maturity of the strategy. Certainly, there is no
reason why qualitative research cannot be employed in
order to test theories that are specified in advance of data
collection. In any case, much qualitative research entails
the testing of theories in the course of the research pro-

A central feature of Chapter 6 was the discussion of

concepts and their measurement. For most qualitative
researchers, developing measures of concepts will not be
a significant consideration, but concepts are very much
part of the landscape in qualitative research. However,
the way in which concepts are developed and employed is
often rather different from that implied in the quantita-
tive research strategy. Blumer's (1954) distinction
between 'definitive' and 'sensitizing' concepts captures
aspects of the different ways in which concepts are
thought about.
Blumer (1954) argued stridently against the use of
definitive concepts in social research. The idea of defin-
itive concepts is typified by the way in which, in quantita-
tive research, a concept, once developed, becomes fixed
through the elaboration of indicators. For Blumer, such
an approach entailed the application of a straitjacket on
the social world, because the concept in question comes
to be seen exclusively in terms of the indicators that have
been developed for it. Fine nuances in the form that the
cess. So, in Figure 16.1, the loop back from Step Sa,
'Tighter specification of the research questionfs)', to Step
5b, 'Collection of further data', implies that a theoretical
position may emerge in the course of research and may
spur the collection of further data to test that theory. This
kind of oscillation between testing emerging theories and
collecting data is a particularly distinctive feature of
grounded theory. It is presented as a dashed line in Figure
16.1, because it is not asnecessary a feature of the process
ofqualitative research as the other steps.
One key point that is implied by Figure 16.1 is that
the typical sequence of steps in qualitative research
entails the generation of theories rather than the testing
of theories that are specified at the outset. Silverman
(1993) is undoubtedly correct that pre-specified theories
can be and sometimes are tested with qualitative data,
but the generation of theory tends to be the preferred
approach.
Conceptsinqualitativeresearch
concept assume or alternative ways of viewing the
concept and its manifestations are sidelined. In other
words, definitive concepts are excessively concerned with
what is common to the phenomena that the concept
is supposed to subsume rather than with variety. Instead,
Blumer (1954: 7) recommended that social researchers
should recognize that the concepts they use are sensitiz-
ing concepts in that they provide 'a general sense of ref-
erence and guidance in approaching empirical instances'.
For Blumer, then, concepts should be employed in such a
way that they give a very general sense of what to look for
and act as a means for uncovering the variety of forms
that the phenomena to which they refer can assume.
In providing a critique of definitive concepts, it is clear
that Blumer had in mind the concept-indicator model
described in Chapter 6. In other words, his views entailed
in large part a critique of quantitative research and a pro-
grammatic statement that would form a springboard for
an alternative approach that nowadays we would recog-
nize as qualitative research.





I
Blumer's distinction is not without itsproblems. Itis ordainedschemesonthesocialworldchimeswithth
atof
not at all clearhow far a very generalformulation ofa manyqualitative researchers.AstheexampleinRes h
.
e
concept can be regarded as a useful guide to empirical in focus 16.1 suggests, the researcher frequentlysta

enquiry. If itistoogeneral,itwillsimplyfailtoprovidea outwithabroadoutlineofaconcept,whichisrevised tts

and
useful starting point because its guidelines are too broad;
narrowed during the course of data collection. ForSub.
iftoonarrow,itislikelytorepeatsomeofthedifficulties sequent researchers, the concept maybe takenupand
Blumer identified in relation to definitive concepts. revised as it is employed in connection with different
However,hisgeneralviewofconceptshasattractedsome social contexts or in relation to somewhat different
support, because his preference for not imposing pre- researchquestions.
Research in focus 16.1
The emergence of a concept in qualitative
research: the case of emotionallabour
Hochschild's (1983) idea of emotional labour-labour that 'requiresone to induceor suppressfeelings inorder
to sustainthe outwardcountenancethatproducestheproper state ofmindinothers'(1983: 7)-hasbecome a
veryinfluentialconcept inthesociologyofworkand inthedevelopingarea ofthesociology ofemotions.
Somewhat ironicallyforapredominantlyqualitativestudy, Hochschild'sinitialconceptualizationappearsto
haveemergedfrom aquestionnaire shedistributedto261 universitystudents. Withinthequestionnairewere
two requests: 'Describea real situationthat was importantto youinwhichyouexperienced a deepemotion'
and 'Describeasfully and concretely aspossible a real situation that was important to youinwhich youeither
changedthe situation to fit yourfeelings or changedyourfeelings to fit the situation'(1983: 13), Thus, although
aself-completion questionnaire wasemployed,theresultingdata werequalitative.Thedata wereanalyzed
interms of theideaof emotionwork, which isthe same as emotional labourbutoccurs inaprivate context.
Emotional labourisessentially emotionworkthat is performed as part of one's paidemployment. Inorderto
developthe idea of emotionallabour, Hochschild lookedto theworld ofwork. Themain occupation shestudied
wastheflightattendant. Severalsourcesof data on emotionallabouramongflightattendantswereemployed,
Shegainedaccessto DeltaAirlines,alarge Americanairline, and inthecourseof herinvestigationsshe:
watched sessions for training attendants and hadmany conversations with bothtraineesandexperienced
attendantsduringthesessions;
interviewed various personnel, suchas managersinvarious sections, and advertising agents;
examinedDeltaadvertisementsspanningthirtyyears;
observedtheflight attendant recruitmentprocessat PanAmerican Airways, sinceshe had not beenallowed
todothisatDelta;
conducted 'open-endedinterviews lasting three tofive hourseach withthirtyflight attendants intheSan
Francisco BayArea'(1983:15).
Inordertoforge acomparisonwithacontrastingoccupationalgroupthatisnonethelessalsoinvolved in
emotional labour,Hochschild alsointerviewedfive debt-collectors. Inherbook, she explores suchtopics asthe
humancostsofemotionallabourand theissueof gender inrelationto it. ItisclearthatHochschild'sconcept of
emotional labour beganas a somewhatimprecise idea that emergedout of aconcern with emotion work and
that wasgraduallydeveloped inorder to address itswidersignificance.Theconcept has beenpickedup by
other qualitative researchersinthesociology ofwork. Forexample, Leidner (1993) has explored through
ethnographicstudiesof aMcDonald'srestaurant and an insurancecompany the waysinwhichorganizations
seekto 'routinize' thedisplay ofemotional labour.
III thi
disc
u
appa
chap
purp
ofS<
unit:
dep;
que:
p
ing
sal
it
re
pr
sa
SE
q
p
"

01



different
n order
Come a
s to
e were
tion'
~ i t r
hough
llyzed
xt
rto
tudied
lyed.
ced
owed
In
.the
otot
nd
1S
iththat of
Research
.tlystarts
vised and
For sub.
1 up and
different
Inthis book, sampling issues in qualitative research are
discussed mainly in Chapters 17 and 18. As will become
apparent from the sections on sampling in these two
chapters, qualitative researchers typically engage in
purposive sampling (see Key concept 17.4). This type
of sampling is essentially to do with the selection of
units (which may be people, organizations, documents,
departments, etc.), with direct reference to the research
questionsbeing asked.
Probability sampling may be used in qualitative re-
search(e.g. Rafaeli et al. 1997), though this is more likely
to occur in interview-based rather than in ethnographic
qualitative studies. There is no obvious rule of thumb that
mightbe used to help the qualitative researcher in decid-
ingwhen it might be appropriate to employ probability
sampling, but two criteria might be envisaged. First, if
it is highly significant or important for the qualitative
researcher to be able to generalize to a wider population,
probability sampling is likely to be a more compelling
sampling approach. This might occur with audiences for
one's work for whom generalizability in the traditional
sense of the word is important. Second, if the research
questions do not suggest that particular categories of
people (or whatever the unit of analysis is) should be
sampled, there may be a case for sampling randomly. This
is not a likely or common occurrence, but, if it occurs,
probability sampling warrants consideration.
When purposive sampling takes place in qualitative
research, it often occurs at more than one level. For exam-
ple, in the study of British symphony orchestras (Maitlis
and Lawrence 2007) referred to briefly in Thinking
deeply 16.1, there were two levels of sampling.
1. Sampling oforchestras. Three orchestras were selected,
with the choice being based 'on the study's aims, seeking
a balanced sample of organizations in which issues and
processes could be compared but in which heterogeneity
was also significant' (Maitlis and Lawrence 2007: 60).
Maitlis and Lawrence argue that the three orchestras are
based on three of the main categories of orchestra in
Britain. These three types are based on the governance
structure of the orchestra and its sources of revenue. The
authors also point out that they selected only orchestras
that were neither the weakest nor the strongest (aestheti-
cally or financially) 'to increase the study's generalizabil-
Samplinginqualitativeresearch
ity' (Maitlis and Lawrence 2007: 60). What we see here
in the selection of units of analysis is a combination of
selection for heterogeneity (governance structure and
revenue) and homogeneity (mid-level performance). In
this way, the researcher is able to exert some control over
the kinds of comparison that can be forged.
Z. Samplingof interviewees. Interviews were conducted
with: each orchestra's executive director and all members
of its management team; its main conductor; and those
musicians who were representatives on the orchestra's
board and its orchestra committee, which represents
musicians' interests. In addition, interviews were con-
ducted with long-term members of the orchestra or who
had previously been members of the board or orchestra
committee. Musicians who expressed an interest were
also interviewed (these latter interviews would be more
characteristic of convenience rather than purposive sam-
pling). A total of 120 interviews were conducted. In addi-
tion, the researchers attended meetings and rehearsals
and examined numerous documents.
Sometimes, when conducting a mixed methods in-
vestigation involving both quantitative and qualitative
research, the findings from a survey might be used as the
basis for the selection of a purposive sample. For ex-
ample, in a study of social policy researchers in the UK, an
e-survey was conducted that sought respondents' views
on a wide variety ofissues concerning criteria for evaluat-
ing the quality of social policy research (Bryman et al.
in press; Sempik et al. 2007). Respondents were asked
whether they would be prepared to be interviewed by
telephone so that issues could be probed more deeply and
other issues that had not been explored in the e-survey
could be addressed. Of the 251 researchers who replied to
the online questionnaire 90 agreed to be interviewed. On
the basis of their replies, 28 of the 90 researchers were
interviewed by telephone using a semi-structured inter-
view approach. The 28 interviewees were selected to
reflect a variety of orientations to social policy research
and to the evaluation of research quality. For example,
one criterion was derived from where the respondent
stood on the issue of whether he orshe felt that social pol-
icy research should contribute to policy and practice or to
knowledge or to a combination of both. This sampling
strategy allowed interviewees to be selected purposively



I
\, ('
I
I .1
I
)1
. ~
( ;

research
InChapters2and 6itwasnotedthatreliabilityandvalid-
ityareimportantqiteriainestablishingandassessingthe
quality ofresearchforthe quantitativeresearcher.How-
ever, there hasbeensomediscussion among qualitative
researchers concerning their relevance for qualitative
research. Moreover, even writers who do take the view
thatthe criteriaarerelevanthaveconsideredthe possib-
ility thatthe meanings ofthe terms need to be altered.
Forexample, the issueofmeasurementvalidityalmost
bydefinition seems to carry connotations of measure-
ment. Since measurement is not a major preoccupa-
tion among qualitativeresearchers, the issue ofvalidity
would seem to have little bearing on such studies. As
foreshadowed brieflyin Chapter2, a numberof differ-
entstanceshavebeentakenbyqualitative researchersin
relationtothese issues.
Adaptingreliabilityandvalidity for
qualitativeresearch
One stance is to assimilate reliability and validity into
qualitative researchwith little change ofmeaningother
thanplaying down the salience ofmeasurementissues.
Mason (1996: 21), for example, inher bookon qualita-
tiveresearch,arguesthatreliability,validityandgeneral-
izability(which isthe main componentofexternal validity
-seeChapter2)'are differentkinds ofmeasures ofthe
quality,rigourandwiderpotentialofresearch,whichare
achieved according to certain methodological and dis-
ciplinary conventionsandprinciples'. She sticksvery close
to the meaning thatthese criteria have in quantitative
research,wheretheyhavebeenlargelydeveloped. Thus,
validityrefers towhether'youareobserving,identifying,
or"measuring"whatyousayyouare' (Mason1996: 24).
LeCompteandGoetz(1982)and Kirkand Miller(1986)
also write about reliability and validity in relation to
qualitativeresearchbutinvestthetermswithasomewhat
found in the next two chapters. Whatlinksthem' h
IS t at
thesamplingisconductedwith reference tothego I
a sof
theresearch,sothatunitsofanalysisareselectedinte
of criteria that will allow the research questions o ~ ;
answered.
interms ofcriteriathatwerecentraltothemain topicof
theresearch-theappraisalofresearch quality.
Theterm 'purposive sampling' coversawidevarietyof
approaches, as this brief discussion should make dear.
Furtherexamplesand ways of thinking about it can be
Reliabilityandvalidityinqualitative
different meaning from Mason. LeCompte and Goetz
writeaboutthefollowing.
Externalreliability, by which they mean the degree
to which astudy can bereplicated. ThisisadiffiCUlt
criterion to meet in qualitative research, since, as
LeCompte and Goetz recognize, it is impossible to
'freeze' a social setting andthecircumstances ofanini-
tialstudytomakeitreplicableinthesenseinwhichthe
term isusually employed (seeChapter6). However.
theysuggest severalstrategiesthatcanbeintroduced
in order to approach therequirements of external reli-
ability. For example, they suggest that a qualitative
researcher replicating ethnographic research needs
to adopt a similar socialrole to thatadopted bythe
original researcher.Otherwisewhataresearchercon-
ductingareplicationseesand hearswillnotbecom-
parabletotheoriginalresearch.
Internal reliability, by which they mean whether.
when thereismorethanoneobserver,membersofthe
research team agree aboutwhatthey see and hear.
This is a similar notion to inter-observer consistency
(seeKeyconcept6.3).
Internalvalidity,bywhichtheymeanwhether there
is a good match between researchers' observations
and the theoretical ideas they develop. LeCompte
and Goetz argue that internal validity tends to be
astrengthofqualitative research, particularlyethno-
graphicresearch, becausetheprolongedparticipation
inthe sociallifeofagroupoveralongperiodoftime
allowstheresearchertoensureahighlevelofcongru-
encebetweenconceptsandobservations.
External validity,whichrefers to the degree to which
findings can be generalized across social settings.
LeCompteandGoetzarguethat,unlikeinternalvalid-
ity, externalvalidityrepresentsaproblemforqualitative
,
I
L



id Goetz
~ degree
difficult
:ince, as
;sible to
)fanini-
hichthe
owever,
~ o u e
nalreli-
ditative
1 needs
bythe
iercon-
Jecom-
hether,
s ofthe
:Ihear.
istency
rthere
'ations
ompte
to be
ethno-
)ation
ftime
ngru-
vhich
tings.
valid-
tative
emisthat
tegoalsof
'dintenns
onstobe
researchersbecauseoftheirtendencyto employcase
studiesand smallsamples.
As thisbrieftreatmentsuggests, qualitative researchers
have tendedto employthe termsreliabilityandvalidityin
very similarways to quantitativeresearcherswhenseek-
ingtodevelopcriteriaforassessingresearch.
Alternativecriteriaforevaluating
qualitativeresearch
However,asecondpositionin relationto reliabilityand
validityin qualitative research canbe discerned. Some
writershavesuggestedthatqualitativestudiesshouldbe
judgedor evaluatedaccordingto quitedifferentcriteria
from those usedbyquantitativeresearchers.Lincoln and
Guba(1985) andGubaandLincoln(1994) proposethat
itisnecessary to specify terms and ways ofestablish-
ingand assessingthequalityofqualitativeresearchthat
provide an alternative to reliability and validity. They
proposetwo primarycriteria for assessing a qualitative
study:trustworthiness andauthenticity.
I
t
Trustworthiness [
to themembersofthe socialworldwhowerestudiedfor
Trustworthinessismadeupoffour criteria,eachofwhich confirmation that the investigator has correctly under-
hasanequivalentcriterioninquantitativeresearch: stood that social world. This latter technique is often
referred to as respondentvalidation or member valida-
1. credibility,whichparallelsinternalvalidity;
tion (seeKeyconcept16.3).Anothertechniquetheyre-
2. transferability, whichparallelsexternalvalidity; commendistriangulation (see Keyconcept16.4).
Keyconcept16.3
Whatis respondentvalidation?
Respondent validation, which is also sometimes called member validation, isa process whereby a researcher
providesthe peopleonwhomheorshehasconductedresearch withanaccount ofhisorherfindings.
The aimof theexerciseistoseekcorroborationorotherwiseoftheaccount thatthe researcher hasarrived
at. Respondent validation has been particularlypopular among qualitative researchers, because they
frequently want to ensure that there is a good correspondence between their findingsand the perspectives
andexperiences of theirresearch participants.Theformthatrespondentvalidationcanassumevaries.
Thereareseveraldifferentformsofrespondentvalidation.
Theresearcherprovides each research participantwithanaccountofwhatheorshehassaidtothe
researcherinaninterviewandconversations,orofwhatthe researcher observed bywatchingthat person
in thecourse of an observational study. For example, Bloor(1978, 1997)reports that he carried out
observations of ear, nose, and throat (ENT)consultants concerning their approaches to makingdecisions
abouttheassessment ofpatients.Hesubmitted areporttoeach consultantonhis orherpractices.
3. dependability, whichparallelsreliability;
4 . confinnability, whichparallelsobjectivity.
Amajorreasonfor GubaandLincoln's uneaseaboutthe
simpleapplicationofreliabilityandvaliditystandardsto
qualitativeresearchisthatthe criteriapresupposethata
single absolute account of social reality is feasible. In
otherwords, theyarecritical ofthe view (described in
Chapter1as realist) thatthereareabsolutetruthsabout
thesocialworldthatit isthe jobofthe socialscientistto
reveal. Instead, theyargue thattherecanbe morethan
one andpossiblyseveralaccounts.
Credibility
The significance of this stress on multiple accounts of
social realityisespeciallyevidentin thetrustworthiness
criterion ofcredibility. Merall, if there canbe several
possible accounts of anaspect of social reality, it is the
feasibilityor credibilityof the accountthata researcher
arrives at thatis going to determineits acceptabilityto
others.Theestablishmentof thecredibilityof findingsentails
bothensuringthatresearchiscarriedoutaccordingtothe
canonsofgoodpracticeand submittingresearchfindings




The researcherfeeds backto a groupof people or an organization his or herimpressions and findings in
relationtothatgroupororganization.Bloor saysthat,for hisresearchon therapeuticcommunities
he conducted groupdiscussions(whichwere taped) withcommunitymembersto gaugereactions todrah
researchreports.
The researcherfeeds backto a groupof peopleor an organization someofhis or herwritings that arebase
on astudyofthatgroupororganization(e.g, articles, bookchapters).Ball (1984) askedteachers ina
inwhichhehadconductedethnographicresearchtocomment ondraftarticlesand chapters,andsimilarl
Willis (1977)askedthe youngworking-classmaleswhowere thefocusof his ethnography tocommenton Y
draftchapters, as didSkeggs (1994) for herparallel studyof youngworking-class women (see Researchin
focus 17.8forfurtherdetails).
Ineach case, thegoal isto seekconfirmationthatthe researcher'sfindingsand impressions are congruent with
theviews ofthose on whomtheresearchwasconducted and to seek out areas inwhichthereisalack of
correspondenceand the reasonsfor it.However,theidea is not without practicaldifficulties.
Respondentvalidation mayoccasion defensive reactions on the part of research participants andeven
censorship.
Bloor (1997: 45)observesthat, becausesome approaches to enquirymayresultinresearchparticipants
developing relationships withtheresearcherof 'fondnessand mutual regard',theremaybe areluctance to
becritical.
Itishighly questionable whether researchparticipants can validatea researcher'sanalysis, sincethis entails
inferences beingmadefor an audience ofsocial sciencepeers.This meansthat, eventhoughthefirsttwo
methods of respondent validation mayreceivea corroborative response,the researcherstill hastomake a
further leap,throughthe developmentof conceptsand theories, inproviding a social scienceframe for the
resulting publications. If the thirdmethodof respondentvalidation is employed, it isunlikely that thesocial
scientific analyseswill be meaningfulto researchparticipants.Hobbs(1993) fed backsomeof hiswritingson
entrepreneurshipinLondon's EastEnd to hisinformants, and it isclearthat they madelittle senseofwhat
he had written.Similarly, Skeggs (1994: 86)reports:. "Can't understandabloody worditsays" was the most
common response' (seeResearch infocus17.8 forfurther detailsofthisstudy).
an
bo
re:
va
Tt
51:
a(
el
Transferability
Becausequalitative research typicallyentails the inten-
sivestudyofasmallgroup, orofindividuals sharing cer-
taincharacteristics(thatis,depthratherthanthebreadth
thatisapreoccupationinquantitativeresearch),qualita-
tivefindingstendtobeorientedtothecontextualunique-
ness and significance of the aspect of the social world
being studied. As Lincoln and Guba put it, whether
findings'hold insomeothercontext, oreven inthesame
contextatsomeothertime,isanempiricalissue'(Lincoln
andGuba1985: 316).Instead,qualitativeresearchersare
encouragedto produce what Geertz (l973a) callsthick
description-that is, rich accounts of the details of a
culture. Lincolnand Gubaargue thatathickdescription
provides others withwhattheyrefer toasadatabasefor
makingjudgementsaboutthe possible transferabilityof
findingsto othermilieux.
Dependability
Asaparalleltoreliabilityinquantitativeresearch,Lincoln
and Guba propose the idea ofdependabilityand argue
that, to establish the merit of research in terms of thiscri-
terion of trustworthiness, researchers should adoptan
'auditing'approach.Thisentails ensuringthat complete
records are kept ofallphases ofthe research process-
problemformulation, selection ofresearchparticipants,
fieldwork notes, interview transcripts, data analysis
decisions, and so on-inan accessible manner. Peers
would then actasauditors, possiblyduring thecourseof
theresearchandcertainlyattheend toestablishhowfar
proper procedures are being and have been followed.
Thiswouldincludeassessingthedegreetowhichtheoret
icalinferencescanbejustified.Auditinghasnotbecomea
popular approachtoenhancing thedependabilityofqual-
itativeresearch.Arareexampleisastudyofbehaviourat



iin
Its
Ice to
ntails
")NO
Ike a
rthe
ocial
19son
'hat
most
ncoln
argue
iscri-
ptan
iplete
ess-
rants,
"lysis
Peers
:seof
wfar
wed.
oret-
rnea
qual-
urat
anAmerican'swapmeet',where second-hand goodsare theprojectforlackofsufficientdata fordrawingitscon-
bought and sold (Belk et al. 1988). A team of three elusionsif theysawsuchavoid'(Belket al. 1988: 456),
researcherscollecteddata overfourdaysthrough obser- Thestudyhighlightssomeproblemsassociatedwiththe
vation, interviews, photography, and video-recording. auditing idea. One is that it isverydemanding for the
Theresearchersconductedseveraltrustworthinesstests, auditors, bearing in mind that qualitative research fre-
suchasrespondentvalidation andtriangulation. But,in quentlygeneratesextremelylargedatasets,anditmaybe
addition, they submitted their draft manuscript and that thisisamajorreasonwhyithasnotbecomeaperva-
entiredatasettothreepeers,whosetask'wastocriticize siveapproach tovalidation.
Key concept 16.4
What is triangulation?
Triangulation entailsusing morethanone methodor sourceof data inthe studyofsocialphenomena.
Thetermhas beenemployed somewhat morebroadly byDenzin (1970: 310)to referto an approach that
uses'multiple observers, theoretical perspectives, sourcesof data, andmethodologies', but theemphasis
hastended to be on methodsofinvestigationand sources of data. Oneofthereasons fortheadvocacy by
Webbet al. (1966) ofa greateruseofunobtrusive methods wastheir potential inrelation to a strategy of
triangulation(see Key concept13.3). Triangulation canoperatewithinand acrossresearchstrategies.Itwas
originallyconceptualizedbyWebbet al. (1966) as an approachto the development ofmeasures of concepts,
whereby morethan onemethod would be employed inthe development ofmeasures, resulting ingreater
confidenceinfindings. Assuch,triangulationwasverymuchassociatedwithaquantitativeresearchstrategy.
However, triangulation canalsotake placewithin a qualitative research strategy.In fact, ethnographers often
checkout theirobservationswithinterview questionsto determinewhethertheymighthavemisunderstood
whattheyhadseen.Bloor (1997) reportsthat hetackled the process of deathcertification inaScottish city
intwoways: interviewing clinicianswith a responsibility forcertifying causesofdeaths,andasking the same
peopleto completedummy death certificatesbasedon casesummarieshe had prepared.Increasingly,
triangulationisalsobeing usedto referto aprocess ofcross-checking findings deriving from bothquantitative
andqualitativeresearch(Deacon et al. 199B). Triangulation representsjustoneway inwhichit maybeuseful
to thinkabout theintegrationofthese two researchstrategiesand iscoveredinChapter25 inthecontextof
mixedmethods research.
Confirmability raiseawider setofissuesconcerningthewiderpolitical
impactofresearch.Thesearethecriteria:
Confirmabilityisconcernedwithensuringthat,whilere-
cognizingthatcomplete objectivityis impossiblein social
Fairness. Doesthe research fairlyrepresent different
research,the researcher canbe showntohave acted in
viewpointsamongmembersofthesocialsetting?
goodfaith;inotherwords,itshouldbeapparentthathe
Ontological authenticity. Doestheresearchhelpmem-
orshehasnotovertlyallowedpersonalvaluesortheoret-
berstoarriveatabetterunderstandingoftheirsocial
icalinclinations manifestly to sway the conduct of the
milieu?
research andfindings deriving fromit. LincolnandGuba
Educative authenticity. Doesthe research help mem-
proposethatestablishingconfirmabilityshouldbeoneof
bers to appreciate better the perspectives of other
theobjectivesofauditors.
membersoftheirsocialsetting?
Authenticity
Catalytic authenticity. Has the research acted as an
Inadditiontothesefourtrustworthiness criteria,Lincoln impetus to members to engage in action to change
andGubasuggestcriteria ofauthenticity. Thesecriteria theircircumstances?


t:
Tactical authenticity. Has the research empowered
! ' l
members to take the steps necessary for engaging in
action?
The authenticity criteria are thought-provoking but have
not been influential, and their emphasis on the wider im-
pact of research is controversial. They have certain points
of affinity with actionresearch (see Key concept 16.5),
which by and large has not been a popular form of social
research, though it has had some impact in fields like or-
ganization studies and education. The emphasis on prac-
tical outcomes differentiates it from most social research.
Recent discussions about quality
criteria for qualitative research
The main point of discussing Lincoln and Guba's ideas is
that they differ from writers like LeCompte and Goetz in
seeking criteria for evaluating qualitative research that
represent a departure from those employed by quantita-
tive researchers. The issue of research quality in relation
to qualitative investigations has become a rather con-
tested area in recent years, with several schemes of cri-
teria being proposed as possible alternatives to reliability
and validity as criteria and to schemes like Lincoln and
Guba's list. For example, Yardley (2000) has proposed
the following four criteria:
Sensitivitytocontext:sensitivity not just to the context
of the social setting in which the research is conducted
Thinking deeply 16.2
but also to potentially relevant theoretiCal positions
and ethical issues.
Commitmentandrigour:substantial engagement With
the subject matter, having the necessary skills, and
thorough data collection and analysis.
Transparencyandcoherence:research methods dearl
specified, clearly articulated argument, and a l x i v ~
stance (see Keyconcept 27.4 on reflexivity).
Impact and importance: importance of having an im.
pact on and significance for theory, the communityOn
which the research is conducted and for practitioners.
When compiling these criteria, Yardley had in mind
health researchers who are likely to emphasize the impact
of a study, which probably accounts for the presence of
the last of these four criteria-impact and importance-
which has some affinities with Lincoln and Guba's
authenticity criteria.
Perhaps in response to the proliferation of different lists
of qualitative research criteria and also because of the lack
of agreed criteria, Spencer et01. (2003) have producedan
extremely comprehensive list (see Thinking deeply 16.2).
This list of quality criteria draws on the schemes that
already existed at the time of their research and alsoon
consultations with researchers in various fields. These
consultations were in the form of semi-structured inter-
views and focus groups with practising researchers and
writers on social research methods. In fact, I was one of
the interviewees and also a focus group participant.
Using checklists for appraising quality in
qualitative research?
Spencer et01. (2003) were commissionedto produce a report forthe UK Government's Cabinet Office that
aimed to provide a frameworkfor assessingthe qualityof evaluation research studies that derived from
qualitative investigations. Although their report focused upon evaluation research (see Key concept 2.5). they
drew on considerations relatingmore generallyto qualitative research. so that their scheme has a relevance
beyond evaluation research.
The authors produced what is probably the most comprehensive listof criteriaaround. Here are the criteria
that they suggest should be used when appraisingthe qualityof a qualitative research study. Inthe case of each
criterion, the original wordinghas been used.
1. How credible are the findings?
2. Has knowledge/understanding been extended by the research?



I
Ipositio!Js
mentWith
,kills, and
Ids clearly
I reflexive
Iganim,
Illnityon
jtioners.
inmind
eimpact
senceof
rtance-
Guba's
'enrlists
thelack
ucedan
Y16.2).
esthat
alsoon
These
Iinter-
irs and
oneof
ley
e
sach
,
,
!
3. Howwell doestheevaluationaddressitsoriginal aimsandpurposes?
!
!
4. Scopefordrawingwiderinfluences-how ~ isthisexplained?
5. How clearisthebasisoftheevaluativeappraisal?
6. Howdefensibleistheresearchdesign?
f
!
7. Howwelldefendedisthesampledesign/targetselectionofcases/documents?
8. Samplecomposition/caseinclusion-howwell istheeventualcoveragedescribed?
9. Howwellwasthedatacollectioncarriedout?
10. How well hasthe approachto, andformulation of,theanalysis been conveyed?
11. Contexts of data sources-howwellare they retainedand portrayed?
12. Howwell has diversity of perspectiveand content been explored?
13. How well hasdetail, depth andcomplexity (richness?) ofthe data beenconveyed?
14. How clearare thelinksbetweendata, interpretation and conclusions-i.e.howwell cantherouteto
anyconclusionsbeseen?
15. Howclearandcoherent isthereporting?
16. How clearare the assumptions/theoreticalperspectives/values that haveshapedtheform and output
oftheevaluation?
17. Whatevidenceisthereofattentiontoethicalissues?
18. How adequatelyhasthe researchprocessbeen documented?
Each of these eighteencriteriacomeswith'qualityindicators' that are designedto helpintheappraisalofa
study. What isnot clearishowsuchaframework shouldbeused. It hasthe appearanceofachecklist, but as
Spenceret 01. (2003: 90)note,there isresistancewithin the qualitative research community to thepossibly
rigid application ofanylistofcriteria that achecklist would entail. The researchersfound that the ideaof
checklists ofquality criteria weregenerally regarded rathernegatively byinterviewees.In fact, Spenceret or
do not promotetheirframeworkas achecklist, notingvariousconcernsabout their use inqualitativeresearch,
suchat therisk ofchecklists becoming too prescriptive or of beingapplied too rigidly. However, the factthat
the authorsdo not treat theirwork as leading to achecklist does not meanthat theframeworkcannot or
should not be usedinthatway. Indeed, aroundthe sametimethatSpencer and hiscolleagues published their
report,Michael Quinn Patton,aleadingqualitativeevaluationresearcher,publishedonlinealistofcriteriathat
was designedtobeusedasachecklist-seehttp://www.wmich.edu/evalctr/checklists/qec.pdf
(accessedon 29March200n.
Whatdoyouthink? Canchecklistsbevaluableforappraisingthequalityofqualitativestudies?Ifyouransweris
no, why isthat?Is it something to dowith thenature ofqualitativeresearchthatmakeschecklists ofquality
inappropriate?Mightchecklistsbe morevaluableinappraising the qualityofquantitativeresearch?
Thefull report bySpenceret 01. canbefound athttp://www.policyhub.gov.uk/docs/qqeJep.pdf
(accessedon29March2007).
Between quantitative and qualitative
research criteria
Hammersley (1992a) lies midway between the prefer-
ence foradapting quantitative research criteria and the
preferenceforalternative qualitycriteriawhen assessing
thequalityofqualitativeinvestigations.Heproposesthat
validity is an important criterion but reformulates it
somewhat. For Hammersley, validity means that an
empirical account must be plausible and credible and
shouldtakeintoaccounttheamountandkindofevidence
usedinrelationtoanaccount.Inproposingthiscriterion,
Hammersley'spositionshareswithrealism(seeKeycon-
cept1.3)thenotion thatthereisanexternalsocialreality
I




that can be accessed by the researcher. However, he
simultaneously shares with the critics of the empirical
realist position the rejection of the notion that such access
is direct and in particular that the researcher can act as a
mirror on the social world, reflecting its image back to an
audience. Instead, the researcher is always engaged in
representations or constructions of that world. The plau-
sibility and credibility of a researcher's 'truth claims' then
become the main considerations in evaluating qualitative
research. Hammersley's subtle realist account, as he calls
it, entails recognizing thatwe can never be absolutely
Cer.
tain about the truth of any account, sincewe havenoc
pletely incontrovertible way of gaining direct
the reality on which it is based. Therefore, he argues 'w
0
must judge the validity of claims [about truth] th:
basis of the adequacy of the evidence offered insUPPOrt of
them' C1992a: 69). This means that an account canbe
held to be 'valid or true if it represents accuratelythose
features of the phenomena that it is intended to describe
explain or theorise' (1992a: 69). '
Key concept 16.5
'." r. ,.

What is action research?
There isno single type ofaction research, but broadly it can be definedas an approach inwhich the action
researcherand membersofa social settingcollaborate inthe diagnosis ofa problem and inthe development or
a solution basedon the diagnosis, Itcantakea variety offorms, from the actionresearcher beinghired bya client
to work on the diagnosis to and solution of a problem, to working witha groupofindividuals whoare identified
as needingto developa capacityfor independent action, The collection ofdata islikely to be involved inthe
formulation of the diagnosis ofa problem and inthe emergenceofa solution, In action research, the investigator
becomes part of the field ofstudy.Action researchcan involve the collection of bothquantitative and
qualitative data. Gibson (2004: 5) describes a Canadian projectthat was interestedinthe social and cultural
factors that havean impact onthe prevention and treatment oftuberculosis (TB) among'foreign-born and
aboriginal populations'. The idea forthe projectcame from a nurseina TB clinic whogarneredsupportfrom
the groupsmostaffectedbythe disease.Anadvisory committee, which drewitsmembership from the local
community ina province of Alberta, as well as from community, government, andacademic constituencies,
wasformed. Two representativesfrom each of the ten distinctsocio-cultural communities were recruited and
acted as researchassociates. Following training, they collected data through interviews and analysed some of
the resulting data. Interviews were conducted in relation to fourgroups: TB sufferers: peopleon prophylaxis:
people who refused prophylaxis; and 'those with a moredistant history ofTB intheircountry oforigin oron
aboriginal reserves' (Gibson 2004: 5).The researchassociates, membersofthe advisory committee, and
academicstaffanalysed the interview data. The findings revealed that, while the healthcare system deals well
will activeTB cases, it is lesseffective inrelation to prevention in relation to communities at risk. Italsorevealed
that health professionals often fail to identify TB becauseit isnot prevalentin Western nations. Theadvisory
groupthen produceda planto disseminate itsfindings and developedother initiatives including 'an information
video, a community education nurse position, and TB factsheet intheir various languages' (Gibson 2004: 5).
Action researchismore commoninsome social scienceareas than others. It ismorecommon infields such as
businessand management researchand social policy than others. It issometimes dismissed byacademics for
lacking rigour and for beingtoo partisaninapproach. However, it isadvocatedbysome researchers because of
itscommitmentto involving people inthe diagnosis ofand solutions to problems rather than imposing onthem
solutions to pre-defined problems.
Action research shouldnot be confusedwithevaluation research (Key concept 2.5), which usually denotesthe
studyofthe impact ofan intervention, suchas a new social policy or a newinnovation inorganizations. The
research referredto in Research in focus 16.6 wasconducted broadly with an evaluation research frame of
reference inthat it was concerned to evaluatethe impact of the introduction ofperformance appraisal in
British universities.
b



ItelYcer.
noCorn.
.ccess to
ues,'we
on the
Pponof
. can be
Iythose
escribe,
In
lent of
3 client
ntified
the
:igator
'al
d
:lm
al
s,
and
~ of
ds;
n
Nell
~ l e
"Y
tion
).
as
)r
e of
tern
he
Hammersley also suggests relevance as an important
criterion of qualitative research. Relevance is taken to be
assessed from the vantage point of the importance of a
topic within its substantive field or the contribution it
makes to the literature on that field. Hammersley also dis-
cusses the question of whether the concerns of practition-
ers(that is, people who are part of the social setting being
investigated and who are likely to have a vested interest
inthe research question and the implications of findings
deriving from it) might be an aspect of considerations of
relevance. In this way, his approach touches on the kinds
ofconsideration that are addressed by Guba and Lincoln's
authenticity criteria (Lincoln and Guba 1985; Guba and
Lincoln 1994). However, he recognizes that the kinds of
researchquestions and findings that might be of interest
to practitioners and researchers are likelyto be somewhat
different. As Hammersley notes, practitioners are likely to
beinterested in research that helps them to understand or
addressproblems with which they are confronted. These
maynot be (and perhaps are unlikely to be) at the fore-
front of a researcher's set of preoccupations. However,
there may be occasions when researchers can combine
the two and may even be able to use this capability as a
means of securing access to organizations in which they
wishto conduct research (see Chapter 17 for a further dis-
cussionof access issues).
Overview of the issue of criteria
There is a recognition-albeit to varying degrees-that a
simpleapplication of the quantitative researcher's criteria
of reliability and validity to qualitative research is not
desirable, but writers vary in the degree to which they
propose a complete overhaul of those criteria. Nor do the
three positions outlined above-adapting quantitative
research criteria, alternative criteria, and Hammersley's
subtle realism-represent the full range of possible
stances on this issue (Hammersley 1992a; Seale 1999).
To a large extent, the differences between the three posi-
tions reflect divergences in the degree to which a realist
position is broadly accepted or rejected. Writers on quali-
tative research who apply the ideas of reliability and
validity with little if any adaptation broadly position
themselves as realists-that is, as saying that social real-
ity can be captured by qualitative researchers through
their concepts and theories. Lincoln and Guba reject this
view, arguing instead that qualitative researchers' con-
cepts and theories are representations and that there
may, therefore, be other equally credible representations
of the same phenomena. Hammersley's position occupies
a middle ground in terms of the axis, with realism at one
end and anti-realism at the other, in that, while acknow-
ledging the existence of social phenomena that are part
of an external reality, he disavows any suggestion that it
is possible to reproduce that reality for the audiences of
social scientific endeavour. Most qualitative researchers
nowadays probably operate around the midpoint on
this realism axis, though without necessarily endorsing
Hammersley's views. Typically, they treat their accounts
as one of a number of possible representations rather
than as definitive versions of social reality. They also
bolster those accounts through some of the strategies
advocated by Lincoln and Guba, such as thick descrip-
tions, respondent validation exercises, and triangulation.
To a certain extent, traditional quantitative research
criteria have made something of a comeback since the
late 1990s. One issue is to do with the perception of qual-
itative research. For one thing, to reject notions like reli-
ability and validity could be taken by some constituencies
(such as funding bodies) as indicative of a lack of concern
with rigor, which is not a desirable impression to create.
Consequently, there has been some evidence of increased
concern with such issues. Armstrong et al. (1997) report
the result of an exercise in what they call 'inter-rater reli-
ability', which involved the analysis by six experienced
researchers of a focus group transcript. The transcript
related to research concerned with links between percep-
tions of disability and genetic screening. The focus group
was made up of sufferers of cystic fibrosis (CF) and the
participants were asked to discuss genetic screening. The
raters were asked to extract prominent themes from tran-
scripts, which is one of the main ways of analysing quali-
tative data (see Chapter 22). They tended to identify
similar themes but differed in how themes were 'pack-
aged'. One theme that was identified was 'visibility'. This
theme was identified as a theme in transcripts by all
researchers and refers to the invisibility of genetic disor-
ders. The CFsufferers felt disadvantaged relative to other
disabled groups because of the invisibility of their dis-
order and felt that the public were more sympathetic to
and more inclined to recognize visible disabilities. How-
ever, some analysts linked it to other issues: two linked
it with stigma; one to problems of managing invisibility.
In a sense the results are somewhat inconclusive but
are interesting for this discussion because they reveal an
interest among qualitative researchers in reliability. A
more recent and similar exercise is described in Research
in focus 16.2.



-------------
Research infocus 16.2

Reliabilityforqualitativeresearchers
Gladney et 01. (2003) report the findings of an exercise in which two multidisciplinary teams ofresearchers
were askedto analyse qualitative interviews witheighty Texas school students.The interviews wereconcerned
withreflections on violence on television; reasonsforviolence amongsomeyoungpeople;and reasons for
some youngpeople not beingviolent One groupof raters read interview transcripts ofthe interviews; the
other grouplistened to the audio-taped recordings. Thus,the dicewere slightly loadedinfavour ofdifferent
themes beingidentified bythe two groups. In spite ofthisthere was remarkable consistency between the
two groups inthe themes identified. Forexample, inresponseto the question 'Whyare someyoung people
violent", GroupOne identified the following themes: family/parental influence; peer influence; social
influence; mediainfluence; and coping. GroupTwo'sthemes were: the waythey were raised; media influence;
appearance; anger, revenge, protection; and environmental or peer influence. Such findings are quite
reassuring and are interesting because oftheir clear interest in reliability ina qualitative research context
Interestingly, exercises suchas this can be viewed as a form of what lincolnand Guba(1985) call ouditmq,
Studentexperience
Thinkingaboutreliability
Hannah Creane was concerned about the reliability of her categorizationof her qualitativedata and
enlisted others to check out her thinking.
There wasa slight concernwhen Iwasgrouping data together that my categorization wasofan arbitrary
nature, and so Icouldbe making assumptions and theorizing on the basis of highly subjective categories.
However, Itried to makesure that allthe categories Iusedwere relevant, and Icheckedthemoverwith
other people to makesure they made sense in relation to the researchand the questions Iwasdealing with.
To read more about Hannah's research experiences, go to the Online Resource Centre that accompanies
this book at http://www.oxfordtextbooks.co.uk/orc/brymansrm3ef
Themainpreoccupationsofqualitative

... .. . .- '..
.
/.
,
researchers
As was noted in Chapter 6, quantitative and qualitative
research can be viewed as exhibiting a set of distinctive
but contrasting preoccupations. These preoccupations
reflect epistemologically grounded beliefs about what
constitutes acceptable knowledge. In Chapter 1, it was
suggested that at the level of epistemology, whereas
quantitative research is profoundly influenced bya nat-
ural science approach to what should count as acceptable
knowledge, qualitative researchers are more influenced
by ;n"'1'retiv;,m (see Key concept 1A). This ""';';"'1
tI
b
[j
,.
t
p
n
[:
n
a
r
ir
p
['
c
c
g
t'
a
a
q
c
'I
c
a
r
c
3
c


ext.
ting.
d
rary
ries.
lith
19with.
anies
rhereas
r anat-
eptable
renced
osition
. Ifbeviewedasthe productofthe confluence of
use
eerelatedstances:Weber'snotion ofVerstehen;. sym-
lieinteractionism;and phenomenology.Inthis sec-
. five distinctive preoccupations among qualitative
~ . .
, archerswill beoutlinedandexamined,
res
e
:Seeing through the eyes of the people
being studied
\ An underlyingpremiss of many qualitative researchers
, isthatthesubjectmatterofthe socialsciences (that is,
" people and theirsocial world)does differ fromthesubject
"matter ofthe natural sciences.Akey difference isthat
theobjectsof analysis of the natural sciences (atoms,
molecules, gases,chemicals, metals, and soon) cannot
attribute meaning to events and to theirenvironment.
However, peopledo. Thisargumentisespeciallyevident
intheworkofSchutzand canparticularlybeseeninthe
passage quoted on page 16,where Schutz draws atten-
tionto the fact that, unlike the objects of the natural
sciences, theobjectsofthesocialsciences-people-are
capable of attributing meaning to their environment.
Consequently, many qualitative researchers have sug-
gestedthatamethodologyisrequiredforstudyingpeople
thatreflectsthese differences between people and the
objectsofthenaturalsciences.Asaresult,manyqualita-
tiveresearchersexpressacommitmenttoviewingevents
andthesocialworld through theeyesofthe people that
theystudy.The social worldmust be interpreted fromthe
perspective of the people being studied, ratherthan as
thoughthosesubjectswereincapableoftheirownreflec-
tions onthe socialworld. The epistemology underlying
qualitative research has been expressed bythe authors
ofonewidely read text asinvolvingtwocentraltenets:
"(l) ...face-to-face interaction is the fullest condition
ofparticipating in the mind of another human being,
and(2) ...youmust participateinthemind ofanother
humanbeing (insociological terms, "taketheroleofthe
other")to acquire social knowledge'(Loflandand Lofland
1995:16).
Itisnot surprising, therefore, that many researchers
makeclaimsintheirreportsoftheir investigationsabout
havingsoughttotaketheviewsofthepeopletheystudied
asthe point of departure. This tendency reveals itself
infrequent references to empathy and seeing through
others'eyes.Herearesomeexamples.
Fielding(1982) carried out research on members of
theNationalFront,aBritishextreme right-wingpolit-
icalparty. Inspite ofhisfeelingsofrevulsion forthe
racist doctrine,hesought to examine the party's position
:as a moral posture and itsmembers' interpretations
weretobeilluminatedbyanempatheticimmersionin
their world. Inthe processof"tellingitlikeitwas for
them", Icouldreproduceanaccount fromwhichout-
siderscouldunderstandtheideology'spersuasiveness
topeoplesoplaced' (Fielding1982:83).
Armstrong (1993) carried out ethnographicresearch
on football hooliganismthrough participantobserva-
tion with Sheffield United supporters. He describes
hiswork aslocatedin'Verstehende sociology-trying
to think oneself into the situations ofthe people one
is interested in...in this case the "Hooligan". This
approach involves recognizing social and historical
phenomenaasbeyondanysingleorsimpleidentifying
cause and attemptingto make sense from the social
actors'viewpoint'(Armstrong1993:5-6).
Like Armstrong, Taylor (1993), in relation to her
ethnographic study of female injecting drug users,
draws attention to the influence ofWeber's idea of
Verstehen onherresearch. Thesignificanceoftheidea
forher wasthatitmeantthat'inorderto understand
socialactions we must graspthe meaningthat actors
attach to their actions' (Taylor 1993: 7). She also
acknowledges the influence of symbolicinteraction-
ismonherposition.
For their research on teenaged girls' views on and
experiences of violence, Burman et al. (2001: 447)
'soughttogroundthestudyinyoungwomen'sexperi-
encesofviolence,hearingtheiraccountsand privileg-
ingtheirsubjectiveviews'.
This predilection for seeing through the eyes of thepeo-
plestudied inthe course ofqualitative research isoften
accompanied by the closelyrelated goal of seeking to
probe beneathsurface appearances. Afterall, bytaking
thepositionofthe peopleyouarestudying, the prospect
is raised that they might view things differently from
what an outsider with little direct contact might have
expected.Thisstancerevealsitselfin:
Foster's (1995) researchonahighcrimecommunity,
whichwasnotperceivedassuchbyitsinhabitants;
Skeggs's (1994: 74) study of young working-class
women,showingthattheywerenot'ideologicaldupes
ofbothsocialclassandfemininity';
Taylor's (1993: 8) studyofintravenous female drug
users, showing the people she studied are not
'pathetic,inadequateindividuals'but'rational,active
peoplemakingdecisionsbasedonthecontingenciesof





I,: I
I,
I I
Student experience
Importance of seeing through research
participants' eyes
Rebecca Barneswasattractedtoqualitative research forherresearch onviolenceinsame-sexrelationshi s
becausetherehadonlybeenaquantitativeres.earchinthisarea and becauseshewantedto understand P
the phenomenoninher research participants'ownwords.
Ichosea qualitative research designfor a number of reasons.First, Iwas aware that verylittlequalitative
research exists in myfieldof research, andat thetime that Istarted myresearch, Icould not findany
comprehensive qualitativestudiesofwoman-to-womanpartnerabuseintheUK.Thus, I wantedmyresearch
to contribute towardsfilling this gap, on a national andinternationallevel. Ialsochose a qualitativeresearch
design because I wanted to achieve an in-depth understanding of the experiences of woman-to-woman
partner abuse that women reported intheir own words andusingtheir own frames of reference, Ialso set
outtoachieve amore texturedanalysisof thedynamicsof abuseandthedifferent impactsthatbeingabused
hasuponwomen,andhowthese maychangeovertime.
To readmoreabout Rebecca's research experiences, go to the Online Resource Centre that accompanies
this book at http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/.
boththeirdrugusingcareersand theirrolesand status possibilitythatthe researcherwillbeable toseethrough
insociety'; the eyes of only some of the peoplewhoform partof a
social scene butnot others, suchas only people ofthe
Armstrong's (1993:11) questinhis researchon foot-
samegender.These andotherpracticaldifficultieswillbe
ball hooliganism to 'see beyond mere appearances'
addressedinthe chaptersthatfollow.
and his finding that, contrary to the popular view,
hooligans are not a highly organizedgroupled by a
clearlyidentifiablegroupofringleaders;
Description and the emphasis
O'Reilly's (2000) ethnography of British expatriates on context
on the Costa del SolinSpain,inwhichsheshowshow
Qualitative researchers are much more inclined than
the widely held view that this group is deeply dis-
quantitative researchers to provide a great deal of
satisfiedwiththeirlivesinthe sunand longtoreturnis
descriptive detail when reporting the fruits of their
by no means an accurate portrayal in terms of how
research. This is not to say that they are exclusively
theyviewthemselvesandtheirsituation.
concerned with description. They are concerned with
The empatheticstance of seekingto see through the explanation, andindeedthe extentto which qualitative
eyes of one's research participants isverymuchin tune researchers ask 'why?' questions is frequently under-
withinterpretivismanddemonstrateswell the epistemo- stated.Forexample,Skeggs (1997: 22) has written that
logical links withphenomenology,symbolicinteraction- herfirstquestionforherresearchonyoungworking-c!ass
ism, andVerstehen. However, it is not withoutpractical womenwas'whydowomen,who are clearlynotjustvic-
problems.Forexample:therisk of'goingnative'and los- tims ofsome ideologicalconspiracy, consenttoasystem
ing sight ofwhat you are studying (see Keyconcept 17.3); of classand gender oppression which appears to offerfew
the problem of howfar the researchershould go, such rewardsandlittlebenefit?'(seeResearchinfocus17.8for
as the potential problem of participating in illegal or furtherdetailsofthisstudy).
dangerous activities, which could be a risk in research Many qualitative studies provide a detailed account
like thatengagedin by TaylorandArmstrong; and the of whatgoes on in the settingbeing investigated.Very




I
;tative
1Y
ryresear'
,research'
Iman !
1150 set
ngabuseJ:.
oonies
through
iarrof a
ofthe
; willbe
i than
eal of
. their
rsively
with
tative
nder-
1that
-class
:t vic-
'stem
rfew
Bfor
Junt
lery
often qualitative studies seem to be.fullof apparently
, 'vial details. However, these details are frequently
. ortan
t
forthequalitativeresearcher,becauseoftheir
, 'fican
ce
fortheirsubjectsandalsobecausethedetails
S1:
de
anaccountofthecontextwithinwhichpeople's
takesplace.Itwaswiththispointinmindthat
GeertZ (l9.73a) recommended the provision of thick
descriptionsofsocial settings,events,andoftenindividu-
als.As a result of thisemphasisondescription, qualitative
studies areoftenfullofdetailed information about the
socialworldsbeingexamined. On the surface,someof
thisdetailmayappearirrelevant, and,indeed,there isa
riskoftheresearcherbecomingtoo embroiledin descrip-
tive detail. Loflandand Lofland(1995: 164-5), for exam-
ple, warnagainst the sin ofwhat they call'descriptive
excess' in qualitativeresearch, whereby the amount of
detail overwhelmsor inhibits theanalysisof data.
Oneofthe mainreasons why qualitative researchers
arekeen to provide considerable descriptive detail is
thattheytypicallyemphasizetheimportanceofthecon-
textual understanding of socialbehaviour. Thismeans
thatbehaviour, values,orwhatevermustbeunderstood
incontext. Thisrecommendationmeansthat wecannot
understand thebehaviour ofmembersofasocialgroup
other than intermsofthe specificenvironmentinwhich
t1}.ey operate. Inthisway,behaviourthatmayappearodd
orirrationalcanmakeperfectsensewhenweunderstand
theparticularcontextwithinwhichthat behaviourtakes
place. The emphasis on context in qualitative research
goesbacktomanyoftheclassicstudiesinsocialanthro-
pology,whichoftendemonstratedhowaparticularprac-
tice,suchasthe magicalritualthatmayaccompanythe
sowingofseeds,madelittlesenseunlessweunderstand
thebeliefsystemsofthatsociety.Oneofthechiefreasons
forthe emphasis on descriptivedetail isthat it isoften
preciselythisdetailthatprovidesthemappingofcontext
intermsofwhichbehaviourisunderstood. Thepropen-
sityfordescriptioncanalsobeinterpretedasamanifesta-
tion of the naturalism that pervades much qualitative
research (see Keyconcept 2.4 and 16.1), because it
placesapremiumondetailed, richdescriptionsofsocial
settings.
Conducting qualitative research in more than one
setting can be helpful in identifying the significance
ofcontext andthewaysinwhichitinfluencesbehaviour
andways of thinking.Research in focus16.3 provides an
illustration of a multiple-casestudy that demonstrates
thispotential.
Research in focus 16.3
Contextual understanding in an ethnographic
study of three schools
Swain(2004)conductedan ethnographic studyofthreejunior schoolsintheUK inthelate1990s. Ethnography
isdiscussedinthenextchapter. Becauseit comparedfindings from threeschools,thiswasamultiple-case
study, which drewonthe strengths ofusing acomparative design inthat it waspossible to explorethe
significance of contextacross the threeschools. Theschoolsweredifferentintermsofthesocial characteristics
ofthepupilstheyrecruited:Highwoods Independent's(theschool namesarepseudonyms)pupilsweremainly
uppermiddleclass;pupilsatPetersfieldJuniorwerepredominantlymiddleclass;andWestmoorAbbey
Junior'spupilsweremainlyworking class.Swain(2004: 169)describeshisdatacollectionmethodsasinvolving
non-participant observation ofpupils inlessons and aroundtheschool and'loosely structured interviews' with
pupils basedon'nominatedfriendship groups'. Inthisarticle,Swain was interestedintheways inwhich boys
constructwhat itmeansto bemasculine intheschool and drawsprimarily on datacollectedon boysrather
thanongirls. Swain shows that masculinity was inseparable from the achievement ofstatusamongschool peer
groups andthat thebodywas the meansofexpressingmasculinity. Thesignificance ofcontextemerges;n
connectionw*hSwain'saccountofhowthebodywasusedtoconveymasculinityinthethreeschools:at
Highwoods sport was themediumthroughwhichthebodyexpressedmasculinity;at Westmoor Abbey the
emphasis wasmachoandfrequentlytookon aviolenttone; and atPetersfieldit was speedandstrength
(predominantly intheplayground ratherthanon thesports field). Context reveals itself inthedifferent
resources inthe threeschools that studentsmustdrawupontoperform masculinity



Emphasis on process
Qualitativeresearchtendsto viewsociallife in termsof
processes.Thistendencyrevealsitself in anumberofdif-
ferentways.Oneofthemainwaysisthatthereisoftena
concern to show how events and patterns unfold over
time.Asaresult,qualitativeevidenceoftenconveysastrong
senseofchangeandflux. As Pettigrew(1997:338)use-
fullyputsit, processis'a sequenceof individualandcol-
l ~ t v events,actions,andactivitiesunfoldingovertime
incontext'.Qualitativeresearchthat isbasedinethnographic
methodsis particularlyassociatedwiththisemphasison
process(although,ironically,Britishsocialanthropology,
whichis oftenassociatedwiththeearlydevelopmentof
ethnographic research, is sometimes thoughtofas pre-
senting a static picture of social reality by virtue of its
associationwithfunctionalism).It istheelementofpar-
ticipantobservationthatisa key featureofethnography
thatisespeciallyinstrumentalin generatingthisfeature.
Ethnographersaretypically immersedinasocialsetting
for a longtime-frequendyyears.Consequently,theyare
able to observe the ways in whichevents develop over
timeorthewaysinwhichthedifferentelementsofasocial
system (values,beliefs, behaviour, andso on) intercon-
nect.Suchfindingscaninjectasenseofprocessbyseeing
sociallife in terms ofstreamsofinterdependentevents
andelements(seeResearchinfocus 16.4foranexample).
This is not to say, however, that ethnograph
ersar
t ~ only qualitative researchers who inject a senSe e
processintoourunderstandingof sociallife. Itc of
analso
be achieved throughsemi-structuredand unstructu
. .. b ki . . red
tntervtewmg, yas ngparticipantstoreflectOnth
di c 11' fr epro.
cesses1ea inguptoor10 owingon omanevent"I<
.lVLC ee
and Bell (1985: 388;seealso Thinkingdeeply2.4), for
example, show, through the use ofa 'largely unst
. al J I ruc
tured, conversation interview stye' with fony-five
couplesin which themanwas unemployed,the accom.
modationsthataremadeovertimebybothhusbandsand
wives to the fact of male unemployment. The various
accommodations are notanimmediateeffect of unem.
ployment but are gradual and incremental responses
over time. The life history approach is an exampleof
a form ofqualitative research. One of the best-known
studies of this kind is Lewis's (1961) studyof a POor
Mexicanfamily. Lewis carriedoutextendedtaped inter.
views with thefamilymembers to reconstructtheirlife
histories.For his studyofdisastersintheUK, and inpar.
ticularofthe fire at aholidayleisurecomplexontheIsle
of Man,Turner (1994) employed publisheddocuments
to arrive ata reconstructionof theeventsleadingupto
thefire anda theoreticalunderstandingofthose events.
Thus,theemphasisonprocessinqualitativeresearchcan
be seenin theuseof quitedifferentapproachestodata
collection.
Research in focus 16.4
Process in (strike) action
Waddington(1994)describes hisexperiencesassociated withhisparticipantobservation ofastrike atthe
Ansellsbrewery inBirminghaminthe 1980s. As aparticipant observer, hewasinvolvedin'attendingpicket
lines,massmeetings and planningdiscussions.and accompanyingthe strikersonflyingpicketingand
intelligencegathering maneeuvres' (1994:113). Inaddition toobservation, hecarriedoutinformal interviews
andlinkedthesedata toothersources,suchas'material derivingfromnewspaperarchives,companyandtrade
union documents,lettersand richlydetailed minutes oftradeunion-managementmeetings' (1994: 115).Asa
result, hewasabletoshow'howthe contemporarybeliefs,valuesandattitudesof the workforce, and the
mutual feelingsofanimosityanddistrust betweenemployeesand management,wereshapedbyasequenceof
historical events stretching back over 20 years' (1994: 115). We can see inthis example the development ofa
senseofprocess inthreeways: through observationofthe strikeoveritsentirety,sothatdevelopmentsand
interconnectionsbetweenevents couldbebrought out;throughconnectingtheseeventswithhistoricaland
otherdata, sothatthe linksbetweenthe strikeand previousandotherevents andactionscouldbeoutlined:
and throughthesketchingof the context (intheformofthe past,aswell ascurrent beliefs and values)andits
linkswithbehaviour duringthe strike.
Man)
10re
form
with
peoj:
data
prod
ofer
wna
soci;
res
e
gen
l
ird.
rese
tion
imj:
xee
tne
the
off
the
res
for
as
mL
qu
ap
ad
ga
ad
fo
er
su
m
re
rc
se
n
ir
q




a sense
It can aJs6

>n thePro.:$
McRee
'12.4), for]
1 unstruc.).
fony-five '
ie accolIl_
>andsand
e various
of unelIl_
:esPOnses
ample of
5t-kno
wn
If a Poor
ledinter-
their life
td in par-
1 the ISle
cuments
ng up to
events.
archean
: to data
Ie
ket
iews
:!trade
Asa
'nee of
ofa
nd
and
ed;
d its
. lexibility andlimited structure
Jt{al1Y qualitative researchers are disdainful of approaches
researchthat entail the imposition of predetermined
::rmats on the social world. This position is largely to do
with the preference for seeing through the eyes of the
'people being studied. if a method of
: data collectionis employed, since this is bound to be the
! roduct of an investigator's ruminations about the object
. certain decisions must have been made about
; what he or she expects to find and about the nature of the
social reality that would be encountered. Therefore, the
researcher is limited in the degree to which he or she can
genuinely adopt the world view of the people being stud-
ied. Consequently, most qualitative researchers prefer a
research orientation that entails as little prior contamina-
tion of the social world as possible. To do otherwise risks
imposing an inappropriate frame of reference on people.
Keeping structure to a minimum is supposed to enhance
the opportunity of genuinely revealing the perspectives of
thepeopleyou are studying. Also, in the process, aspects
ofpeople'ssocial world that are particularly important to
them, but that might not even have crossed the mind of a
researcher unacquainted with it, are more likely to be
forthcoming. As a result, qualitative research tends to be
a strategy that tries not to delimit areas of enquiry too
much and to ask fairlygeneral rather than specificresearch
questions (see Figure 16.1 and Thinking deeply 16.1).
Because of the preference for an unstructured
approach to the collection of data, qualitative researchers
adoptmethods of research that do not require the investi-
gator to develop highly specific research questions in
advanceand therefore to devise instruments specifically
forthose questions to be answered. Ethnography, with its
emphasis on participant observation, is particularly well
suited to this orientation. It allows researchers to sub-
mergethemselves in a social setting with a fairly general
researchfocus in mind and gradually to formulate a nar-
rower emphasis by making as many observations of that
settingas possible. They can then formulate more specific
research questions out of their collected data. Similarly,
interviewing is an extremely prominent method in the
qualitative researcher's armoury, but it is not of the kind
we encountered in the course of most of Chapter 8--
namely, the structured interview. Instead, qualitative
researchers prefer less structured approaches to inter-
viewing, as we will see in Chapter 18. Blumer's (1954)
argument for sensititizing rather than definitive concepts
(that is, the kind employed by quantitative researchers) is
symptomatic of the preference for a more open-ended,
and hence less structured, approach.
An advantage of the unstructured nature of most quali-
tative enquiry (that is, in addition to the prospect of gain-
ing access to people's world views) is that it offers the
prospect of flexibility. The researcher can change direc-
tion in the course of his or her investigation much more
easily than in quantitative research, which tends to have
a built-in momentum once the data collection is under
way: if you send out hundreds of postal questionnaires
and realize after you have started to get some back that
there is an issue that you would have liked to investigate,
you are not going to find it easy to retrieve the situation.
Structured interviewing and structured observation can
involve some flexibility, but the requirement to make
interviews as comparable as possible for survey investiga-
tions limits the extent to which this can happen. O'Reilly
(2000) has written that her research on the British on the
Costa del Sol shifted in two ways over the duration of her
participant observation: from an emphasis on the elderly
to expatriates of all ages; and from an emphasis on per-
manent residents to less permanent forms of migration,
such as tourism. These changes in emphasis occurred
because of the limitations of just focusing on the elderly
and on permanent migrants, since these groups were not
necessarily as distinctive as might have been supposed.
Similarly, Kathleen Gerson has explained that, in her
research on changing forms of the family, she conducted
an early interview with a young man who had been
brought up in his early years in a traditional house-
hold that underwent a considerable change during his
childhood. This led her to change her focus from an
emphasis on family structures to processes of change in
the family (Gerson and Horowitz 2002). See Research
in focus 16.5 for a further illustration of the ways in
which the unstructured data collection style of qualitative
research can be used to suggest alternative avenues of
enquiry or ways of thinking about the phenomenon being
investigated.




I
, I,
Research in focus 16.5
Flexibility in action
In the course of a studyof youngpeoplewith learningdifficultiesusing Qualitative interviews, Davies (1999)
reportsthat shefound that on manyoccasions her interviewees mentionedfood in the course of conversation
Initially, shefollowed these conversations up largelyin order to establish rapport with these young people. s
However,shegradually came to realizethat in fact food wasof considerable significance for her research,
because it represented alens throughwhichher participants viewedtheir anxieties about the ways people
attempted to control them. Food wasalsoa focus for their strategies of resistance to control.
Research in focus 16.6
Emerging concepts
In the late 1980s and early 1990s, most UK universities were in the throesof introducingstaffappraisal schemes
for both academic andacademic-related staff. Staffappraisal isemployed to reviewthe appraisee's
performance and activitiesovera periodof usually oneor two years. Alongwith some colleagues, I undertook
an evaluation of staffappraisal schemes in four universities (Bryman et 01. 1994). The research entailed the
collectionof both Quantitative and qualitativedatawithin the framework of a comparative research design.
The qualitativedatawerederivedfrom large numbers of interviewswith appraisers, appralsees. senior
managers, andmanyothers. Inthe course of conductingthe interviewsand analysing the subsequent data
we became increasingly awareof a cynicism amongmanyof the people we interviewed. Thisattitude revealed
itself in several ways, such as: aviewthat appraisal hadbeenintroducedjust to pacify the government: abeliel
that nothing happenedof anysignificance in the aftermathof an appraisal meeting; theviewthat it was not
benefitinguniversities; anda suggestion that manyparticipants to the appraisal process werejust going
through the motions. Asoneof the interviewees said in relationto this lastfeature: 'It's likegoingthrough
the motions of it (appraisal]. It'sjust get it over with and signed anddatedandfiled andthat'stheendof it'
(quoted in Bryman et 01.1994: 180).
On the basis of thesefindings it was suggested that theattitudes towards appraisal andthe behaviour of those
involvedin appraisal werecharacterized by procedural compliance, whichwasdefinedas 'a response to an
organizational innovationin whichthe technical requirements of the innovation... arebroadlyadhered to,but
wherethere aresubstantial reservations about its efficacy and onlypartial commitmentto it, sothat there is a
tendencyfor the procedures associated with the innovationto beadhered to with less thana total
commitment to its aims' (Bryman et oi. 1994: 178).
Concepts and theory grounded in data
This issue has already been addressed in much of the
exposition of qualitative research above. For qualitative
-
researchers,conceptsand theories areusually inductively
arrived at from the data that are collected (seeResearch
in focus 16.1 and 16.6).
lr
q
p
S



Thecritiqueofqualitativeresearch
schemes
Iertook
the
iign.
ata
:vealed
3 belief
not
h
it'
those
In
to, but
: isa
ively
.arch
Inasimilarway to the criticisms that have been levelled at
uantitativeresearch mainly by qualitative researchers, a
~ r l l e l critique has been built up of qualitative research.
Some of the more common ones follow.
Qualitativeresearchistoosubjective
Quantitative researchers sometimes criticize qualitative
research as being too impressionistic and subjective. By
these criticisms they usually mean that qualitative find-
ings rely too much on the researcher's often unsystematic
views about what is significant and important, and also
uponthe close personal relationships that the researcher
frequentlystrikes up with the people studied. Precisely
because qualitative research often begins in a relatively
open-ended way and entails a gradual narrowing-down
of research questions or problems, the consumer of the
writingsderiving from the research is given few clues as
towhy one area was the chosen area upon which atten-
tionwas focused rather than another. Bycontrast, quanti-
tative researchers point to the tendency for the problem
formulation stage in their work to be more explicitly
stated in terms of such matters as the existing literature
on that topic and key theoretical ideas.
Difficult to replicate
Quantitative researchers also often argue that these
tendencies are even more of a problem because of the
difficulty of replicating a qualitative study, although
replication in the social sciences is byno means a straight-
forward matter regardless of this particular issue (see
Chapter 6). Precisely because it is unstructured and often
reliant upon the qualitative researcher's ingenuity, it is
almost impossible to conduct a true replication, since
there are hardly any standard procedures to be followed.
In qualitative research, the investigator him- or herself is
the main instrument of data collection, so that what is
observed and heard and also what the researcher decides
to concentrate upon is very much a product of his or her
predilections. There are several possible components of
this criticism: what qualitative researchers (especially
perhaps in ethnography) choose to focus upon while in
the field is a product of what strikes them as significant,
whereas other researchers are likely to empathize with
other issues; the responses of participants (people being
observed or interviewed) to qualitative researchers is
likely to be affected by the characteristics of the re-
searcher (personality, age, gender, and so on); and,
because of the unstructured nature of qualitative data,
interpretation will be profoundly influenced by the sub-
jective leanings of a researcher. Because of such factors, it
is difficult-not to say impossible-to replicate qualita-
tive findings. The difficulties ethnographers experience
when they revisit grounds previously trodden by another
researcher (often referred to as a 'restudy') do not inspire
confidence in the replicability of qualitative research
(Bryman 1994).
Problemsofgeneralization
It is often suggested that the scope of the findings of
qualitative investigations is restricted. When participant
observation is used or when unstructured interviews are
conducted with a small number of individuals in a certain
organization or locality, they argue that it is impossible to
know how the findings can be generalized to other set-
tings. How can just one or two cases be representative of
all cases? In other words, can we really treat Holdaway's
(1982) research on the police in Sheffield as representa-
tive of all police forces, or Armstrong's (1998) research
on Sheffield United supporters as representative of all
football supporters, or Waddington's (1994) study of a
strike as generalizable to all lengthy strikes? In the case of
research based on interviews rather than participation,
can we treat interviewees who have not been selected
through a probability procedure or even quota sampling
as representative? Are Taylor's (1993) female intra-
venous drug users typical of all members of that category
or are Skeggs's (1994; see Research in focus 17.8) young
working-class women typical?
The answer in all these cases is, of course, emphatically
'no'. A case study is not a sample of one drawn from
a known population. Similarly, the people who are
interviewed in qualitative research are not meant to be
representative of a population, and indeed, in some
cases, like female intravenous drug users, we may find it
more or less impossible to enumerate the population in
any precise manner. Instead, the findings of qualita-
tive research are to generalize to theory rather than to
~ ~ ~ ~ ~ .



populations. Itis 'the cogency of the theoretical reason-
ing' (Mitchell 1983: 207), rather than statistical criteria,
that is decisive in considering the generalizability of the
findings of qualitative research. Inother words, itis the
quality of the theoretical inferences that are made out
of qualitative data that is crucial to the assessment of
generalization.
However, not all writers on the issue of generalization
in relation to qualitative research (and case study research
inparticular) accept this view. Williams (2000: 215) has
argued that, in many cases, qualitative researchers are in
a position to produce what he calls moderatum general-
izations-that is, ones in which aspects of the focus of
enquiry (a group of drug users, a group of football hooli-
gans, a strike) 'can be seen to be instances of a broader set
of recognizable features'. In addition, Williams argues
that not only is it the case that qualitative researchers
can make such generalizations but that in fact they often
do make them, Thus, when generating findings relating
to the hooligans who follow a certain football club, a
researcher will often draw comparisons with findings by
other researchers relating to comparable groups. Indeed,
the researcher may also draw comparisons and linkages
with still other groups: followers of other professional
sports teams or violent groups that are not linked to sport.
When forging such comparisons and linkages, the
researcher is engaging in moderatum generalization.
Moderatum generalizations will always be limited and
somewhat more tentative than those associated with
statistical generalizations of the kind associated with
probability sampling (see Chapter 7). On the other hand,
they do permit a modicum of generalization and help to
counter the view that generalization beyond the immedi-

':
'.
This was a heading that was employed in Chapter 6 in
relation to quantitative research, but it is perhaps less
easy to answer in relation to qualitative research. To a
large extent, this is because qualitative research is less
codified than quantitative research-that is, it is less
influenced by strict guidelines and directions about how
to go about data collection and analysis. As a result, and
this may be noticed by readers of the chapters that follow
this one, accounts of qualitative research are frequently
less prescriptive in tone than those encountered in rela-
tion to quantitative research. Instead, they often exhibit
Is italwayslikethis?
ate evidence and the case is impossible in quall't '
atlve
research.
These three criticisms reflect many of the preoccu
, f .. h th pa,
nons 0 quantitanve researc at were diScussed i
Chapter 6. A further criticism that is often made ofq l ~
tative research, but that is perhaps less influenced b
quantitative research criteria, is the suggestion thatquai.
itative research frequently lacks transparency in how the
research was conducted.
Lackoftransparency
It is sometimes difficult to establish from qualitative
research what the researcher actually did and how he
or she arrived at the study's conclusions. For example,
qualitative research reports are sometimes unclearabout
such matters as how people were chosen for observation
or interview. This deficiency contrasts sharply with the
sometimes laborious accounts of sampling procedures in
reports of quantitative research. However, it does not
seemplausible to suggest that outlining insomedetailthe
ways in which research participants are selected consti-
tutes the application of quantitative research criteria.
Readers have a right to know how far research particip-
ants were selected to correspond to a wide range of
people. Also, the process of qualitative data analysis is
frequently unclear (Bryman and Burgess 1994<1). It is
often not obvious how the analysis was conducted-in
other words, what the researcher was actually doing
when the data were analysed and therefore how the
study's conclusions were arrived at. To a large extent,
these areas of a lack of transparency are increasingly
being addressed by qualitative researchers.
more of a descriptive tenor, outlining the different ways
qualitative researchers have gone about research or
suggesting alternative ways of conducting research or
analysis based on the writer's own experiences or those
of others. To a large extent, this picture is changing, in
that there is a growing number of books that seek to
make clear-cut recommendations about how qualitative
research should be carried out.
However, if we look at some of the preoccupationsof
qualitative research that were described above, we can
see certain ways in which there are departures from the
prac
the
tim
t
that
tion
arri
no I
eml
exa
wa:
me
exa
ies
ing
tea
(H
I
SE
tit
at
1(
at



--------
ther
1 research
prlJ
J
-, ,- - ..4lld
formulating
research questions
Chapter outline
Introduction 66
Getting to know what is expected of you by your institution 66
Thinking about your research area 67
Using your supervisor 67
Managing time and resources 66
Formulating suitable research questions 69
Writing your research proposal 75
Preparing for your research 76
Doing your research and analysing your results 76
Checklist 78
Key points 78
Questions for review 79


t t
_ ' I
__________________________ d9X,1 'A ZS3QH I
6
9d%J
_
1
I
Chapter guide
--------------------------------------...
I
Thegoalofthis chapteristo provideadviceto studentsonsomeoftheissuesthattheyneedto consider
rn.
iftheyhaveto prepareadissertationbasedupon arelativelysmall-scale project,Increasingly, social
abl
sciencestudentsarerequiredto producesuchadissertationaspartofthe requirementsfor their
are
degrees.Inadditionto helpwiththeconductofresearch, whichwillbetheaim ofthe chapters that
ing
come laterinthis book, more specific advice ontacticsincarryingoutand writingupsocialresearchfor
adissertationcanbeuseful. It isagainstthis backgroundthatthis chapterhasbeenwritten.Thechapter
exploresawidevarietyofissues, suchas:
advice ontiming;
I
advice ongeneratingresearchquestions;

advice onconductingaproject;
advice onwritingaresearch proposal.
Introduction
This chapter has been written to provide some advice for
readers who might be carrying out a research project of
your own. The chapters that follow in Parts Two, Three,
and Four of this book will then provide more detailed
information about the choices available to you and how
to implement them. But beyond this, how might you go
about conducting a small project of your own? I have
in mind here the kind of situation that is increasingly
common among degree programmes in the social sciences
Mo
cor
val'
otb
mu
cat
fret
tbe
full
-the requirement to write a dissertation often of around
yOl
8,000 to 15,000 words. In particular, I have in mind the
eel'
needs of undergraduate students, but it may be that stu-
pre
dents on postgraduate degree programmes will also find
fee
some of the observations I make helpful. Also, the advice
ins
is (
is really concerned with students conducting projects
sd
with a component of empirical research in which they col-
to
lect new data or perhaps conduct a secondary analysis of
existing data.
f
I,

Getting to know what is expected of you

.
j;
,,'
.
by your institution
Your institution or department will have specificrequire-
ments concerning a wide variety of different features that
your dissertation should comprise and a range of other
matters relating to it. These include such things as: the
form of binding; how it is to be presented; whether an
abstract is required; how big the page margins should be;
the format for referencing; number of words; perhaps the
structure of the dissertation; how much advice you can
get from your supervisor; whether or not a proposal is
required; plagiarism; deadlines; how much (if any) finan-
cial assistance you can expect; and so on.
The advice here is simple: follow the requirements,
instructions, and information you are given. If anything
in this book conflicts with your institution's guidelines
and requirements, ignore this book! I very much hope
this is not something that will occur very much, but
if it does, keep to the guidelines your institution gives
you.



toconsider
',social
heir <\(
:';;"
rs thatt
!searchfor "
.'1.'
'hechapter

of around
I mind the
e that stu-
1alsofind
1Jeadvice
projects
1 theycol-
nalysis of
J
:y)finan-
rements,
mything
iidelines
ch hope
ich, but
)fi gives

, project and formulating research questions 67
Thinkingaboutyourresearcharea
The chances are that you will be asked to start thinking various modules, begin to think about whether there are
about what you want to do research on well before you any topics that might interest you and that might provide
are due to start work on your dissertation, It is worth giv- you with a researchable area.
ing yourself a good deal of time. As you are doing your
..

:
Usingyoursupervisor
"
Most institutions that require a dissertation or similar or she provides, since the criticisms will invariably be
component allocate students to supervisors. Institutions accompanied by reasons for the criticisms and sugges-
varyquite a lot in what can be expected of supervisors; in tions for revision. It is not a personal attack. Supervisors
other words, they vary in terms of what kinds of and how regularly have to go through the same process themselves
much assistance supervisors will give to students allo- when they submit an article to a peer-refereed journal or
cated to them. Equally, students vary a great deal in how apply for a research grant or give a conference paper.
frequently they see their supervisors and in their use of So respond to criticisms and suggestions positivelyand be
them. Myadvice here is simple: use your supervisor to the glad that you are being given the opportunity to address
fullest extent that you are allowed and follow the pointers deficiencies in your work before it is formally examined.
you are given by him or her. Your supervisor will almost Afurther point is that students who get stuck at the start
certainly be someone who is well versed in the research of their dissertations or who get behind with their work
process and who will be able to provide you with help and sometimes respond to the situation by avoiding their super-
feedback at all stages of your research, subject to your visors. They then get caught up in a vicious circle that
institution's strictures in this regard. If your supervisor results in their work being neglected and perhaps rushed
is critical of your research questions, your interview at the end. Try to avoid this situation byconfronting the fact
schedule, drafts of your dissertation, or whatever, try that you are experiencingdifficulties in getting going or are
to respond positively. Follow the suggestions that he getting behind and seek out your supervisor for advice.
Studentexperience
Usingsupervisors
Several students wrote about the role that their supervisors played in their research projects. Isabella Robbins
mentions that her supervisor played an important role in relation to her analysis of her qualitative data.
The emerging themes were strong and inthat sense the analysiswas not problematic, but I guess the
problems came in mapping the analysis onto the theory. Myway of dealing with this was to talk about the
analysis at supervisions and to incorporate the ideas that came of these discussions.
Cornelius Grebe provided the following advice about relationships with supervisors:
I have learned to be very clear about my expectations of my supervisors: what kind of professional and
personal relationship I thrive in and what form of support exactly I need from them.
To readmoreabout Isabella's and Cornelius's research experiences, go to the Online Resource Centre that
accompanies thisbookat http://www.oxfordtextbooks.co.uk/orc/brymonsrm3e/.



011
I
f

Managingtimeand'resources
we
rio
up
all
Allresearch isconstrained bytime and resources. There is 2. Find out what, if any, resources can be put at Your
no point in working on research questions and plans that disposal for carrying out your research. For example, Will
cannot be seen through because of time pressure or be- you receive help from your institution with such things
cause of the costs involved. Two points are relevant here. astravel costs, photocopying, secretarial assistance, posi,
f
age, stationery, and so on? Will the institution be able to
1. Work out a timetable-preferably in conjunction with
loan you hardware such as recording equipment and
your supervisor-detailing the different stages of your
transcription machines if you need to record and tran.
research (including the review of the literature and writ-
scribe your interviews? Has it got the software you need,
ing up). The timetable should specifythe different stages
such as SPSS or a qualitative data analysis package like
and the calendar points at which you should start and
NVivo? This kind of information will help you to establish
finish them. Some stages are likely to be ongoing-
how far your research design and methods are finan,
for example, searching the literature for new references
cially feasible and practical. The imaginary 'gym study'
(see below)-but that should not prove an obstacle to
used in Chapter 15 is an example of an investigation that
developing a timetable.
Studentexperience
Managingtime
One of the most difficult aspects of doinga research project for manystudents ismanagingtheir time.
SarahHansonwasexpliciton this point:
Neverunderestimate howlong it will take youto completea largeprojectlike a dissertation. Choosea
topicyouhave passion about. The moreyou enjoy your researchthe moreinteresting it will be to read.
Beorganized: post-it notes, folders, wall planners, anythingthat keepsyouontrackfrom dayto day will
helpyounot to be distracted from the purposeof yourstudy.
BothHannah Creane and Lily Taylorfelt that, unless yourtime is managed well, the analysis phase tends
to be squeezed-often withundesirableconsequences. Indeed, it is myexperience too from supervising
students' dissertationsthat they allow far too littletime for data analysis and writing up. Here is what
Hannahand Lily respectively wrote in response to a question askingwhat one singlebit of advicethey
wouldgive to others.
Get your researchdone as soonas possible. The processofanalysis isprettymuchan ongoing one and can
take averylongtime,sothe sooner youhaveallyourdata compiled the better. Italsomeansthat you have
moretimeto makemoreextensive analysis rather than just noticingthe surface emergenttrends.
Make sureyougive yourself enoughtimeto carryout the project, don't underestimate the amount oftime
data analysis cantake!
Similarly, Rebeccca Barneswrote that, ifshe was doingher research again:
Iwould alsoallocatemoretimefordata analysis and writing, as largely becauseof the longperiodof time
which ittookto recruit participants, these phasesof myresearchweresubjectto considerable timepressures
To readmoreaboutSarah's, Hannah's, Lily's. andRebecca's research experiences. goto the Online Resou.:e
Centre that accompanies thisbookat http://www.oxfordtextbooks.co.uk/ore/brymansrm3e/.



put at yoUr
xample,Will
suchthingS
itance, POSt-
nbeableto
'pmenr and
d andtran-
eyouneed,
ackage like
toestablish
;are finan,
gym study'
gationthat
~
ime.
sea ,;;;J
ead. t
'{
{will
!tends
vising
iat
hey
:ndcan
IUhave
Iftime
time
essures.
esource
wouldbe feasible within the kindof timeframeusually
programs; photocopying covering letters and question-
allocated to undergraduate and postgraduate disserta-
naires;postageforsendingthequestionnairesoutandfor
tions. However,itwouldrequiresuchfacilitiesas: typing
anyfollow-uplettersto non-respondents;returnpostage
u the questionnaire,whichnowadaysstudentscanusu-
for thequestionnaires;andthe availabilityofaquantita-
~ do forthemselveswiththe helpofword-processing tivedataanalysispackagelikeSPSS.
Studentexperience
Devisingatimetableforwritingup
LilyTaylorfound ithelpfulto haveatimetableofdeadlinesforthe differentsectionsofthe reportshe had
towrite.
I produced afirst draftofmyreportand madesurethatI gotitdone inplentyof timebeforethe deadline.
I wasthen abletogoovermyworkand makethenecessarychanges.I madesurethat I hadachecklist with
mini deadlines for each section, this made surethat I kept on top of mywork and progressed at a steady rate.
IsabellaRobbinswritesthatshe 'devisedawritinguptimetablewithaplanofthe thesis'.CorneliusGrebe
adoptedasimilarapproachto hiswritingup.Hewrites: 'Iagreedsubmissiondatesforindividualdraft
chapterswithmysupervisors.'
To readmoreabout Lily's, Isabella's, and Cornelius's research experiences, goto the Online Resource
Centre that accompanies thisbookat http://www.oxfordtextbooks.co.uk/ore/brymonsrm3el.

Formulatingsuitableresearchquestions
Manystudentswantto conductresearchintoareasthat Research questions are, therefore, important. No re-
areofpersonalinteresttothem.Thisisnotabadthingat searchquestionsorpoorlyformulatedresearchquestions
all and, as Inotedin Chapter1, manysocialresearchers will lead to poor research. If you do not specifyclear
startfrom this pointaswell(seealso Lofland andLofland researchquestions,thereisagreatriskthatyourresearch
1995: 11-14). However, you must move on to develop willbe unfocusedandthatyouwillbeunsureaboutwhat
research questions. This recommendation applies to yourresearchisaboutandwhatyou arecollectingdata
qualitativeresearchas well as quantitative research. for. It does notmatterhowwell you design a question-
Asisexplained inChapter 16, qualitative research ismore naire or how skilled an interviewer you are; you must be
open-ended thanquantitative research, and in Chapter clearaboutyourresearchquestions. Equally,it does not
17I refertosomenotablestudiesthatappearnotto have matterwhetheryourresearch is for a researchgrantof
beendrivenbyspecificresearchquestions.However,very 250,000, a doctoral thesis, or a small mini-project.
open-ended researchis risky andcan leadto thecollec- Researchquestionsare crucialbecausetheywill:
tionoftoomuchdataand,whenitcomestowritingup,to
guideyourliteraturesearch;
confusion about your focus. So, unless yoursupervisor
guideyou indecidingwhatdatayouneedtocollect;
advisesyou to thecontrary,Iwoulddefinitelyformulate
guideyouranalysisof yourdata;
someresearchquestions,evenif theytumouttobesome-
whatlessspecific thanthekindswe oftenfindinquantit-
guideyourwriting-upofyourdata;
ativeresearch.Inother words,whatisitaboutyourarea stopyou fromgoingoffinunnecessarydirectionsand
ofinterestthatyouwanttoknow? tangents.



Therefore,researchquestionshavemanyusesandyou researchquestionsmusthaveaclearsocial scientific(for
shouldresistthe temptationof notformulatingthemor example,sociological)angle.
delaying theirformulation. But do rememberthatyour
Step5
Thinkingdeeply3.1
Marx'ssourcesofresearchquestions
Marx(1997) suggeststhefollowing aspossiblesourcesofresearch questions.
Intellectualpuzzlesandcontradictions.
Theexistingliterature.
Replication.
Structures and functions. Forexample, ifyou point to a structure such as a type of organization, you can ask
questionsabout the reasonswhy there aredifferent typesandthe implicationsof thedifferences.
Opposition. Marx identifiesthe sensation of feelingthat a certain theoretical perspective or notable piece of
workismisguidedandofexploringthe reasonsforyouropposition.
Asocial problem. But remember that this isjust the source of a research question; you still have toidentify
social scientific(e.g, sociological)issues in relation toa social problem.
'Gaps betweenofficial versions of realityand the facts on the ground' (Marx1997:113).An example here is
something likeDelbridge's (1998) fascinating ethnographic account of company rhetoric aboutJapanized
workpracticesandhowtheyoperateinpractice.
The counter-intuitive. For example, when common sense seems to flyinthe face of social scientifictruths.
'Empirical examples that trigger amazement' (Marx1997: 114). Marxgives, as examples, deviant cases and
atypicalevents.
Newmethods andtheories.Howmightthey beappliedinnewsettings?
'New social and technical developments and social trends' (Marx 1997: 114).
Personalexperience.
Sponsorsandteachers. Butdonotexpect yourteacherstoprovideyouwithdetailedresearch questions.
Marx (1997) has suggested a wide range of possible AsFigure3.1 implies,weusuallystartoutwithagen-
sourcesofresearchquestions (seeThinkingdeeply3.1). eralresearchareathatinterestsus.It mayderivefromany
As this list makes clear, research questions can derive ofseveralsources;
from awidevarietyofcontexts.Figure3.1 bringsoutthe
Personalinterest/experience.AsIpointedoutinChap-
main steps in developing research questions. Research
terI,myinterestinthemeparkscanbetracedbacktoa
questions in quantitative research are sometimes more
visitto DisneyWorldinOrlandoin 1991 andmyinter-
specific thanin qualitativeresearch.Indeed,somequali-
est in the representationof social scienceresearch in
tativeresearchersadvocateaveryopenapproachwithno
themassmedia to a woundingencounter with thepress.
researchquestions.This isaveryriskyapproachandcan
Theory. Someone might be interested in testing or
result in collectinglots of datawithouta clearsenseof
exploring aspects of labour process theoryor in the
whattoobserveorwhattoaskyourinterviewees.Thereis
theoryoftherisk society.
a growingtendencyfor qualitative researchersto advo-
cate a somewhat more focused approach to their craft The research literature. Studiesrelatingto a research
(e.g. HammersleyandAtkinson1995: 24-9). area like modem consumerism might stimulate an



IOU can ask
blepiece of
:0 identify
pIehere is
ipanized
ictruths.
.asesand
estions,
ith a gen-
from any
in Chap-
backtoa
my inter-
search in
:he press.
!sting or
)f in the
research
alate an
Research area
Concerns about risk .
:Ij ... ""
Student experience
Theory as an influence on research questions
Rebecca Barnes's interest in feminist theories relatingto patriarchyinfluencedher selection of woman-to-
woman partner abuse asa focusfor her enquiries.
I became interested inthe topicof woman-to-woman partnerabuse asanundergraduate. Myfirst encounter
with this subject area tookthe formof atheoretical engagement with feministexplanations for domestic
violence-primarily emphasizing patriarchy-and the ways in which emerging knowledge aboutviolence
andabuse in female same-sex relationships challenges this understanding. It was asaresultof thisfirst
encounter that I became awareof the scarcity of research in thisarea, particularly inthe UK, wherethis
subject was virtuallyuncharted territory. I wasat this point interested in pursuing postgraduate study,
andthusdecided to conductmyown UK-based study of woman-to-woman partnerabuse for my Ph.D.
Theoretical ideas stimulatedGarethMatthews'sinterest in migrant labour. In his case, it waslabour process
theory that wasthe focus of his theoretical enquiry.
Primarily, myintereststems fromamoregeneral interestin Marxistlabourprocess theory, whichI believe to
behighlyrelevant to anunderstanding of the content of modern work-forms aswell as the claims that are





made by academics about these. Since Braverman published Labour andMonopolyCapitalin 1974, the
of t
labour process debate has taken many twists and turns, and the 'core' elements ofthe theory are now
(20
somewhat different from those expounded by Ido not seek simplyto reiterate the importance
not
of Braverman's formulation, but instead have attempted to explore the space between this and more
modern theoretical propositions-in the light of real and perceived changes inthe world of workand
The
workers.... Essentially, myapproach stems fromthe beliefthat the employment relation cannot simplybe
For e)
'read off fromanalyses of the content of jobs, and that it must instead be examined through an analysis of
focus
sourct
forces that operate at various levels(i.e. the workplace, the labour market, the state, etc), and fromthe
capita
interaction between these forces and employers' necessarily contradictory aims and pressures.
I
rrifica
As t
Toreadmore aboutRebecca'sand Gareth'sresearchexperiences, go to theOnline Resource Centrethat
start (
accompaniesthisbook at http://www.oxfordtextbooks.co.uk/ore/brymonsrm3el.
This r
interest in the nature of the shopping experience in
contemporary society.
Puzzles. An interesting example of this can be found
in a research article by Hodson (2004) in which
he employs data from the Workplace Ethnography
Project (see Research in focus 12.4). In this article he
notes that writings on modern work imply two rather
inconsistent views concerning the extent to which
workplaces today are a source of social fulfilment.
Some writers construe modern workplaces as intrinsi-
cally attractive environments to which people are
drawn; others writers view people's commitment to
social life at the workplace as stemming from job and
career insecurities. Hodson set up these two different
points of view explicitly as essentially rival hypo-
theses. Similarly, Wright etal. (2006) collected semi-
Student experience
can d
rions I
structured interview data on street robbers in the UK

to shed light on two different views of the motivation
sugge
for engaging in this crime. One view, which draws on
indies
rational choice theory, depicts street robbery as motiv.
resea
ated by a trade-off between the desire for financial
down
gain against the necessity to reduce the likelihood of
to spr
detection. The other view of street robbery portrays it
we h:
as a cultural activity from which perpetrators derived
an emotional thrill and which helped to sustain a par-
ticular lifestyle.
Newdevelopments insociety.Examples might include
the rise of the Internet and the diffusion of newmodels
of organization, e.g, call centres.
Social problem. An example might be the impact of
asylum-seekers being viewed as a social problem by
some sectors of society. This seems to have been one
New developments in society as a spur to
research questions
Lily Taylor was interested in the role of debt on the student experience. What, in other words. is the impact
of top-up fees on students' experiences of higher education?
Increasingly today more students are put off universitybecause ofthe amount of debt most students will
leave with. Particularlywith the topical debate at the time over the tuition fee system and top-up fees. I
believed it was an interesting area to lookat. Students are supposed to be concerned and worriedabout
essay deadlines and attending lectures and seminars, yet finance today seems to be the main anxiety for
most universitystudents.
Toreadmore aboutLily's research experiences, go to theOnline Resource Centre that accompanies this
bookathttp://www.oxfordtextbooks.co.uk/orc/brymonsrm3el.




4,the
now.
portance
.ore
3nd
iimplybe
ialysisof
mthe
-tre that
5intheUJ<
notivation
I drawson
vasmotiv,
r financial
elihoodof
portraysit
rsderived
tainapar-
htinclude
ewmodels
impactof
.oblemby
beenone
relmpact
tswill
es,I
ibout
tyfor
?s this
\\ "
a.researchprojectand formulatingresearchquestions 73
. ...._::::T s, <'
ofthemainfactorsbehind theworkofLynnandLea
We cannot answer all the research questions that
(2003), who examined the discourses surrounding the
occur to us. Thisis not just todo with issuesof time and
notionoftheasylum-seekerintheUK.
thecostofdoingresearch. Itisverymuchtodo with
the fact that we must keepaclear focussothatour
Thesesources ofinterestare not mutually exclusive.
researchquestionsmustrelatetoeachotherandform
Forexample,the investigation reported in Research in
acoherentsetofissues.
focus 1.1 was motivated by at least two of the above
Wethereforehavetoselectfromthepossibleresearch
sources: an interest in exploring the conceptof social
questionsthatwearriveat.
capital (theory) and understanding the processofgen-
trification(anewdevelopmentinsociety). Inmakingourselection,weshouldbeguidedbythe
/IJ3 these types of source suggest, in research we often principle that the research questions we chooseshould
startoutwith a general research area thatinterests us. berelatedtooneanother. Iftheyarenot,ourresearch
This researcharea hastobenarroweddownsothat we willprobablylackfocusandwemaynotmakeasclear
can developatighter focusout ofwhich researchques- acontributiontounderstandingaswouldbethecaseif
tionscanbedeveloped.Wecandepicttheprocessofgen- researchquestionswereconnected.Thus,intheexam-
eratingresearch questions as a series ofsteps that are pleinFigure3.1,theresearchquestionsrelatingtorisk
suggested inFigure3.1. Theseriesofstagesismeantto arecloselyconnected.
indicate that, when developing research questions, the
researcheris involvedin aprocess ofprogressivefocusing In Tips and skills 'Criteria for evaluating research ques-
downsothatwemovefromageneralresearchareadown tions'somesuggestionsarepresented aboutthekindsof
tospecificresearchquestions.Inmakingthismovement, considerations thatshould be taken into accountwhen
wehavetorecognizethat: developingyourownresearchquestions.
Student experience
The natureof research questions
Someof the students workedwithquite explicit and narrowly formulatedresearchquestions.Forexample.
RebeccaBarneswrites:
My researchquestions were:Whatformsand dynamicsofabuse do womenexperienceinsame-sex
relationships? Whatopportunitiesandchallenges do womenexperiencewithrespecttoseeking supportfor
woman-to-woman partnerabuse?Whatimpacts does beingabused byafemale partnerhaveuponwomen's
identities and biographies?How are women's accountsofwoman-to-womanpartnerabusesimilar to and
different from heterosexual women's accounts of partnerabuse?
IsabellaRobbinswassimilarlyexplicitaboutherresearchquestions:
1. How do mothers frame their decisions regarding childhood vaccination? Inparticular, do theypresentthis
as a matterofmoralobligation (totheir child/tothecommunity)?
2. Domothersconsider theyhaveachoice regarding childhood vaccination? Ifso, inwhat sensedothey see
thisas achoice andwhat,ifany,constraintsdo theyidentify as theyseektoexercise that choice?
3. How dowomen place themselves andtheirdecisions about childhood vaccination, intermsofthediscourse
ofrisk, responsibility, autonomy. and expertise?
4. Whatroledowomen accord to partners,mothers. siblings, and professionals intheirdecision-making about
childhoodvaccination?
Others opted for researchquestionsthat were somewhat moregeneral and wider infocus. Erin Sanders
writesof herresearchquestionsfor herstudy:




..'
r
What arethe policygoalsof women'sNGOsin Thailand?Howdo thesegoalsrelateto theneedsof women
inthesex industry?
Inasimilar vein, GarethMatthewswrites:
My research questions werequitegeneral.(i) what isthe roleof migrantworkers in the UK'shospitality
sector?(ii) What canthistellusabouttherelevanceand usefulnessof Marxistlabourprocesstheory?
Garethwentontowrite:
Thesequestionsstemfrommytheoreticalconcerns,andadesireforthethesistobeguidedbyttJefindings
andtheoretical developments in relationto thesefindings duringthecourse ofthe research. I did not wantto
beginwithaspecific hypothesis,andthen to proceedby attemptingto 'prove' or 'disprove' this, butsought
insteadto start with ageneraltheoreticalbelief about work, andthentoremainopen-mindedsoas to allow
the directionofresearchto beguidedbythe qualitativefindings asthey unfolded.
To readmoreabout Rebecca'5, Isabella'5, Erin '5, and Gareth's research experiences, goto the Online
Resource Centre that accompanies thisbookat http://www.oxfordtextbooks.co.uk/ore/brymansrm3el.
Researchquestionsfor adissertationor project exhibit the followingcharacteristics.
r.. Theyshouldbeclear, in thesenseof beingintelligible.
'.' Theyshouldberesearchable-that is,they shouldallowyouto doresearch in relation to them.This means
that theyshould not beformulatedin termsthat aresoabstractthat theycannot be converted into
researchableterms.
, Theyshouldhavesomeconneaionts) with established theory andresearch. Thismeansthatthereshouldbe
aliteratureon which youcan drawto help illuminatehow yourresearch questions should beapproached.
Even if youfind atopic thathas beenscarcely addressed by social scientists, it isunlikelythat therewill be
no relevantliterature(e.g,on relatedor paralleltopics).
" Yourresearchquestionsshouldbelinked toeachother. Unrelatedresearchquestions areunlikelyto be
acceptable, since you should bedeveloping anargument in yourdissertation. You couldnot veryreadily
constructasingle argumentin relationto unrelatedresearchquestions.
.0 Theyshouldat theveryleastholdouttheprospectofbeingabletomakeanoriginal contribution-however
small-tothetopic.
e. Theresearchquestionsshouldbeneither toobroad (sothat youwould needamassive granttostudythem)
nortoonarrow (sothat youcannotmakeareasonablysignificant contributionto yourareaof study).
If youarestuckabout howto formulateresearch questions (or indeedotherphases of yourresearch), it is
alwaysagoodideatolookat journal articlesor researchmonographsto seehowotherresearchershave
formulatedthem.Also, lookat pastdissertations for ideasas well. Marx(1997) has suggested a widerange of
sourcesof researchquestions(seeThinkingdeeply3.1).


;ofwomen
:>itality
ory?
Iefindings
notwantto;
lut sought
astoallow

iline
ansrm3e/.
ismeans
)
;hould be
cached.
will be
to be
~ i l y
-however
1ythem)
).
tis
/e
ngeof
Writingyourresearch proposal
In preparationforyour dissertationyoumayberequired
towrite a short proposal or plan outlining what your
researchproject will be about and how you intend to
goabout it. This is a useful way of preparing foryour
. researchanditwillencourageyoutothinkaboutmanyof
theissuesthatarecoveredinthenextsection.Inaddition
tooutliningyourproposedresearchdesignand methods,
thetopicarea inwhichyourstudyisgoingtobelocated,
andthe research questionsthat you intend to address, the
proposalwillaskyoutodemonstratesomeknowledge of
theliteraturein yourchosenfield, forexamplebyidentify-
ingseveral key authors or important research studies.
This information may be used as the basis for allocat-
inga supervisor who isknowledgeable in yourarea of
researchinterestorwho has experiencewith your pro-
posedresearch approach. The proposal isalso a useful
basis for discussion of your research project with your
supervisor, and, if itincludes atimetableforthe project,
thiscan provide a basis for planning regular meetings
withyoursupervisortoreviewyourprogress.Developing
atimetable can bevery importantin making you think
about aspects of the overall research process like the
differentstages ofyour researchand their timingand in
givingyouaseriesofongoinggoalstoaimfor.Evenifyou
are not required to produce a research proposal, it is
worthwhile constructing a timetable for your research
andaskingyoursupervisorto lookat it,so thatyoucan
assesshow (un)realisticyourgoalsare and whetheryou
areallowingenoughtime foreachofthe componentsof
theresearchprocess.
Whenwriting aresearchproposal, thereare anumber
ofissues that you will probably need to cover.
Whatis yourresearchtopicor,alternatively,whatare
yourresearchobjectives?
Whyis your research topic (or why are those research
objectives)important?
What is your research question or what are your
researchquestions?
What does the literature have to say about your
researchtopic/objectivesand researchquestion(s)?
Howareyougoingtogoaboutcollectingdatarelevant
to yourresearch question(s)? In other words, what
researchmethodsareyouintendingtouse?
Whyare the research methods/sources youhaveselected
the appropriate ones for your research question?
Whatresourceswillyouneed toconductyourresearch
(e.g. postage, travel costs, software) and how will
those resourcesbefunded?
What isyour timetableforthe different stages ofthe
project?
Whatproblemsdoyouanticipateindoingtheresearch
(e.g.accesstoorganizations)?
Whatarethepossibleethicalproblemsassociatedwith
yourresearch?
Howwillyouanalyseyourdata?
Writing a proposal is therefore useful in getting you
startedonyour researchproject and encouragingyouto
set realistic objectivesforyourresearchproject. Insome
highereducationinstitutions,the researchproposalmay
form part(albeitasmall one) ofthe overall assessment
ofthe dissertationor reportthatisproducedout ofthe
project. While the researchproposal isa working docu-
mentandtheideasthatyousetoutinitcanberefinedand
developedasyourresearchprogresses, itisimportantto
bearin mind thatifyou keepchangingyour mind about
your area of research interest and research design you
will be using up valuable time needed to complete the
dissertationwithinthedeadline.



I
1
.
76 Planning a research project
, , r'"

Preparing for your research
Do not begin your data collection until you have lead you into sampling considerations, such as the
identified your research questions reasonably clearly. following.
Develop your data-collection instruments with these
Whodoyouneed tostudyinordertoinvestigateYour
research questions at the forefront of your thinking. If
researchquestions?
youdonot dothis,there istheriskthat your resultswill
not allow you to illuminate the research questions. If at Howeasily can you gain accessto a sampling frame?
all possible, conduct a small pilot study to determine
Whatkindofsampling strategywillyouemploy(e.g.
howwellyourresearch instrumentswork.
probability sampling, quota sampling, theoretical
You will also need to think about access and sam-
sampling,conveniencesampling)?
plingissues. Ifyour researchrequires youtogainaccess
Canyoujustifyyourchoiceofsamplingmethod?
to or the cooperation of one or more closed settings
like an organization, you need to confirm at the earli- Also, while preparing for your data collection, you
est opportunitythatyou have the necessary permission should consider whetherthere are any possibleethical
to conduct your work. Youalso need to consider how problemsassociatedwithyourresearchmethodsoryour
youwillgoaboutgaining accesstopeople. Theseissues approachtocontactingpeople(seeChapter5).

Doing your research and analysing
, 1
"

your results
Sincedoingyourresearchand analysingyourresultsare conducting a questionnaire survey, begin coding
what the bulk of this book will be about, it is not necessary your data and entering them into SPSS or whatever
at this stage to gointo detail, but here are some useful package youare usingafter you have put togethera
hintsaboutpracticalities. reasonablysized batch ofcompleted questionnaires.
In the case of qualitative data, such as interview
Keepgood records ofwhatyoudo. Aresearch diary
transcripts, the same point applies, and, indeed, it
canbehelpfulhere, butthere areseveralotherthings
is a specific recommendation of the proponents of
to bearin mind. For example, ifyou are doing a survey
grounded theory that data collection and analysis
bypostal questionnaire,keep good records ofwho
shouldbeintertwined.
has replied, so that you know who should be sent
reminders. Ifparticipantobservationisacomponent
Remember that the transcription of recorded inter-
ofyour research, remembertokeepgoodfield notes
viewstakesalongtime.Allowatleastsixhours'tran-
andnottorelyonyourmemory.
scriptionforeveryonehourofrecordedinterviewtalk,
atleastintheearlystagesoftranscription.
Makesure thatyou are thoroughly familiarwith any
hardwareyouareusingincollectingyourdata,suchas Becomefamiliar with any data analysis packagesas
tape recorders for interviewing, and make sure it is in soon as possible. This familiaritywill help you toestab-
goodworking order (e.g.batteriesthat are notflator lishwhetheryoudefinitelyneed themandwillensure
closetobeingflat). thatyoudonotneedtolearneverythingaboutthemat
theverytimeyouneedtousethemforyouranalysis.
Donotwait untilallyourdata havebeen collectedto
begin coding. This recommendation applies to both Donotatanytimetakerisks withyourpersonal safety
quantitative and qualitative research. If you are (see Tips and skills'Safety in research').


uch as the
stigateYOUr
lngframe?
mploy Ie.g,
theoretical
thod?
ection, you
ible ethical
ads or your
I.
in coding
. whatever
together a
:ionnaires.
interview
indeed, it
onents of
j analysis
Ied inter-
iursrran-
view talk,
ckages as
t to estab-
'ill ensure
It them at
malysis,
nal safety
.
research questions 77
Tips and skills
Safety in research
Inthe middleof December 2002 a19-year-old female student who hadjust started a degree course in sociology
andcommunitystudies at Manchester MetropolitanUniversitywent missing. It was believed that, in order to
complete a coursework assignment, shehad gone to conducta life history interview with a person aged over
50. Since.she was interested inthe homeless, it wasthought that shehad gone to interviewa homeless person.
Because of concerns about her safety, her tutor hadadvised her to takeafriend andto conductthe interviewin
apublicplace. Infact, she hadnot gone to conductthe interviewandto everyone's relief turnedup in Dublin.
Thereisanimportant lesson in thisincident:youmust bear in mind that social research mayon occasions place
youin potentiallydangerous situations. Youshouldavoidtaking personal risks at all costs andyoushouldresist
anyattemptsto place yourselfin situations wherepersonal harmisa real possibility. Justasyou should ensure
that noharmcomes to research participants (asprescribed in the discussion of ethical principles in Chapter 5),
individuals involved in directingothers' research shouldnot place students andresearchers in situations in
whichthey might cometo harm. Equally, loneresearchers shouldavoid such situations. Sometimes, as with
the interviews with the homeless, there issome possibility of beingin a hazardous situation, in whichcase, if
the researcher feels confidentabout goingahead with the interview, heor sheneeds to takeprecautions
beforegoingahead with the interview. The advice givenby the student's tutor-to takesomeone with her and
to conduct the interviewin apublic place-s-was verysensible for a potentially dangerous interview. If youhave
a mobiletelephone, keepit with youandkeepit switched on. Personal attackalarms mayalso beuseful. You
should also make surethat, if your interviews or your periods of observation are part of a programme of work.
youestablish a routinewherebyyoukeepin regular contact with others. However, thereare situations in
whichthere isno obvious reason to think that asituationmaybe dangerous, but wherethe researcher is
faced with a sudden outburst of abuse or threateningbehaviour. Thiscanarise whenpeoplereactrelatively
unpredictably to aninterviewquestion or to being observed. If there aresigns that such behaviour isimminent
(e.g. throughbody language), beginawithdrawal fromthe research situation. Further guidelines onthese
issues canbefoundin Craig et 01. (2000).
Lee (2004) drawsanimportant distinctionbetweentwo kinds of dangerin fieldwork: ambientandsituational.
Theformer refers to situations that areavoidable andin whichdangerisaningredientof the context.
Fieldwork in conflict situations of the kindencountered bythe researcher whotook on the roleof a bouncer
(Hobbset 01. 2003) would bean example of this kind of danger. Situational danger occurs 'whenthe
researcher's presence or activities evokeaggression, hostility or violence from thosewithin the setting'
(Lee2004: 1285). Whileproblems surrounding safety maybe easier to anticipate inthe case of ambient danger,
they are less easy to foresee in connection with situational danger. However, that isnot to say that ambient
danger isentirely predictable. It was only some time after shebegan her research in a hospital laboratory that
Lankshear (2000) realized that there was a possibility of her beingexposed to dangerous pathogens.
Sources: P. Barkham and R. Jenkins, 'Fears forFresher who Vanished onMission totalk totheHomeless', The Times.
13 Dec. 2002; S. Mcintyre, 'How didVicky Vanish?', DailyMail, 13 Dec. 2002; R.Jenkins, 'Wasteland Search forMissing
Student', The Times, 14 Dec. 2002.



-------------
Checklist
Planning a research project
o Do youknowwhat the requirements for your dissertationare, as set out byyour university or
department?
0'Haveyoumadecontactwithyoursupervisor?
oHaveyouallowedenough time forplanning, doing,andwriting upyourresearch project?
oDo youhavea cleartimetablefor your research project withclearly identifiable milestones for the
achievementofspecifictasks?
o Haveyougot sufficientfinancial andpractical resources (e.g, money to enable travel toresearch site,
recording device) to enableyouto carryout yourresearch project?
o Haveyouformulatedsomeresearchquestions anddiscussedthesewith your supervisor?
o Aretheresearch questions youhaveidentifiedcapable of beinganswered throughyourresearch
project?
o Do youhavetheaccessthat yourequirein order to carryout yourresearch?
o Are youfamiliarwith the dataanalysissoftware thatyouwill beusing to analyseyour data?
o Haveyouallowedotherstocommenton your work sofar andrespondedtotheirfeedback?
o Haveyoucheckedout whether therearelikelyto beanyethical issues that might beraisedin
connectionwithyourresearch?
o Haveyouallowedenoughtime for gettingclearance throughanethics committee,if that isrequired
foryourresearch?
Key points
Followthe dissertationguidelines provided by your institution.
Thinking about yourresearch subject canbetimeconsuming, soallowplenty of time for thisaspect
ofthedissertationprocess.
Useyour supervisorto the fullest extent allowedand follow the adviceofferedby himor her.
Planyour time carefullyand be realistic about what you canachieve in the time available.
Formulatesomeresearch questionstoexpress what it isabout your areaof interest that you want to
know.
Writing aresearch proposal isagoodway of getting started on yourresearch project and
encouragingyou toset realisticobjectives.
Consideraccess andsamplingissues at an earlystage and consider testingyourresearch methodsby
conductingapilotstudy.
Keepgoodrecordsof what you do in yourresearch asyou goalongand don't wait until all your data
havebeencollected beforeyoustart coding.



. ' ,
.' ;iIl$'{F
tell
andformulating research questions 79
".
Questions for review
lor
;for the
earch site,
search
?
in
required
saspect
r.
I want to
thods by
our data
Managingtime and resources
Why isit important to devisea timetable for your research project?
Formulating suitableresearch questions
What arethe main sources of research questions?
What arethe main steps involved in developing research questions?
What criteria canbe usedto evaluateresearch questions?
Writingyour research proposal
What isthe purposeof the research proposal and how canit be useful?
Online Resource Centre
http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/
Visit the OnlineResource Centre that accompanies this book to enrichyour understanding of
planning a research project andformulating research questions. Consult weblinks, testyourself
usingmultiple choice questions, andgain further guidanceand inspiration fromtheStudent
Researcher's Toolkit.



Writing up social
research
Chapter outline
Introduction 661
Writing up yourresearch 662
Startearly 662
Bepersuasive 662
Getfeedback 663
Avoidsexist,racist, and disablist language 663
Structureyourwriting 663
Writingupquantitative.qualitative. andmixedmethods research 668
Writingupquantitativeresearch 669
Writingupqualitativeresearch 672
Writingupmixedmethodsresearch 675
Postmodernism anditsimplications for writing 679
Writingethnography 684
Experiential authority 685
Typicalforms 685
Thenative'spointofview 685
Interpretativeomnipotence 686
Checklist 686
Key points 688
Questions forreview 688

Chapter guide
It is easy to forget that one of the main stages in any research project, regardless of its size, is that it has
to be written up. Not only is this how you will convey your findings, but being aware of the significance
of writing is crucial, because your audience must be persuaded about the credibility and importance of
your research. This chapter presents some of the characteristics of the writing-up of social research.
The chapter explores:
why writing, and especially good writing. is important to social research;
using examples, how quantitative and qualitative research are composed;
the influence and implications of postmodernism for writing;
key issues raised by discussions about the writing of ethnography, an area in which discussions about
writing have been especially prominent.

Introduction
The aim of this chapter is to examine some of the strat-
egies that are employed in writing up social research. Initi-
ally, we will explore the question of whether quantitative
and qualitative research reveal divergent approaches. As
we will see, the similarities are frequently more striking and
apparent than the differences. However, the main point
of this chapter is to extract some principles of good prac-
tice that can be developed and incorporated into your
own writing. This is an important issue, since many peo-
ple find writing up research more difficult than carrying it
out. On the other hand, many people treat the writing-up
stage as relatively unproblematic. But no matter how well
research is conducted, others (that is, your readers) have
tobe convinced about the credibility of the knowledge
"..
Keyconcept27.1
Whatis rhetoric? Q
claims you are making. Good writing is therefore very
much to do with developing your style so that it is persua-
sive and convincing. Flat, lifeless, uncertain writing does
not have the power to persuade and convince. In explor-
ing these issues, I will touch on rhetorical strategies inthe
writing of social research (see Thinking deeply 27.2).
As Atkinson (1990: 2) has observed in relation to social
research, 'the conventions of text and rhetoric are among
the ways in which reality is constructed'. This chapter
will review some of the ways in which social research is
written up in a way that will provide some basic ideas
about structuring your own written work if you have to
produce something like a dissertation.
The study of rhetoric isfundamentally concerned with the ways in which attempts to convince or persuade
an audience are formulated. We often encounter the term in a negative context, such as 'mere rhetoric' or the
opposition of 'rhetoric and reality'. However, rhetoric is an essential ingredient of writing, because when we
write our aimis to convince others about the credibilityof our knowledge claims. To suggest that rhetoric
should somehow be suppressed makes little sense. since it is in fad a basic feature of writing, The examination
of rhetorical strategies in written texts based on social research is concerned with the identification of the
techniques in those texts that are designed to convince and persuade.




Writingupyourresearch
It iseasy to neglect the writing stage of your work because
to present and justify the research questions that are driv_
of the difficulties that you often encounter in getting your
ing your research or how to structure the theoretical and
research under way. But-obvious though this point is
research literature that will have been used to frame Your
-your dissertation has to be written. Your findings research questions. Students often tend to underestimate
must be conveyed to an audience, something that all of the time that it will take to write up their research, so it is
us who carry out research have to face. The first bit of a good idea to allow plenty of time for this, especiallyif
advice is ... you are expecting your supervisor to read and comment
onan early draft, since you will need to allow him orhera
reasonable amount of time for this. A further reason Why
Startearly
it is advisable to begin writing earlier rather than later is
It is easy to take the view that the writing-up of your an entirely practical one: many people find it difficult to
research findings is something that you can think about get started and employ (probably unwittingly) procrasti-
after you have collected and analysed your data. There is, nation strategies to put off the inevitable. This tendency
of course, a grain of truth in this view, in that you could can result in the writing being left until the last minute
hardly write up your findings until you know what they and consequentlyrushed. Writing under this kind ofpres-
are, which is something that you can know only once you sure is not ideal. How you represent your findings and
have gathered and analysed your data. However, there conclusions is a crucial stage in the research process. If
are good reasons for beginning writing early on, since you you do not provide a convincing account of your research,
might want to start thinking about such issues as how best you will not do justice to it.
Studentexperience
Writingupis difficult
Several of the students mentioned that they found writing up difficult. Gareth Matthews comments that
he 'found this stage the most difficult'. Isabella Robbins admits that writing the chapters presenting her
findings was 'the most difficult task of the Ph.D. process'. Havingenough time for writingup isa common
refrain in their questionnaires. Sarah Hanson's advice is:
The only problemwith a writingproject ofthis size istime. As it is always against you, start early. and be
organized, do one thing at a time. Work chronologically. Lecturersand markers liketo see that you have
gone on a journey of exploration into an interesting world and at the end have come out with something
worthwhilethat has changed your thinking and will hopefullychallenge theirs.
To read more about Gareth's, Isabella's, and Sarah's research experiences, go to the Online Resource
Centre that accompanies this book at http://www.oxfordtextbooks.co.uk/orc/brymansrm3el.
conclusions. Writing up your research will contain many
Bepersuasive
other features, such as referring to the literature on which
This point is crucial. Writing up your research is not sim- you drew, explaining how you did your research, and out-
ply a matter of reporting your findings and drawing some lining how you conducted your analysis. But above all,
yOU rn:
vince)
Simpl)
not en
findinl
plausil
Get
Tryta
andre
what
sauro
sors a
with c
tions
feedb
St.
It rr
10,1
stn,
dis;
Tit
YOI
she
Ac
Yo
pe-




:U and

itnate
;0 it is
allyif
[ment
'hera
1Why
Iter is
ult to
rasn.
lency
inute
pres-
: and
ss. If
arch,
n
my
ich
ut-
311,
yoU must be persuasive. This means that you must con-
vinceyour readers of the credibility of your conclusions.
Silnplysaying 'this is what I found; isn't it interesting' is
not enough. You must persuade your readers that your
findings and conclusion are significant and that they are
plausible.
Get feedback
Try to get as much feedback on your writing as possible
and respond positively to the points anyone makes about
what they read. Your supervisor is likely to be the main
source of feedback, but institutions vary in what supervi-
sors are allowed to comment on. Provide your supervisor
with drafts of your work to the fullest extent that regula-
tions will allow. Give him or her plentyof time to provide
feedback. There will be others like you who will want

':
Tips and skills
f,

1'.':-
r
Non-sexist writing
your supervisor to comment on their work, and, if he or
she feels rushed, the comments may be less helpful. Also,
you ask others on the same degree programme to
read your drafts and comment on them. They may ask
you to do the same. Their comments may be very useful,
but, by and large, your supervisor's comments are the
main ones you should seek out.
Avoid sexist, racist, and disablist
language
Remember that your writing should be free of sexist,
racist, and disablist language. The British Sociological
Association provides very good general and specific
advice about this issue, which can be found at http://
www.britsoc.co.uklequality/ (accessed on 16 July
2007).
One ofthe biggest problems(but by no means the onlyone) when tryingto write ina non-sexist wayis
avoidingcomplex his/her formulations. The easiest wayof dealingwith this isto write inthe plural insuch
circumstances.Consider, for example: 'I wanted to giveeach respondent the opportunityto complete the
questionnaire in his or her own time and ina locationthat was convenient for himor her.' Thisisa rather
tortuous sentence and, although grammatically correct, it could be phrased more helpfully as: "wanted to give
respondents the opportunityto complete their questionnaires intheir own time and ina locationthat was
convenient for them,
Structure your writing
It may be that you have to write a dissertation of around
10,000-15,000 words for your degree. How might it be
structured? The following is typical of the structure of a
dissertation.
Title page
You should examine your institution's rules about what
should be entered here.
Acknowledgements
You might want to acknowledge the help of various
people, such as gatekeepers who gave you access to an
organization, people who have read your drafts and
provided you with feedback, or your supervisor for his or
her advice.
Listof contents
Your institution may have recommendations or prescrip-
tions about the form this should take.
An abstract
A brief summary of your dissertation. Not all institu-
tions require this component, so check on whether it is
required. Journal articles usually have abstracts, so you
can draw on these for guidance on how to approach
this task.



10
Introduction
The followingaresomepointsto considerwhenwriting
an introduction.
You should explain what you arewritingaboutand
why it is important. Saying simply that it interests
you because of a long-standing personal interest is
notenough.
You might indicate in general terms the theoretical
approachorperspectiveyouwillbeusingandwhy.
You should also at this point outline your research
questions.In the case ofdissertationsbasedon quali-
tative research, it is likely that your research ques-
tionswill be rathermoreopen-endedthanisthecase
withquantitativeresearch.Butdotrytoidentifysome
research questions. A totally open-ended research
focusisriskyand canleadtothecollectionoftoomuch
data,and,whenitcomestowritingup, itcan resultin
alackoffocus.
The openingsentenceor sentencesareoftenthemost
difficult ofall. Becker(1986)advisesstronglyagainst
openingsentencesthathe describesas 'vacuous' and
'evasive'. He gives the example of 'This study deals
withtheproblemofcareers', and addsthatthis kind
of sentence employs 'a typically evasive manoeuvre,
pointing to something without saying anything, or
anythingmuch,aboutit.What aboutcareers?'(Becker
1986: 51). He suggests that such evasiveness often
Tips and skills
occursbecauseofconcernsaboutgivingawaythe
111 fact, he argues, it ismuchbetterto give t.
quickandclearindicationofwhatisgoingtobe
outtothemandwhereitisgoing.
literature review
See Chapter 4 for more detailed advice on how to 0
aboutwritingthischapterofyourdissertation. g
Research methods
The term 'research methods' is meant here asa kind
of catch-all for several issues thatneed to be outlined:
your research design; your sampling approach; how
accesswas achievedifrelevant;the proceduresyouused
(such as, if you sent out a postal questionnaire, did
youfollowupnon-respondents);thenatureofyourques-
tionnaire,interviewschedule,participantobservation
role, observationschedule,codingframe,orwhatever
(thesewill usuallyappearinanappendix,butyoushould
commenton suchthings as yourstyle ofquestioning or
observationandwhyyouaskedthethingsyoudid);prob-
lems ofnon-response; note taking; issues of ongoing
access and cooperation; coding matters; and how you
proceeded withyour analysis. When discussing eachof
theseissues, you shoulddescribeanddefendthe choices
thatyou made,suchaswhyyou used a postal question-
naire rather than a structuredinterview approach, or
why you focused upon that particular population for
samplingpurposes.

In1
yOl
Iik,
me
yOl
wil
of
lite

The importance of an argument

Inmy experience,oneof thethings thatstudentsfind mostdifficultaboutwritingup their research isthe
formulationofanargument.Thewriting-upofresearchshouldbeorganizedaroundanargumentthatlinks
all aspects of the research process fromproblemformulation, through literature reviewand the presentation
of research methods, to the discussionand conclusion. Too often, students make a series of points without
askingwhat the contribution of those points isto the overall argument that they are tryingto present.Consider
what your claimtoknowledge isand tryto organize your writingto support and enhance it. Thatwill be
your argument.Sometimes it is useful to think interms of seekingtotell a story aboutyour research andyour
findings. Tryto avoidtangentsand irrelevant material that may mean that your readerswill lose the thread
of your argument. Ifyou arenot able to supplya clear argument, you are veryvulnerable to the 'so what?'
question. Askyourself: 'What isthe keypoint or message thatIwant my readers to take awaywith themwhen
they have finished reading mywork?' Ifyou cannot answer that simplequestion satisfactorily (and it maybe
worth tryingit outonothers), almost certainlyyou donothave an argument.The argumentisa threadthat
runsthroughyourdissertation(seeFigure27.1 foranillustrationofthis).



heplot.
adersa
~ m t
vtogo
akind
Itlined:
h; how
JUused
re, did
lrques-
rvation
hatever
should
ningor
);prob-
mgoing
JW you
eachof
choices
iestion-
ach,or
ion for
nks
ation
)ut
)nsider
your
:ad
?'
when
(be
hat
The role of an argumentina dissertation
Introduction A
R
G
Literature review
Research methods
U
M
Results
E
N
T
Discussion
Conclusion .
Results
Inthischapteryoupresentthe bulk ofyour findings.If
youintend to have a separate Discussion chapter, it is
likelythatthe results willbe presentedwith little com-
mentaryin terms ofthe literatureor the implicationsof
yourfindings.Ifthere willbe noDiscussionchapter,you
willneed to providesomereflectionsonthesignificance
ofyour findingsforyour research questions and forthe
literature. Bearthesepointsinmind.
Whichever approach you take, remember not to
include all your results. Youshould present and dis-
cussonly those findingsthat relate to your research
questions. This requirement may mean a rather
painful process of leaving out many findings, but it
isnecessary, so that the thread ofyour argumentis
not lost (see Tips and skills 'The importance of an
argument' for more on the significance of having a
goodargument).
Yourwriting should point to particularly salient as-
pectsofthe tables, graphs, orotherformsofanalysis
you present. Do not just summarize what a table
shows;youshoulddirectthereadertothecomponent
orcomponents ofit that are especiallystriking from
thepoint of viewof your researchquestions.Try to ask
yourselfwhatstoryyouwant thetable toconveyand
trytorelaythatstorytoyourreaders.
Another sin to be avoided is simply presenting a
graphortableorasectionofthe transcriptofasemi-
structuredintervieworfocus groupsessionwithout
any comment whatsoever, because the reader isleft
wonderingwhyyouthinkthefindingisimportant.
When reporting quantitative findings, it is quite a
good idea to vary wherever possible the method of
presentingresults-forexample,provideamixtureof
diagrams and tables. However,you must remember
the lessons of Chapter 14 concerning the methods
of analysis that are appropriate to differenttypes of
variable.
A particular problem that can arise with qualitative
research isthatstudentsfind itdifficult to leave out
large parts oftheir data. Asone experienced qualita-
tiveresearcherhasputit:'Themajorproblemweface
inqualitative inquiryisnottogetdata,buttogetridof
it!' (Wolcott 1990a: 18). Hegoeson to say that the
'critical taskinqualitativeresearch isnot to accumu-
late allthedatayou can,but to"can"[i.e.get ridof]
mostofthedatayouaccumulate'(Wolcott1990a:35).
Yousimply have to recognize that much of the rich
datayouaccumulatewillhavetobejettisoned.Ifyou
donotdothis,anysenseofan argumentinyourwork
is likely to be lost. There is also the risk that your
account ofyour findingswill appear too descriptive
andlackananalyticaledge.Thisiswhyitisimportant
touseresearchquestions asafocusand toorient the
presentation of yourfindings to them.It is also import-
ant tokeepinmindthetheoretical ideasand theliter-
ature that have framed your work. The theory and
literaturethat haveinfluencedyourthinkingwillalso
haveshapedyourresearchquestions.




Student experience
Do not try to write up everything
Youwill not be able to write up everything thatyou havefound.Sophie Masonrecognized this.She writes:
Thegreat quantity of data meant thatI hadtousemy ownjudgementasto what datawasthemostrelevant
to the aimsof theresearch. I alsohadto be carefulto usevisual aids whenusingcomplicatedstatistics to
emphasizetheimportanceoftheresults.
RebeccaBarneswrites:
Becausesomanyimportant and interestingissues haveemergedin theanalysisof mydata,I havehadtobe
selective;I havechosen to do justicetoasmallernumber ofthemes. rather thanresorting tosuperficial
coverageof a largernumber of themes.
To readmoreabout Sophie's and Rebecca's research experiences. go to the Online Resource Centre that
accompanies thisbook at http://www.oxfordtextbooks.co.uk/orc/brymansrm3el.
Student experience
The importance of research questions, theory,
and the literature in writing up findings
Several studentsmentioned how important itwasfor them to keepin mind theirresearch Questions and
thetheoryandliterature that weredrivingtheirresearchwhilewritingup.Foronething,theyhelpthe
studenttodecidewhichfindingstoincludeortoemphasizewhenwriting up.RebeccaBarneswrites:
I chose to havethree chaptersof my thesisthat reported myfindings.and Ichosethe themes that I would
includein eachofthese chapters. These were not,however, set instone, andhavechanged ina numberof
respectsfromwhen I first startedtoplan the writing-up.Eachof thesechaptersaddresses oneof my main
researchquestionsoraims.
ErinSanders writes: 'First I wrote down the main points and ideasI wanted to get across-and how my
findings related to[my] researchQuestion.'HannahCreane's writing-up of her findingswasgeared to her
research Questions.
Igroupedtogetherquestionsandresponsesthatconcernedsimilaraspectswithinthechildhooddebateand
formedthreemainchapters:What makesachild achild?;Childhoodpasttimes;andThechildtoday.Within
thesechaptersI interwovethemesthatemergedfromthe dataandseemedto be presentin most responses.
For GarethMatthews the theoretical debates about the labourprocess werecrucial: 'Thishasallowedme
to frame my thesistheoretically,and to lay the foundations for adiscussion of my empirical findings.'
To readmoreabout Rebecca's. Erin '5, Hannah's, and Gareth's research experiences, go to the Online
Resource Centre that accompanies thisbook at http://www.oxfordtextbooks.co.uk/orc/brymansrm3el.
If yo'
orP)
than
pres
ing
issu
shoi
are
sigr
In 1
cles
linl
Discu
Inthe
findm
reseal
your
these
hypot
migh
impli
Con.
The!
A
it
g
'"
o
s
)'
.')
(,



---------------- -
'e that
"
15and
Ithe
es:
would
mberof
I main
rrny
Itoher
bateand
v. Within
esoonses,
vedme
s.'
'ne
srmse/.
Ifyouarewritingathesis-forexample,foranM.Phil.
or Ph.D. degree-itislikely thatyou will have more
thanone and possiblyseveralchaptersin whichyou
presentyourresults.Cryer(1996)recommendsshow-
ing at the beginning ofeach chapter the particular
issues that are being examined in the chapter. You
shouldindicatewhichresearchquestionor questions
are beingaddressed inthe chapterandprovidesome
signpostsaboutwhatwill be includedinthechapter.
In the conclusion of the chapter, you should make
clearwhatyour resultshave shownanddrawoutany
linksthatmightbemadewiththenextresultschapter.
Discussion
Inthe Discussion, you reflectonthe implicationsofyour
findingsfortheresearchquestionsthathave drivenyour
research. Inotherwords,howdoyourresultsilluminate
your research questions? If you have specified hypo-
theses, thediscussion will revolve around whetherthe
hypotheseshave beenconfirmedornot, and, ifnot, you
mightspeculateaboutsome possiblereasonsforandthe
implicationsoftheirrefutation.
Conclusion
Themainpointshereare asfollows.
AConclusionisnotthesameasasummary.However,
itisfrequentlyusefultobringout inthe openingpara-
graphoftheConclusionyourargumentthus far. This
will meanrelatingyourfindings andyour discussion
ofthemto yourresearchquestions. Thus, yourbrief
summaryshould be a meansof hammeringhometo
yourreadersthe significanceofwhatyouhavedone.
Youshould make cleartheimplications ofyourfind-
ingsforyourresearchquestions.
Student experience
Youmightsuggestsomeways inwhichyourfindings
haveimplicationsfortheoriesrelatingtoyourarea of
interest.
You mightdrawattention to any limitations ofyour
researchwiththe benefitofhindsight, butit isprob-
ablybestnottooverdothiselementandprovideexam-
iners withtoo much ammunitionthatmightbe used
againstyou!
Itisoftenvaluabletoproposeareasoffurtherresearch
thatare suggestedbyyourfindings.
Two thingsto avoid are engaginginspeculationsthat
takeyoutoofarawayfromyourdata,orthatcannot be
substantiated by thedata, and introducing issues or
ideasthathavenotpreviouslybeenbroughtup.
Appendices
Inyourappendicesyoumightwanttoincludesuchthings
as your questionnaire, coding frame, or observation
schedule,letterssenttosamplemembers,andletterssent
to andreceivedfrom gatekeeperswherethecooperation
ofanorganizationwas required.
References
Includehereallreferencescitedinthetext.Fortheformat
of the References sectionyou should follow whichever
one isprescribedbyyourdepartment.Nowadays,thefor-
matisusuallyavariationofthe Harvardmethod,such as
the one employedforthisbook.
Finally
Remembertofulfilanyobligationsyouenteredinto,such
as supplyingacopy ofyourdissertation,if,for example,
youraccessto anorganizationwas predicatedonprovid-
ing one, andmaintainingthe confidentialityof informa-
tionsuppliedandthe anonymityofyourinformantsand
otherresearchparticipants.
Structure of the dissertation or thesis
Some of the students wrote up their work with a similar structure to the one that has been outlined inthis
section.SophieMasonwrites:
The research project was written invarious stages and split into several differentsections; thesewere as
follows: IntroductionandAims, LiteratureReview, Research Designand Data Gathering, DataAnalysis and
Research Findings, Conclusions and Recommendations, AppendixandBibliography.
-



ErinSanders writes:
I wrote it in order, introduction, literature review, research design, findings, discussion, and conclusion. It
each section as ifit were an essay inand of itself, and attempted to break it down intochunks soas nott oak
lost in a long document.
To read more about Sophie's and Erin'sresearch experiences, go to the Online ResourceCentrethat
accompanies this book at http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/'
Ti ps and skills
Proof reading your dissertation
Beforesubmitting your dissertation, make sure that it is spell-checkedand check it forgrammaticaland
punctuation errors. There are many useful guides and handbooks that can be used forthis purpose. It mayalso
be useful to ask someone else, such as a friend or family member, to proof read your workincase there are
errors that you have missed. Aswell as being an important presentational issue, this will affect the ease with
which your written work can be read and understood. It therefore has the potential to affect the qualityofyour
dissertation significantly.

Writing up quantitative, qualitative,
and mixed methods research
of how
explor
t
practit
metho
other 1
tions f
area tI
oget
Wri1
Toilh
rative
take I
refen
(see '
sugg'
repre
that
of pi
secc
natii
mas
Soci
Am.
pub
refe
de
the
its
art:
jou
de
ace
wi
au
ve
fu
at
re
G
bl
tl
a
n
P
d
In the next three sections, research-based articles that
have been published in journals are examined to detect
some helpful features. One is based on quantitative re-
search, one on qualitative research, and another on mixed
methods research. The presentation of the quantitative
and the qualitative research articles raises the question of
whether practitioners of the two research strategies em-
ploy different writing approaches. It issometimes suggested
that they do, though, when I compared two articles based
on research in the sociology of work, I found that the dif-
ferences were less pronounced than I had anticipated on
the basis of reading the literature on the topic (Bryman
1998). One difference that I have noticed isthat, injournals,
quantitative researchers often give more detailed accounts
of their research design, research methods, and approaches
to analysis than qualitative researchers. This is surprising,
because, in books reporting their research, qualitative re-
searchers provide detailed accounts of these areas. Indeed,
the chapters in Part Three of this book rely heavily on
these accounts. Wolcott (1990a: 27) has also noticed this
tendency: 'Our [qualitative researchers'] failure to render
full and complete disclosure about our data-gathering
procedures give our methodologicallyoriented colleagues
fits. And rightly so, especially for those among them will-
ing to accept our contributions if we would only provide
more careful data about our data.' Being informed that
a study was based on a year's participant observation or
a number of semi-structured interviews is not enough
to gain an acceptance of the claims to credibility that a
writer might be wishing to convey.
However, this point aside, in the discussion that fol-
lows, although one article based on quantitative research
and one based on qualitative research will be examined,
we should not be too surprised if they tum out to be more
similar than might have been expected. In other words,
. although we might have expected clear differences
between the two in terms of their approaches to writing,
the similarities are more noticeable than the differences.
In addition to looking at examples of writing in quanti-
tative and qualitative research, I will examine the matter



layalso
are
with
ofyour
render
thering
leagues
imwill-
provide
ed that
itionor
enough
r that a
hat fol-
eseareh
mined,
'emore
words,
erences
vriting,
ences.
quanti-
matter
ofhow mixed methods researchcan be writtenup and
explore some guidelines that are being proffered by
practitioners. The approach to dealing with the mixed
methods research article is slightly different from the
othertwOin thatIwill begin withsome generalsugges-
tionsforwritingup mixedmethodsresearchasthisisan
areathathasnotbeengivenagreatdeal ofattention.
.Writing upquantitativeresearch
Toillustratesomeofthecharacteristicsofthewayquanti-
tativeresearchiswrittenup foracademicjournals,Iwill
takethe articleby KelleyandDeGraaf (1997) thatwas
referredtoonseveraloccasionsinChapters1,2,6,and13
(seeespeciallyResearchin focus 1.4 and6.3). Iam not
suggesting that this article is somehow exemplary or
representative, butratherthatit exhibits some features
that are often regarded as desirable qualities in terms
ofpresentation and structure. The article isbasedon a
secondaryanalysisof survey dataon religioninfifteen
nations and was accepted for publication in oneof the
most prestigious journals in sociology-the American
Sociological Review, which is the officialjournal of the
American Sociological Association. The vast majorityof
published articles in academicjournals entail the blind
refereeingofarticlessubmitted.Thismeansthatan arti-
clewillbe read bytwo or threepeers, who commenton
thearticleandgivetheeditorsajudgementaboutitsmer-
its andhence whetherit isworthyof publication. Most
articles submittedare rejected. With highly prestigious
journals,itiscommonforinexcessof90percentofarti-
cles to be rejected. It is unusual for an article to be
accepted on its first submission. Usually, the referees
will suggestareas thatneedrevising andtheauthor(or
authors)isexpectedtorespondtothatfeedback. Revised
versionsofarticlesareusuallysentbacktotherefereesfor
further comment, and this process may result in the
authorhaving to revise the draftyet again. Itmay even
result in rejection. Therefore, anarticle like Kelleyand De
Graaf'sisnotjustthe culminationofaresearchprocess,
butis also theoutcomeof a feedback process. The fact
that it has been accepted for publication, when many
othershavebeenrejected,testifies toitsmerits ashaving
met the standardsofthe journal.Thatisnot to say it is
perfect,butthe refereeingprocess isan indicationthatit
doespossesscertaincrucialqualities.
Structure
Thearticlehasthe followingcomponents,aside from the
abstract:
1. introduction;
2., theory;
3. data;
4. measurement;
5. methodsandmodels;
6. results;
7. conclusion.
Introduction
Right at thebeginningof the introduction, the opening
foursentences attemptto grab our attention, to give a
clear indication ofwhere the article's focus lies, and to
provideanindicationoftheprobablesignificance ofthe
findings. Thisiswhattheauthorswrite:
Religionremains acentralelementofmodernlife,
shapingpeople'sworld-views,moralstandards,family
lives,and inmanynations,theirpolitics.Butinmany
Westernnations, modernizationand secularization
maybeerodingChristian beliefs,withprofound
consequencesthathave intrigued sociologistssince
Durkheim.Yetthismuchtoutedsecularizationmay
beoverstated-certainlyitvarieswidelyamong
nations and isabsentintheUnitedStates (Benson,
Donahue,and Erickson1989:154-7;Felling, Peters,
and Schreuder1991; Firebaughand Harley1991;
Starkand Iannaccone1994). Weexplorethedegreeto
whichreligiousbeliefsare passed onfromgeneration
togenerationindifferent nations. (Kelleyand De
Graaf1997:639)
This is an impressive start, because, in justover 100
words,theauthorssetoutwhatthearticleisaboutandits
significance.Letuslookatwhateachsentenceachieves.
Thefirstsentencelocatesthearticle'sresearchfocusas
addressinganimportantaspectofmodernsocietythat
touchesonmanypeople'slives.
The secondsentencenotesthatthereisvarietyamong
Westernnationsintheimportanceofreligionandthat
the variations may have 'profound consequences'.
But this sentencedoes more than the first sentence:
it also suggests that this is an area thathas been of
interestto sociologists. Tosupportthis point, one of
sociology'smostveneratedfigures-EmileDurkheim-
ismentioned.
The third sentence suggests that there is a problem
with the notionof secularization, which has been a
.._---------



I:
Iii
research focus for many sociologists of religion.
,
I
'
Several fairly recentarticles are cited to supportthe
authors' contention that there is a possibility that
secularization is being exaggerated by some com-
mentators. In this sentence, the authors are moving
towards a rationale for their article that is more in
terms ofsociologicalconcernsthanpointingtosocial
changes,which arethe mainconcernofthetwoopen-
ingsentences.
\'
Then in the fourth sentence the authorsset up their
specificcontributionto this area-theexploration of
thepassing-onofreligiousbeliefsbetweengenerations.
So,bytheend offoursentences,thecontributionthatthe
article isclaiming to make to ourunderstandingofreli-
gion in modemsocietyhas been outlined and situated
withinanestablishedliteratureonthe topic.Thisisquite
apowerfulstartto the article, becausethe readerknows
what the article is about and the particular case the
authorsaremakingfortheircontributiontotheliterature
onthe subject.
Theory
Inthissection, existingideasand researchonthe topicof
religioussocializationarepresented.Theauthorspointto
the impactofparentsand otherpeople onchildren'sreli-
gious beliefs, but then assert that 'a person's religious
environment is also shaped by factors otherthantheir
own and their parents' religious beliefs, and hence is
a potential cause of those beliefs ...' (Kelley and De
Graaf1997:641).Thissuggestionisthenjustified,which
prompts the authors to argue that 'prominent among
these"unchosen"aspectsofone's religious environment
isbirthplace'(1997: 641).Kelleyand DeGraaf'srumina-
tions on thisissuelead them toproposethe firstofthree
hypotheses,whichispresentedinResearch infocus 104.
This hypothesis stipulates thatcontextualfactors have
an impact on religious beliefs. Thisleads the authors to
suggestintworelatedhypothesesthat, inpredominantly
secularsocieties, family backgroundwillhave a greater
impacton a person'sreligious beliefs thaninpredomin-
antly devout societies, because in the former parents
and otherfamilymembersaremore likelytoseekto iso-
latechildrenfromsecularinfluences. However, indevout
societiesthisinsulationprocess islessnecessaryand the
influenceofnationalfactorswillbegreater.Thus,weend
up with very clearresearchquestions, which have been
arrivedat byreflecting onexisting ideas and researchin
thisarea.
Data
Inthis section,the authorsoutlinethe data theydrew
for theirresearch. This expositionentailsageneral On
~
line 0 f the datasets. Thequotationon page299is k
ta en
fromthiscommentary.Thesamplingproceduresar
eOUt-
linedalongwithsamplesizesandresponse rates.
Measurement
Inthissection, Kelleyand DeGraafexplainhowthern .
am
conceptsintheirresearchweremeasured.Theconce ts
were:religious belief (the questionnaireitemsusederein
Researchinfocus6.3);parents'churchattendance; secular
and religious nations (thatis, the scoring procedurefor
indicatingthe degreetowhich a nationwas religiousor
secular in orientation on a five-point scale); othercan.
textual characteristics of nations (forexample, whethera
formerCommunistnationornot); andindividual charac-
teristics (forexample,ageandgender).
Methodsand models
Thisisaverytechnicalsection,which outlines thediffer.
entways in which the relationships betweenthevari-
ables might be conceptualized and the implicationsof
using differentmutivariate analysis approaches forthe
ensuingfindings.
Results
The authors provide a general description of their
findings and then considerwhether the hypotheses are
supported. In fact, it turnsoutthe hypotheses are sup-
ported. The significance of other contextual character-
isticsofnationsand individualdifferencesareseparately
explored.
Conclusion
In this final section, Kelleyand De Graaf return to the
issues thathave beendrivingtheir investigation. These
are the issuesthey had presentedintheIntroductionand
Theory sections. They begin the section with a strong
statementoftheirfindings: 'Thereligiousenvironmentof
a nationhas amajorimpact onthe beliefsofitscitizens:
Peoplelivinginreligiousnationsacquire, inproportionto
the orthodoxyoftheirfellowcitizens, more orthodoxbeliefs
than those living in secular nations'(Kelleyand DeGraaf
1997: 654).They thenreflecton the implications ofthe
confirmationoftheirhypothesesforourunderstandingof
theprocessofreligioussocializationandreligiousbeliefs.
They also address the implications of theirfindings for
certai
whid
OUf
abo
the
(19
nat
me
wi'
thl
rei
T
the
of1
fac
rio:
ter
lev
on




rewon
.al OUt-
Staken
Ire OUt-
lemam
)ficepts
I arein
secular
ure for
iousor
er con-
ethera
:harac-
differ-
Le vari-
ionsof
for the
f their
sesare
re sup-
racter-
arately
to the
These
onand
strong
nentof
tizens:
1ionto
beliefs
Graaf
ofthe
dingof
oeliefs.
19S for
certaintheoriesaboutreligiousbeliefsinmodemsociety,
whichwere outlined intheir Theory section:
Ourresultsalsospeak tothe long-runningdebate
about USexceptionalism(Warner1993):Theysupport
theviewthattheUnited Statesisunusuallyreligious.
...Ourresultsdo notsupportStarkand Iannaccone's
(1994)'supply-side' analysisofdifferences between
nationswhichargues thatnationswithreligious
monopolieshavesubstantialunmetreligiousneeds,
while churches in religiouslycompetitive nations like
theUnited States doabetterjobofmeeting diverse
religiousneeds. (Kelleyand DeGraaf1997:655)
Thefinal paragraphspells outsome inferences about
thewaysinwhichsocialchangeshaveanimpactonlevels
ofreligious beliefin a nation. The authorssuggest that
factorssuchas modernizationarid the growthofeduca-
tion depress levels of religious beliefand that their impact
tendstoresultinaprecipitousratherthanagradualfallin
levelsofreligiosity. Intheirfinalthreesentences,theygo
ontowriteaboutsocietiesundergoingsuchchange:
Theoffspringofdevoutfamiliesmostlyremain
devout, buttheoffspringofmoresecularfamilies
nowstronglytendto besecular. Aself-reinforcing
spiralofsecularizationthensets in,shiftingthe
nation's average religiosityever further awayfrom
orthodoxy. Soafter generationsofstability,religious
beliefdeclines abruptly inthecourse ofafew
generationstothe modestlevelsseeninmany
Western nations. (Kelleyand DeGraaf1997:656)
Itmightbearguedthatthese reflectionsaresomewhat
risky, because the data from which the authors derive
theirfindingsarecross-sectionalinresearchdesignterms
ratherthanlongitudinal. They are clearly extrapolating
fromtheirscoringofthefifteen nationsintermsoflevels
of modernization to the impact of social changes on
national levels of religiosity. However, these final sen-
tences make for a strongconclusion,which itselfmight
formaspringboardforfurtherresearch.
Lessons
What lessons can be learnedfrom Kelleyand DeGraaf's
article? To some extent, these have beenalluded to in
the course of the above exposition, butthey are worth
spellingout.
Thereis aclearattempttograb the reader'sattention
with strong opening statements, which also act as
signpoststowhatthe article isabout.
The authors spell out clearly the rationale of their
research. This entails pointing to the continued sig-
nificanceofreligioninmanysocietiesandtothelitera-
tureon religious beliefsand secularization.
The research questions are spelled out in a very
specific way. Infact, the authorspresenthypotheses
that are a highly specific form of research question.
As notedin Chapter6, by no means all quantitative
researchisdrivenbyhypotheses,eventhoughoutlines
ofthe natureofquantitativeresearchoftenimplythat
itis.Nonetheless,Kelleyand DeGraafchosetoframe
theirresearchquestionsinthisform.
The natureofthe data,the measurementofconcepts,
thesampling,the researchmethodsemployedandthe
approachesto the analysisofthe dataare clearlyand
explicitlysummarizedinsections 3,4,and 5.
Thepresentationofthefindingsinsection6isoriented
very specifically to the research questions thatdrive
the research.
The conclusionreturnsto the researchquestionsand
spells out the implications of the findings for them
and for the theories examined in section 2. This is
an important element. It is easy to forget thatyou
shouldthinkofthe researchprocessasclosingacircle
in which you must return unambiguously to your
researchquestions. Thereisno point insertingextra-
neousfindings if they donot illuminateyourresearch
questions.Digressionsofthiskindcanbeconfusingto
readers,who mightbe inclined to wonderaboutthe
significanceoftheextraneousfindings.
We also see that there is a clear sequential process
moving from the formulation of the research questions
throughthe expositionofthe natureofthe dataand the
presentation of the findings to the conclusions. Each
stage islinkedtoand followsonfromitspredecessor(but
see Thinkingdeeply27.1).The structureused by Kelley
and De Graafis based on a commonone employed in
the writing-up of quantitative research for academic
journalsinthesocialsciences.Sometimesthereisasepar-
ateDiscussion sectionthatappearsbetweenthe Results
and the Conclusion. Anothervariation is that issues of
measurementandanalysis appearinthe same sectionas
the onedealingwith researchmethods,butperhapswith
distinctsubheadings.



Thinkingdeeply27.1
Anempiricistrepertoire?
At thispoint, it isworthrecalling the discussion inChapter20 of Gilbert and Mulkay's (1984) research on
scientists. Theauthors drewadistinction between anempiricist repertoire andacontingent repertoire. The
former derived from 'the observation that the texts of experimental papersdisplay certain recurrent stylistic
and lexical featureswhich appear to be coherentlyrelated'(Gilbert and Mulkay 1984: 55-6).We
shouldbear inmindthat the sameis true of papers writtenforsocial sciencejournals.Thesetoodisplay Certain
features thatsuggest a certaininevitability to the outcomeof the research.In other words, the reader is given
a sensethat. infollowing the rigorous proceduresoutlinedinthearticle, the researcherslogically arrived at
their conclusions. The contingent repertoire, with its recognition ofthe roleof the researcher intheProduction
offindings, isfar lessapparent inscientists' publishedwork. Thus, we havetorecognize the possibility that
the impression ofa seriesoflinked stagesleading to an inescapableculminationis to a largeextent a
..
reconstructionofeventsdesignedto persuade referees(who,of course,usethe sametacticsthemselves)
ofthecredibility and importance of one'sfindings. This means that the conventions aboutwriting upa
quantitative research project, some of which are outlined inthischapter, are inmany waysal1 invitation to
reconstruct aninvestigation ina particularway. The wholeissueof theways inwhich the writing-up of research
representsa meansof persuadingothers ofthecredibility of one'sknowledgeclaimshas beenaparticular
preoccupationamongqualitativeresearchers(seebelow)and has beengreatly influencedbythesurge of
interest inpostmodernism. However,inThinkingdeeply27.2, some oftherhetorical strategiesinvolvedin
writing up quantitativesocial researchare outlined. Three pointsare worthmaking about thesestrategiesin
the present context. First. theyare characteristic of the empiricist repertoire.Second, while the writing of
qualitative research hasbeen a particularfocus inrecent times(seebelow), someattentionhasalsobeenpaid
to quantitativeresearch.Third. whenIcomparedthewriting of quantitativeandqualitativeresearcharticles,
Ifound theywere not asdissimilar intermsofrhetorical strategiesas issometimesproposed(Bryman 1998).
However, Ididfind greater evidenceofa managementmetaphor (seeThinking deeply 27.2), which isalso
evident inKelley and DeGraaf'sarticle; for example. 'we excluded the deviantcasesfrom ouranalysis'(1997:
646) and 'wedivided the nationsintofive groups' (1997: 647).
1. introduction;
Writingupqualitativeresearch
2. theanalysisofthe socialdimensions offoodandeating;
Nowwe will look atan exampleof a journal articlebased
3. studiesofvegetarianism;
on qualitative research. Again,Iam not suggesting that
the article is exemplary or representative, but that it 4. thedesignofthestudy;
exhibitssome features that are often regarded as desir-
5. thefindingsofthestudy;
ablequalitiesintermsofpresentationandstructure.The
6. explainingcontemporaryvegetarianism;
articleisonethat hasbeenreferredtoinseveralprevious
7. conclusions.
chapters (especially Research in focus 2.10, 18.2, and
18.8):astudyofvegetarianism byBeardsworthand KeiI
What isimmediatelystrikingabout thestructure isthat
(1992).Thestudyisbasedonsemi-structured interviews
it is not dissimilar to Kelleyand De Graaf's (1997).
and was published in the Sociological Review, a leading
Nor shouldthis be all thatsurprising.Afterall, a structure
Britishjournal.
that runs
Structure
Introduction Literaturereview Researchdesign/
Thestructurerunsasfollows: methods Results Discussion Conclusions
isno
rathE
res
e2
thee
WOVI
thee
and
tenT
Intl'l
The
whz
n
to
w
a'
VI
a
n
il
1
t

on
The
ylistic.
-6). We
Iy certain "
isgiven
ecat
oduction
that
-es)
Into
f research
ular
:e of
'd in
gies in
: of
~ n paid
rtlcles,
1998).
also
;'(1997:
tdeating;
reisthat
(1997).
structure
I design!
ins
isnot obviously associated with one researchstrategy
rather thanthe other. One differencefrom quantitative
researcharticlesisthatthepresentationoftheresultsand
thediscussionofthemare frequentlyrathermore inter-
woveninqualitativeresearcharticles. Wewill seethisin
thecaseofBeardsworthandKeil'sarticle.Aswith Kelley
and De Graaf's article, we will examine the writing in
termsofthe article'sstructure.
Introduction
Thefirst four sentences give us an immediate sense of
whatthe article isaboutand whereitsfocuslies:
The purposeof thispaperistooffer acontribution
totheanalysisoftheculturalandsociologicalfactors
whichinfluence patternsoffoodselection andfood
avoidance. Thespecificfocusiscontemporary
vegetarianism.acomplexofinter-relatedbeliefs,
attitudesand nutritional practices whichhastodate
receivedcomparativelylittleattentionfromsocial
scientists.Vegetariansinwesterncultures, inmost
instances,are notlife-longpractitionersbutconverts.
Theyare individualswhohavesubjectedmore
traditional foodwaysto criticalscrutiny,and
subsequentlymade adeliberatedecisiontochange
their eating habits,sometimesinaradicalfashion.
(Beardsworthand Keil1992:253)
LikeKelleyand DeGraaf's,this isa strongintroduction.
Wecanlookagain atwhat each sentenceachieves.
Thefirstsentencemakesclearthattheresearchiscon-
cernedwithissuestodowiththe studyoffood.
The secondsentenceprovidesuswith the specificre-
searchfocus-thestudyofvegetarianism-andmakes
a claim for our attentionby suggestingthatthis isa
topic thathas beenunder-researchedbysociologists.
Interestingly,this isalmostthe oppositeof the claim
made by Kelleyand De Graaf in their second sentence,
in thatthey pointto a line ofsociological interestin
religiongoing back toDurkheim. Eachisalegitimate
textualstrategyforgainingthe attentionofreaders.
Our attention is jolted even more by an interesting
assertionthatbeginstodrawthereaderintooneofthe
article's primarythemes-theidea of vegetarians as
converts.
Thefourthsentenceelaboratesuponthe ideaofvege-
tarianismas beingformost peoplean issue ofchoice
ratherthanatraditionintowhichone isborn.
Thus,afteraround100words, thereaderhasaclearidea
ofthe focusofthe researchand hasbeenledtoanticipate
t ~ t there is unlikely to be a greatdealof pre-existing
socialresearchonthisissue.
The analysisofthesocial dimensionsoffood
andeating
This and the next section review existing theory and
researchinthis area.Inthis section,the contributionsof
varioussocialscientiststo socialaspects offood and eat-
ingare discussed.Theliteraturereviewedactsasaback-
clothtothe issueofvegetarianism.Beardsworthand Keil
(1992: 255)proposethattheirreview ofexistingtheory
andresearchsuggeststhat'thereexistsarange oftheor-
eticalandempirical resources which can be broughtto
bearupontheissueofcomemporaryvegetarianlsm'.This
pointisimportant,as the authorsnote onceagainat the
endofthesectionthatvegetarianismhas receivedlittle
attentionfrom socialscientists.
Studiesofvegetarianism
Thissectionexaminesaspectsoftheliteratureonvegetar-
ianismthathasbeencarriedoutbysocialscientistsorthat
hasasocialscientificangle. Thereviewincludes:opinion
pollandsurveydata,which point tothelikelypercentage
of vegetarians in the British population; debates about
animalrights; sociological analysis of vegetarianideas;
andone study(Dwyer etal. 1974)ofvegetariansin the
USAcarriedoutby ateamofsocialscientists using sur-
vey research. In the final paragraphofthis section, the
authorsindicatethecontributionofsomeoftheliterature
theyhave covered.
The designofthestudy
Thefirstsentenceofthissectionforges ausefullinkwith
theprecedingone: 'Thethemesoutlinedabove appearto
warrant further investigation, preferably in a manner
which allows for a much more richly detailed examina-
tionof motivations andexperiences thanisapparentin
the studyby Dwyer et al.' (BeardsworthandKeil1992:
260).Thisopeninggambitallows the authorstosuggest
thatthe literatureinthis areaisscantand thatthereare
many unanswered questions. Also,they distance them-
selves from the one sociological study of vegetarians,
whichinturnleadsthemtosetupthe groundsforprefer-
ringqualitativeresearch.Theauthorsthenoutline:
whowastobestudiedandwhy;
how respondents were recruited (see Research in
focus18.8) and thedifficultiesencountered;



,i
the semi-structured interviewing approach (see
Researchinfocus18.2)and therationaleforit;
the numberof people interviewedand the contextin
which theinterviewstookplace;
the approach to analysing the interview transcripts,
whichlargelycomprisedtheidentificationofthemes.
Thefindings ofthestudy
Thechieffindings are outlinedunderseparateheadings:
respondents'characteristics;types ofvegetarianism;the
process of conversion; motivations; nutritional beliefs;
social relations; and dilemmas. The presentation of the
results iscarriedout so thatthereissome discussion of
theirmeaningorsignificanceinsuchawayastoleadonto
the next section, which provides exclusivelyadiscussion
ofthem. Forexample,inthefinalsentenceinthesection
reporting findings relating to nutritional beliefs, the
authorswrite:
Justasmeattendedtoimplystronglynegative
connotationsforrespondents,conceptslike'fruit'
and 'vegetable'tendedtoelicitpositivereactions,
although lessfrequentlyand inamoremuted form
than mighthavebeenanticipatedonthe basis
ofthe analysisoftheideologicalunderpinningsof
'wholefoods' consumptionputforwardbyAtkinson
(1980,1983),oronthebasisofthe analysisof
vegetarianfoodsymbolismadvancedbyTwigg
(1983: 28).(Beardsworth and Keil1992:276)
Inthisway,the presentationoftheresults ispointingfor-
ward tosome themesthatare takenup inthe following
sections and demonstrates the significance of certain
findingsforsomeofthepreviouslydiscussedliterature.
Explainingcontemporaryvegetarianism
This section discusses the findings in the light of the
study'sresearchquestionsinconnectionwithfoodselec-
tionand avoidance.The results are also relatedtomany
oftheideasencounteredinthe twosectionsdealingwith
the literature. The authors develop an idea emerging
from their research, which they call'food ambivalence'.
This concept encapsulates for the authors the anxieties
and paradoxesconcerningfoodthatcan bediscernedin
the interviewtranscripts (forexample,foodcan be con-
strued both as necessary for strength and energy and
simultaneouslyasasourceofillness).Vegetarianismisin
many respects a response to the dilemmas associated
withfoodambivalence.
Conclusions
In,thissection, the authors return to manyoftheid
leas
andthemesthat have driven theirresearch. Theyspell
thesignificance oftheideaoffoodambivalence,
probablythearticle'smaincontributiontoresearchin
area.Thefinalparagraphoutlinestheimportanceof
ambivalenceforvegetarians,but the authors arecareful
nottoimplythatitisthe salereason fortheadoptionof
_ vegetarianism.Inthefinalsentencetheywrite:'However
for a significantsegmentofthe population [vegetarian:
ism] appearsto represent a viabledevicefor re-establishin
somedegreeofpeaceofmindwhen contemplatingsam:
of the darker implications of the carefully arranged
message on the dinner plate' (Beardsworth and Keil
1992:290).Thissentenceneatlyencapsulatesoneofthe
"
<
article's masterthemes-theidea ofvegetarianismasa
responseto food ambivalence-andalludes throughthe
referenceto'the carefullyarrangedmessage'tosemiotic
analysesofmeatandfood.
lessons
Aswith KelleyandDe Graaf'sarticle, it is useful toreview
some of the lessons learned from this examinationof
BeardsworthandKeil'sarticle.
Justlikethe illustrationofquantitativeresearchwrit-
ing,therearestrongopeningsentences,whichattract
our attentionand giveaclearindication ofthenature
and contentofthearticle.
Therationaleoftheresearchisclearlyidentified.Toa
largeextent,thisrevolvesaroundidentifyingthesoci-
ologicalstudyoffoodandeatingasagrowingareaof
research but noting the paucity of investigationsof
vegetarianism.
Research questions are specifiedbut they are some-
whatmoreopen-endedthaninKelleyand DeGraaf's
article, which isin keeping with the general orienta-
tionofqualitativeresearchers. Theresearchquestions
revolvearoundtheissueof vegetarianismasadietary
choiceandthemotivationsforthatchoice.
Theresearchdesign and methods areoutlinedandan
indication is given of theapproach to analysis. Thesec-
tioninwhich these issuesarediscusseddemonstrates
greatertransparencythan issometimes the casewith
articlesreportingqualitativeresearch.
The presentation and discussion of the findingsin
sections 5 and 6 are geared to the broad research
questions that motivated the researchers' interestin
vegetarianism.However,section6alsorepresents the

v
p
IT
n
p
o
1<
o
a
v
a




theideas
fspellout
, whichis
rchinthis
ceoffood
re careful
[optionof
However,

tablishing
tingsome
arranged
and Keil
)neofthe
nismasa
roughthe
) semiotic
toreview
nation of
archwrit-
chattract
henature
ified.Toa
thesod-
tgareaof
of
ire some-
)e Graaf's
Jorienta-
questions
; adietary
edandan
s,Thesec-
lonstrates
casewith
ndings in
research
nterestin
ssentsthe
major opportunityfor the idea of food ambivalence
and its dimensions to be articulated. The inductive
natureofqualitativeresearchmeansthattheconcepts
and theoriesthatare generatedfrom an investigation
must be clearly identified and discussed, as in this
case.
The conclusionelucidatesin a morespecific way the
significanceofthe resultsforthe researchquestions.It
alsoexplorestheimplicationsoffoodambivalencefor
vegetarians,sothat' oneofthe article'smajortheoret-
icalcontributionsisclearlyidentifiedandemphasized.
Writing up mixed methods research
partly because interest in and the practice of mixed
methodsresearchhasgainedmomentumonlyinrelatively
recent times, ithasfew ifanywritingconventions.More
particularly, it is difficult to say what an exemplary
ormodel mixedmethodsresearchjournalarticlemight
looklike.Toacertainextent,itisboundtoborrowsome
of theconventions associatedwith writing upquantitative
andqualitativeresearchintermsofneedingto startout
witha researchfocus in the senseofa researchproblem
and!orsomeresearchquestions.CreswellandTashakkori
(2007: 108),the editorsoftheJournalofMixed Methods
Research, have suggested that 'good original/empirical
mixedmethodsarticles'shouldbe:
'well-developed in both quantitative and qualitative
components' (Creswell and Tashakkori 2007: 108);
and
'morethanreportingtwo distinct"strands"ofquanti-
tativeandqualitativeresearch;thesestudiesmustalso
integrate,link,or connectthese"strands"insomeway'
(CreswellandTashakkori2007:108).
Theyactuallyadd athirdfeatureofgood mixedmethods
articles-namely,thattheycontributeto theliteratureon
mixedmethodsresearchinsomeway.Thisseemsarather
tall order for many writers and researchers, so that I
wouldtendtoemphasizethe othertwo features.
The first impliesthatthe quantitativeandthe qualita-
tive components of a mixed methods article should be
at the very leastcompetentlyexecuted. This meansthat
intermsofthe fundamentalcriteriaforconductinggood
quantitativeandgoodqualitativeresearch,mixedmeth-
ods research should conform to both quantitative and
qualitativeresearchcriteria.Intermsofwriting,itmeans
that,foreachofthecomponents,itshouldbe clearwhat
theresearchquestionswere,howthesamplingwasdone,
whatthedatacollection technique(s) was or were, and
qowthedatawereanalysed.
Thesecondfeatureimpliesthatagoodmixedmethods
articlewill be morethanthe sumofits parts.This issue
relates to a tendency that has beenidentified by some
writers (e.g. Bryman2007c; O'Cathain et al. 2007) for
somemixedmethodsresearchersnottomakethebestuse
oftheir quantitative and qualitative data, in that they
often do not link the two sets of findings so that they
extractthe maximumyield from theirstudy.AsCreswell
andTashakkori(2007:108)putit:
The expectationisthat,bythe endofthe manuscript,
conclusionsgleanedfrom the twostrandsare
integratedto provideafullerunderstandingofthe
phenomenonunderstudy. Integrationmight be in
theformofcomparing,contrasting.buildlng on, or
embeddingonetypeofconclusionwiththeother.
Tosomeextent,whenwritingupthe resultsfromamixed
methods study, researchers might make it easier for
themselves to getacrossthe extrayield associatedwith
theirinvestigationsiftheymakecleartheirrationalesfor
includingbothquantitativeandqualitativecomponents
in theiroverallresearchstrategy.Theissue ofrationales
for conductingmixedmethodsresearchisonethatwas
addressedinChapter25.
Furtheradviceon writingup mixed methodsresearch
canbe foundinsuggestionsinCreswell andPlanoClark's
(2007: 161)delineationofastructurefor amixedmeth-
odsjournalarticle.Theysuggestthatthestructureshould
bealongthe followinglines.
Introduction. This would include suchfeatures as: a
statementofthe researchproblemor issue; anexam-
ination of the literature on the problem/issue; an
examinationofthe problemswiththe priorliterature,
whichmightincludeindicatingwhyamixedmethods
approachwouldbe beneficialperhapsbecausemuch
of the previous research is based mainlyonjust quanti-
tativeorqualitativeresearch;andthespecificresearch
questions.
Methods.Thiswouldincludesuchfeaturesas:indicat-
ingtherationaleforthe mixedmethodsapproach;the
type of mixed methods design (see e.g. Morgan's
classification of approaches to mixed methods re-
searchin Thinkingdeeply25.3);datacollectionand
data analysis methods; and indications of how the
qualityofthedatacanbejudged.



I I
I, I
,I
Tipsandskills
I havenoticedthatsomestudents ~ conduct mixed methods investigationstreat their quantitative and
qualitative findingsas separate domains, so thattheypresent oneset and then the other. In.PhD theses and
Masters dissertations, this can take the formof separate chapters labelled something like'surveyfindings' and
'qualitativeinterviewfindings'. This maynotbeaproblemifthetwo(or more) setsof findings are then
integrated in the Discussionsections or chapters. However, treating findingsinthis way does tend to
encourage a viewof the quantitative and the qualitative findingsas separate spheres and maytherefore
militate against integration, which, as writers likeCreswell andTashakkori (2007)imply,is increasingly an
expectation in mixed methods studies. Instead, try to think ofthe quantitative aridthe qualitativefindings
thematicallyacrossthetwosetsofresults,sothatthefindingsarepresentedintermsofsubstantiveissues
ratherthan intermsofdifferentmethods.
Results. Thequantitativeandthequalitativefindings
mightbepresentedeitherin tandemor sequentially,
but,ifthe latter,theywouldneedto bemergedinthe
Discussion.
Discussion. Summarizeandexplainresults,emphasiz-
ingthesignificanceofthemixedmethodsnatureofthe
researchand whatisgainedfrom thepresenceofboth
quantitativeand qualitative findings; drawattention
to anylimitations of the investigation; and possibly
suggestavenuesforfutureresearch.
In terms of the overall structure, Creswell and Plano
Clark's (2007) suggestionsaremoreor less the sameas
for an article based on quantitative research or anart-
iclebasedonqualitativeresearch(seeabove).Itisinthe
needtooudinethe mixedmethodsnatureoftheresearch
andtobringthetwosetsoffindingstogetherthatthedis-
tinctiveness of a mixed methods journalarticle can be
discerned.
Anexample ofmixedmethodswriting
Many of thesefeatures can be seen in thestudyof the
food-and-mouth crisis by Poortinga et al. (2004). This
article has beenpreviouslyencountered in Research in
focus 1.8and25.3.Itmaybeworthlookingbackat these
two accounts as a reminderof thestudy. The following
examinationof thewritingof this articleisorganizedin
termsofitsstructure.
Introduction
The article begins with a very strong and clear state-
mentof the focus of the article and its methodological
leanings:
Thirtyyears ofempirical workonpublicperceptions
have generatedanimpressivebodyoffindingson
attitudestotheconsequences,benefitsand
institutionalprofilesofarangeofimportantriskissues
...However,much oftheavailable researchtends to
have beenconductedwhenthe riskissuesstudied are
not particularlysalientinpublicdebate.Although
thereissomeevidencefrom opinion polling,risk
perceptionstudiesare rarelyconductedduringa
major riskcrisis.The presentstudyexaminespublic
attitudesto riskanditsmanagementduring one such
crisis:the2001Footand Mouth Disease (FMD)
epidemicinBritain.Amixed methodstudy designwas
employed,specificallyaquantitativesurvey
conductedattheheightoftheepidemicfollowed up
byqualitativefocus groups comprising individualswho
had participatedinthesurvey. Recentstudieshave
shownthatcombiningdifferentresearch methodscan
provideamore comprehensiveviewon riskissues
thancananyonemethodologyalone ...(Poortinga
et01.2004: 73-4)

TJ
c1
T
T
e:
cl
c;
F
1
e
a
c
E




gs'and
e
an
19s
res
ir state-
ological
ions
issues
Idsto
edare
f
[
blic f
such I
f
Inwas I
dup t
liswho '.
we .
cis can t
es I
nga
nd
esanc
Thisopeningpassageaccomplishesthefollowing:
It locates the study immediatelyin the literature on risk.
Itprovides ajustificationforconductingthe studyat
thetimeofthe FMDcrisis.
Itidentifies itselfas a mixed methodsstudyand pro-
videsarationaleforamixedmethodsapproach.
Theauthorsthengoontooutlinethestructureofthearti-
clesothat thereader.has a route map for what isto come.
TheBritish2001FootandMouthCrisis
Theauthorsoutlinetheoriginsofthecrisis,itstiming, its
extent,and itseffects.As aresult, thereaderisleftwitha
clear understanding of thenatureof theFMDcrisis.
Governmentpolicy,trust,and public reactionsto the
FMDepidemic
Thissection provides ajustificationfor the researchers'
emphasis on the significance oftrustinthe government
anditspolicies and draws attentionto relatedliterature
onthetopic.Forexample,theauthorsdrawattentiontoa
studyoftrustinrelationtoanotherfood-relatedcrisisin
Britain,theBSEcrisis:
Losingtrust, asoccurred totheBritishgovernment
over theBSE (madcow)crisis in the mid-1990s, may
havefar-reachingconsequences(Slavic, 1993), as
peoplebecomesuspiciousaboutnewgovernment
policyinterpreted inthelight of earlier experiences,
perhapsturningelsewhereforinformationandadvice.
So, it isvitallyimportant tohave somegauge of public
response. NotonlyregardingperceptionsoftheFMD
crisisasaneventwithinsociety,butalsoasatestcase
of the impacts of government policyand industry
responsiveness intheUK inthewakeofthe BSEcrisis.
(Poortingaet 01. 2004:75)
Theythenoutlinethenatureoftheirstudyinbroad-brush
terms,pointingout thatitcomprisedasurvey and focus
groups.Theauthorsexplainthattheyemphasizedintheir
researchfour aspects ofFMDand its management (see
below) and that theywere also keen to examine how
perceptions of them differed between the two commu-
nities(seeResearchinfocus1.8).
Methodology
The discussion of the research design and research
methodsisdivided intothreesections.
1. Study locations. The two communities-Bude and
Norwich-are examined, along with a justification for
u ~ i n thesetwo communities,whenthey write thatthey
wanted 'to find outmore about differences in attitudes
betweencommunitiesthatweredifferentiallyaffectedby
theepidemic'(Poortingaetal. 2004:75).
2. The questionnaire survey. The authors explain how
andwhen the questionnaires weredistributed in Bude
andNorwich. They outlinethe kinds and formatsofthe
questions that were asked. They provide the response
ratesforthe twosurveys and examinethe comparability
oftheensuingsamples.
3. Focus groups. Theauthorsexplainthatthefocusgroup
participantswereselectedfromthe questionnairesurvey
samples.They providedataon the numbersofparticip-
ants and offocusgroups, whentheytookplace,and how
longthesessionslasted.Thetopicsfordiscussionarealso
summarized.
Results
The findings areorganizedinto fournumberedsections,
each ofwhich dealswithone ofthe four aspects ofFMD
and its management that were indicated earlierin the
article: public risk perceptions of FMD;blame; govern-
menthandlingoftheFMDcrisis:andtrustininformation
aboutFMD.Itisverystrikingthat,whenpresentingdata
for each of the four aspects ofFMDthey explored, the
authorspresentboththequantitativeand thequalitative
findings, examininghow the two interrelate. Forexam-
ple,whendiscussingthe firstofthefour aspects-public
riskperceptionsofFMD-theybeginbypresentingsome
questionnairedataaboutrespondents'levels ofconcern
aboutFMD. These questionnairedataderivefromLikert
items that asked aboutlevels of agreementwith state-
mentslike'MymainconcernsaboutFMDare to dowith
the possible impacts on the health and welfare of ani-
mals'.A tableispresentedshowingmeanlevelsofagree-
mentwiththis and fiveotheritems, withthe databeing
presentedfor the whole sample, aswellas forBudeand
Norwich separately. They thenpresent thefocus group
findings, noting that the 'findings of the focus groups
reinforce those of the questionnaire regarding general
concern' (Poortingaetal. 2004: 78). The focus groups
found thatparticipantsweredeeplyconcernedaboutthe
slaughter of animals and the rotting carcases, whereas
thequestionnairesdidnotpickupthispoint.Thepossible
healtheffectsofthese ratherthanofthediseaseitselfwas
a concern (the surveyandthe focus group results both
~ _.



suggestthattherewas a low level ofconcernsaboutthe
directhealtheffectsofFMD).
Discussion
The Discussionsectionbeginsby outliningthe rationale
forthe mixed methodsstudyandwhathasbeengleaned
from it:
Theaimofthis-mixedmethodologystudywasto
investigatepublic reactionsto theFMDepidemic,
supportforgovernmentpoliciesto get FMDunder
control,and trustininformationaboutFMD. More
specifically,aquantitativesurveyand qualitativefocus
groupswereconductedto examinehowtwo separate
communitiesthatwereaffected-todifferentdegrees
bytheepidemicrespondedto the crisis. Inthisstudy,
thefocusgroupsweremainlyused to illustratethe
findingsofthequestionnaire.Thefocus groups
providedvaluableadditionalinformation,especially
onthereasons,rationalizationsand arguments
behindpeople'sunderstandingoftheFMDissue.
(Poortingaet al. 2004:86)
Thus ,the authors restate the mixed methods nature of
the investigationand the rationalefor the differentcom-
ponents.Theythenproceedtoprovideadetailedsummary
of the mainfindings. This accountof the key findings is
set in the contextofothercrises,like the BSEcrisis, and
existing literature on crisis management. They reflect
in some detail on the differences between Bude and
Norwich.The finalparagraphprovidesaverystrongcon-
cludingstatement:
Inconclusion,the combinationofaquestionnaire
surveyand afocusgroupstudygave acomprehensive
viewon people'sperceptionsand responsesto the
2001FMDepidemic.Theuniqueaspectofthisstudy
isthatithascapturedperceptionsduring theFMD
crisis.Although itonlygivesasnapshotofpublic
attitudesto riskand itsmanagement,itprovideda
vividpictureofpeople'sperceptionsand debates
on FMDattheheightoftheepidemic.Further
researchmayprovide insightinthedynamics and
thelong-term effects ofthedisease.Some studies
have shown thatriskperceptioncan berelatedto
theamountofpresscoveragethatisgivento that
particularrisk(Renn et al. 1992).Additionalstudies
may provideanswerson howarange ofdifferent
drive's,such asthemedia, policymeasures,and local
andindividualevents(see e.g, Pidgeon et 0/. 2003)
takeon variouslevelsofimportanceforpeople's
reactionto acrisissuch asFMD. Taken asaWhole,
this studysuggeststhatriskperceptionsofacriSisare
embeddedinbothlocaland nationalsocialcontexts.
(Poortingaet ai. 2004:89)
. -,
This final paragraph is significant and well crafted for
severalreasons:
The first sentencerestates the mixedmethodsnature
ofthestudyandthatitsprimaryrationalewastopro-
videa'comprehensive'overviewofthe topic.
The major contribution of the research-thatit was
conductedin the courseof the crisis-issuggestedto
thereaderinthe secondsentence. .
Thethirdsentenceprovidesabriefindicationofalim-
itationof the study('onlygives asnapshot') but then
invitesthereadernottodwellonthislimitationbysug-
gestingthatthe research'providedarichdescription'.
The next three sentences suggest future potentially
fruitfulavenuesforinquiry.
The finalsentenceprovidesafinalmessageforreaders
totakeawaywiththem,namely,that'riskperceptions
of a crisis are embedded in both local and national
social contexts'.
This isaverystrongfinal andconcludingparagraphthat
leaves readersin nodoubt about what theauthors believe
is the major contribution of their findings and which
reminds them of the significance of the fact thatitisa
mixedmethodsstudy.
Onefeatureofthis articlethatisquitestrikingisthatin
termsofstructureandoverallapproachitisquitesimilar
to the quantitative and the qualitative researcharticles
previouslyexamined.Indeed,itwas notedthatthe quali-
tativeresearcharticlewas notdissimilarto the quantita-
tive one. Inlargepart,thesesimilaritiescanbeattributed
to thefact thattherearegeneralconventionsabouthow
findings should be written up for academic audiences,
andtheseconventionsactasa templatefor, andto some
extentrestrict, muchacademicwriting. Whatis striking
aboutthe articlebyPoortingaet al. istheirinclinationto
makeasmuchofthe mixedmethodsstatusandcontextof
theirresearchas possible,asrecommendedinthe guide-
lines suggestedbyCreswellandPlanoClark (2007).
While an
researchis;
writers me
(Z007) am
I
Postn
diffio
a fon
ingtl
It qu
scien
Inste
cour



t
local
13)
le,
isare
exts,
fted for
i nature
; to pro-
t it was
ested to
ofa lim-
jut then
l bysug-
iption'.
:entially
readers
.eptions
rational
iph that
:believe
I which
It it is a
s that in
, similar
articles
Iequali-
uantita-
:ributed
iut how
Iiences,
to some
striking
arionto
ntext of
guide-
7).
While attention to the writing-up of mixed methods
researchis an area that is in its infancy, the suggestions of
writers mentioned above like Creswell and Tashakkori
(2007) and Creswell and Plano Clark (2007) along with
Studentexperience
Writer's block
Sometimeswhen writingwe feel as though the words willnot come out. RebeccaBarneswrites that, when
this happened to her, it usuallymeant that she needed to return to her data to workout what exactlyshe
was tryingto say.
There have been frustrating times when I have been unsure of what to writeand havespent manyhours
staringat a largely blankcomputer screen. I have now realizedthat when I experiencethis, it isusually
because I need to return to the data and spend more time planning what I want to say, how,and whyit
matters.
Isabella Robbins's response to similarproblems was to try to write every day:
Sometimes just gettingwordson the page isdifficult. Ihaveset myself the taskofwriting 1,000wordsa day,
no matter howincoherentthey are. I can usually achievethis. I have tried to put the thesis intothe realmof
'goodenough' and 'the last part of myresearchtraining' rather than it being'somethingexceptional'.
To readmoreabout Rebecca's and Isabella's research experiences, go to the Online Resource Centre that
accompanies thisbook at http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/.
Postmodernism and its implications

for writing
Postmodemism (see Key concept 27.2) is an extremely
difficult idea to pin down. In one sense, it can be seen as
a form of sensitivity-a way of seeing and understand-
ing that results in a questioning of the taken-for-granted.
It questions the very notion of the dispassionate social
scientist seeking to uncover a pre-given external reality.
Instead, postmodemists view the social scientist's ac-
count as only one among many ways of rendering social
strong exemplars like the article by Poortinga et al. pro-
vide helpful pointers to the ways in which this task should
be
reality to audiences. The social world itself is viewed as a
context out of which many accounts can be hewn. As a
result, 'knowledge' of the social world is relative; any
account is just one of many possible ways of rendering
social reality. As Rosenau (1992: 8) puts it, postmod-
ernists 'offer "readings" not "observations," "interpreta-
tions" not "findings" .. :.



Keyconcept27.2
Whatis postmodernism?
Asnotedin the maintext, postmodernism isextremely difficult to pin down. Part of the problem isthat, as an
approach, postmodernism isat leasttwo things. One isthat it isanattempt to get to gripswith the nature of
modernsociety andculture. Theother, which isthe more relevantaspectfor this book, isthat it represents a
wayof thinkingabout andrepresenting the nature of the social sciences andtheir claims to knowledge. In
particular:it isa distinctivesensitivity regardingthe representation of social scientific findings. Postmodernists
tend to bedeeplysuspicious of notionsthat imply that it is possible to arriveat a definitiveversion of any
reality. Reports of findings areviewedasversions of an external reality, sothat the keyissue becomes oneof
the plausibilityof thoseversions rather than whether they are right or wrongin anyabsolute sense. Typically,
writersof a postmodernist persuasion haveless to say about data-collectionissues than about thewritingand
representation of social science findings, though it isprobably the case that they aremoresympathetic to -?
qualitativethan quantitative research (Alvesson 2002). Indeed, postmodernist's have probably been most
influential in qualitativeresearch when discussing the nature of ethnographicaccounts andquestioning the
ethnographer'simplicit claimthat heor shehas provideda definitive account of a society. Thisthinking can
be discerned in VanMaanen's (1988) implicit critique of 'realist tales' ashecalledthem(Key concept 27.5),
Forpostrnodernists, there canbeno sense of anobjective reality out there waitingto be revealed to and
uncovered by social scientists. That reality isalways goingto be accessed through narratives in theformof
research reportsthat provide representations. With this shift in orientation came aninterestinthe language
employed in research reports, like written ethnographies, to revealthe devices researchers use to convey the
definitiveness of their findings(Delamont and Atkinson 2004). Postmodernists tendto emphasize thenotion
of reflexivity(see Keyconcept 27.4), which positsthe significance of the researcher for the research process
andconsequently the tentativeness of anyfindings presented in a research report (since the researcher is
always implicatedin hisor her findings), Asthis account of postmodernism implies, postmodernists tendto be
deeplysuspicious of anyviewof research that impliesthat there areor canbe accepted foundations to
knowledge, asis suggested by positivists (seeKey concept 1.2).Postrnodernisrn isa deeplydisruptive stance
on social research, in that it problernatizes and questions our capacityeverto know anything. Views vary on
postmodernism's current appeal. Matthewmanand Hoey (2006) depict its influenceas having waned to a
significant extent, while Bloland(2005) argues that it has hadan irnpacton thinking in manyfieldsin higher
education andthat this isespecially noticeableamongthosewho do not identify themselves aspostmodernists.
One of the effects of the impact of postmodernism since produced by ethnographers that have been a particular
the 1980s has been a growing interest in the writing of focus of attention, This focus has led to a particular inter-
social science. Forpostmodernists, reporting findings in a est in the claims to ethnographic authority that are
journal article provides merely one version of the social inscribed into ethnographic texts (Clifford 1983). The
reality that was investigated. Postmodernists mistrust the ethnographic text 'presumes a world out there (the real)
knowledge claims that are frequently boldly made when that can be captured by a "knowing" author through the
findings are reported and instead they adopt an attitude careful transcription and analysis of field materials
of investigating the bases and fOnTIS of those knowledge (interviews, notes, etc.)' (Denzin 1994: 296). Post-
claims and the language that is used to represent them. modernism problematizes such accounts and their
This has led to what is described as a linguistic tum authority to represent a reality because there 'can never
within the social sciences (Key concept 27.3). While the be a final, accurate representation of what was meant or
writing of all types of social science is potentially in the said, only different textual representations of different
postmodernist's firing line, it has been the kinds of text experiences' (Denzin 1994: 296).
Ho'"
attenti
sively
(1995
t r n ~
intere
astne
the Ii
and I
(e,g.
theSE
it]' c
and
~




t, asan
Jreof
;entsa
In
jemists
ny
)neof
pically,
ingand
:to
ost
*
g the
19can
:7.5).
td
n of
luage
ley the
lotion
rocess
ris
Id to be
tance
ry on
oa
gher
idernists,
articular
arinter-
that are
\3). The
:hereal)
iughthe
iaterials
). Post-
.d their
in never
neantor
lifferent
However, it would be wrong to depict the growing an evaluationof scientific and social scientific writing.
attention being focused on ethnographicwriting as exclu- Some illustrations of these analyses can be discerned in
.
sivelyaproductofpostrnodernism.Atkinsonand Coffey Thinkingdeeply27.1 and 27.2.Atkinsonand Coffeyalso
(1995) have argued that there are other intellectual pointrotheantipathywithinfeminismtowardstheimage
trends in the social sciences that have stimulated this oftheneutral'observer-author'whoassumesaprivileged
interest.Writersintheareaoftheoryandresearchknown stanceinrelationto members ofthe socialsettingbeing
asthe social studies of science have been concerned with studied. This stance is regarded as revealing a position of
thelimitations ofaccepteddistinctionsbetweenrhetoric dominationoftheobserver-authorovertheobservedthat
and logic and between the observer and the observed isinconsistentwiththegoalsoffeminism(seeChapter16
(e.g. Gilbert and Mulkay 1984).The problematizing of foranelaborationofthisgeneralpoint).Thisconcernhas
thesedistinctions,along with doubts aboutthe possibil- led to an interest inthe ways inwhich privilege iscon-
ity of a neutral language through which the natural veyedinethnographictextsand howvoices,particularly
andsocialworlds can be revealed, opened the door for ofmarginalgroups, aresuppressed.
Keyconcept27.3
t'.'./.
W
Whatisthelinguisticturn?
Postmodernismcanalsobe seen as the stimulus forthelinguistic turninthesocialsciences. Thelinguistic turn
isbasedontheideathatlanguageshapesourunderstandingoftheworld. Moreover,becauseknowledgeis
constructedthroughlanguage,andlanguagecan never create an objectiverepresentationof externalreality,
meaningisuncontrollable and undiscoverable.Thisleadsto a rejectionofpositivist scientists' claims tobe able
toproducereliableknowledge througha neutral processof exploration. Postmodernistsarguethat knowledge
isnever neutral and isconstantly open torevision. Theyreject what they seeasscientific 'grand' or 'meta'
narratives that seek toexplain the worldfrom an objectiveviewpoint. Scientific investigation isthus suggested
bypostmodernists tobe nothingmorethana type of 'languagegame'(Rorty1979) usedbythis particular
community to producelocalized understandings. Postmodernists have alsosuggestedthat certain methods can
bemoreeasilyadaptedtothelinguisticturn,suchasqualitativeresearchandinparticularethnography,
because it can be usedto deconstruct claims torepresentrealityand can providealternative versions ofreality
thatattemptto blurtheboundarybetween 'fact' and 'fiction'(Alvesson 2002). The impactofthelinguistic turn
can be seen inthegrowinguse ofvarious forms ofdiscourse analysis, which wascoveredinChapter20.
Discourse analysis has provedaparticularly useful approachfor unpacking the roleoflanguage inshaping
particularversionsofsocial reality.
Thinkingdeeply27.2
Rhetoricalstrategiesinwritingup
quantitativeresearch
The rhetoricalstrategiesused byquantitativeresearchers includethefollowing.
There isa tendency to removethe researcherfrom the text as an activeingredient ofthe researchprocess
inorder toconvey an impression ofthe objectivenature of the findings-thatis, as part of an external
realitythatisindependentoftheresearcher(Gusfield 1976).wootgar(1988) referstothisasan
externalizingdevice.



Theresearchersurfacesinthetextonlytodemonstratehisorheringenuityinovercomingobstacles
(Bazerman 1987;Bryman 1998).
Keyfigures in the field areroutinely cited to bestowcredibility ontheresearch (McCloskey 1985).
Theresearch process ispresented asa linear one to conveyanair of inevitabilityabout thefindings arrived
at(Gusfield 1976).
Relatiyely strict rulesarefollowed about what shouldbe reported in publishedresearch andhowitshould
bereported(Bazerman1987).
Theuijf ofamanagement metaphor iscommon in thepresentationof findingsin whichtheresearcher is
e p i t ~ asingeniously' "designing"research,"controlling"variables, "managing"data,and"generating"
tables' (Bryman1998: 146). See Shapiro (1985-6) andRichardson (1990) onthis point.
Notethat the first two points aresomewhat inconsistent.Thereissomeevidence that disciplineswithinthe
socialsciences differ in respectof theiruseof an impersonal styleof writing. But it maywell alsobethat it
sometimesdependson what the writeristrying todo; forexample,sometimesgettingacrossasense of one's
running.inovercomingpracticaldifficulties canbe just asuseful asgivingasense of the external natureof the,
findings.Therefore, sometimes the styleof presentation mayvarysomewhat.
Key concept 27.4
What is reflexivity?
Reflexivity hasseveral meanings in thesocial sciences. The term isemployedby ethnomethodologists torefer
to the wayin whichspeech andaction areconstitutive of thesocialworld in whichtheyarelocated;inother
words,they do morethan merelyact asindicators of deeper phenomena(seeChapter20).The other meaning
of the termcarries the connotation thatsocialresearchersshouldbe reflectiveabout the implications of their
methods,values,biases,anddecisionsfortheknowledgeofthesocial worldtheygenerate.Relatedly,
reflexivityentailsasensitivity to the researcher's cultural. political, andsocial context. AsSUCh. 'knowledge'
fromareflexivepositionisalways a reflection of a researcher's locationin time andsocialspace. Thisnotionis
especiallyexplicitinPink's(2001)formulationofareflexiveapproach totheuseofvisual images(seeChapter
17) andinPlummer's(2001)delineation of a reflexive approachtolifehistories(seeKeyconcept 18.1).
Therehasbeenevidenceof agrowing retlexivityinsocialresearchin the formof an industryofbooks that
collecttogether insidestories of theresearch process that detail the nutsandboltsofresearch asdistinctfrom
the oftensanitized portrayal inresearch articles.An earlyvolume edited byHammond (1964) paved the way
for alargenumber of imitators(e.g. Bell and Newby1977; Bell andRoberts 1984; Bryrnan 1988b). andthe
confessional tales referredto in Keyconcept 27.5are invariablymanifestations of this development.Therefore.
the rise of reflexivitylargelypredates the growingawareness of postmodernthinkingsince the late1980s.
What distinguishes the reflexivity thathasfollowedin thewakeof postmodernismisa greaterawareness and
acknowledgement of the role of the researcher aspart andparcel of the construction ofknowledge. Inother
words, the reflexiveattitude withinpostmodernismishighly critical of the notion that theresearcherIS
someone who extractsknowledgefromobservations andconversationswith othersandthentransmits
knowledge toanaudience.Theresearcheris viewed asimplicatedin the constructionofknowledgethrough
thestancethatheorsheassumesinrelationtotheobservedandthroughtheways inwhichanaccountis
transmittedin the formof atext. Thisunderstandingentailsanacknowledgementof the implicationsand
significanceofthe researcher'schoicesasbothobserverand writer.




fer
~
ling
~
1is
ter
)m
ly
ore,
Id
h
Theconcernswithintheseandothertraditions(includ-
ingpostrnodernism) haveled to experimentsin writing
ethnography(Richardson1994).An exampleis theuseof
a'dialogic' formof writing that seeks to raise the profileof
themultiplicityofvoicesthatcanbe heardin thecourse
offieldwork. AsLincolnand Denzin(1994: 584) putit:
'Slowlyitdawnson us thattheremay...be ...notone
"voice", butpolyvocality; notonestory, butmanytales,
dramas, pieces of fiction, fables, memories, histories,
autobiographies, poems, and othertexts to inform our
sense of lifeways, to extend our understandings ofthe
Other..:.
Manning(1995) cites, as an exampleofthe postrnod-
ern preference for allowinga varietyof voices to come
throughwithinan ethnographictext,theworkofStoller
(1989), who conducted research in Africa. Manning
(1995:260) describesthetextas'periodically'dialogicin
thatit is'shapedby interactionsbetweeninformantsor
"the other" and the observer'. This postrnodern prefer-
encefor seekingoutmultiplevoices andfor turningthe
ethnographer into a 'bit player' reflects the mistrust
amongpostrnodernistsof'meta-narratives'-that is,posi-
tionsorgrandaccountsthatimplicitlymakeclaimsabout
absolutetruthsandthatthereforeruleoutthepossibility
of alternative versions of reality. On the other hand,
'mini-narratives, micro-narratives, local narratives are
However, reflexivity is a notoriously slippery concept. Lynch(2000) has complained that too oftenit is assumed
that areflexive positionis somehowsuperiortoan unreflexiveone.Thecasefor the superiorityofreflexivityis
rarelymade. Moreover,hepointsoutthattheterm hasdifferentmeanings.Oneofthese ismethodological
reflexivity,whichcomes closesttothe kindofreflexivitythatisbeingreferredto inthischapter. However,this
meaninghasanumberofsub-meanings,threeofwhichare especiallyprominentinmethodologicalwritings.
1. Philosophicalself-reflection:anintrospection involving'an inward-looking, sometimes confessional and self-
critical examinationofone'sownbeliefs andassumptions'(Lynch2000: 29).
2. Methodologicalself-consciousness:takingaccountofone'srelationshipswiththosewhomone studies.
3. Methodologicalself-criticism:the confessionalstyle of ethnography (see Keyconcept27.5), butLynchnotes
thatthe injunctiontobeself-criticalthatis associated withsuchethnographicwritingismuchmore
pervasiveinacademicdisciplines,
Theterm 'reflexivity' hasto beusedwithadegreeofcaution,asLynch'sdiscussionimplies.
juststories thatmakeno truthclaimsandaretherefore
more acceptable to postmodernists' (Rosenau 1992:
p.xiii).
Postmodernism has also encouraged a growing
reflexivityin considerationsabouttheconductofsocial
research, and the growing interest in the writing of
ethnographyisverymucha manifestationofthis trend
(seeKeyconcept27.4).This reflexivitycan be discerned
in the wayin which manyethnographers have turned
inwards to examine the truth claims inscribed in their
ownclassictexts,whichisthefocus of the nextsection.
In the end,whatpostrnodernismleaves us withisan
acutesenseofuncertainty. It raisestheissue ofhowwe
caneverknoworcapturethe social realitythatbelongsto
othersandinsodoingitpointstoanunresolvabletension
thatwill notgo awayandthatisfurtherrevealedin the
issuesraisedinthenextsection,because,toquoteLincoln
andDenzin (1994: 582) again:'On theonehandthereis
theconcernforvalidity,orcertaintyinthetextasaform
ofisomorphism and authenticity. On the other hand,
thereisthesureandcertainknowledgethatall textsare
socially, historically, politically, and culturally located.
We,likethetextswewrite,canneverbe transcendent.'At
the same time,ofcourse,such a viewrenders problematic
the very idea ofwhat social scientificknowledge is or
comprises.

...:
Writing ethnography'
". .
The term 'ethnography', as noted in Chapter 17, is inter-
The ethnographic text is permeated by stylistic and
esting, because it refers both to a method of social
rhetorical devices whereby the reader is persuaded to
research and to the finished product of ethnographic enter into a shared framework of facts and interpreta_
research. In other words, it,is both something that is car- tions.observations and reflections. Just like the scientific
. ried outin doing research and something that one reads." paper and the kind of approach to writing found in repon.
Thus, writing seems to be at the heart of the ethnographic ing quantitative social research, the ethnographer typic.
enterprise. In recent years, the production of ethno- ally works within a writing strategy that is imbued with
graphic texts has become' a focus of interest in its own , realism. This simply means that the researcher presents
right. This means that there has been a growth of interest an authoritative, dispassionate account that represents
not just in how ethnography is carried out in the field but an external, objective reality. In this respect, there is very
also in the rhetorical conventions employed in the prb- ' little difference between the writing styles of quantitative '
duction of ethnographic texts. and qualitative researchers. Van Maanen (1988) calls
Ethnographic texts are designed to convince readers ethnography texts that conform to these characteristics
of the reality of the events and situations described, and realist tales. These are the common type of ethnographic
the plausibility of the analyst's explanations. The ethno- writing, though he distinguishes other types (see Key
graphic text must not simply present a set of findings: it concept 27.5). However, thefonn that this realism takes
must provide an 'authoritative' account of the group or differs. Van Maanen distinguishes four characteristics of
culture in question. In other words, the ethnographer realist tales: experiential authority; typical forms; the
must convince us that he or she has arrived at an account native's point of view; and interpretive omnipotence.
of social reality that has strong claims to truth.
Key concept 27.5
Three forms of ethnographic writing
Van Maanen (1988) has distinguished three major types of ethnographic writing,
1. Realist tales-apparently definitive,confident, and dispassionate third-person accounts of a culture and of
the behaviour of members of that culture. This isthe most prevalent formof ethnographic writing.
2. Confessional toles-personalized accounts in which the ethnographer isfully implicated inthe data-
gathering and writing-upprocesses. These are warts-and-all accounts of the trialsand tribulations of doing
ethnography. They have become more prominent since the 1970sand reflect a growingemphasis on
reflexivity in qualitative research in particular. Several of the sources referred to in Chapter 17are
confessional tales (e.g. Armstrong 1993; Hobbs 1993;Giulianotti 1995).However, confessional tales are
more concerned with detailing how research was carried out than with presenting findings. Veryoften the
confessional tale istold in one context (such as an invited chapter in a bookof similartales), but the main
findingsare written up in realist tale form,
3. Impressionist toles-accountsthat place a heavy emphasis on 'words, metaphors, phrasings, and, .. the
expansive recall of fieldworkexperience' (Van Maanen 1988: 102).There isa heavy emphasis on storiesof
dramatic events that provide 'a representational means of crackingopen the culture and the fieldworker's
way of knowing it' (Van Maanen 1988: 102). However, as Van Maanen (1988: 106)notes, impressionisttales
'are typicallyenclosed within realist, or perhaps more frequently, confessional tales'.
Just a
disap
told \
the 01
vides
/\Sa
prest
resec
acce
whil
pers
tiall'
the
nav.
is SL
thei
crec
and
aut
gra
I
wh
rna
the
ha
vic
Th
an
fiE
ef
qi
al
01
et
fc
d
t;
v
,




'c and
led to
preta.
'. mtific
~ P n
typic
With
sents
sents
:very
ative
calls
iStics
Iphic
Key
a1ces
cs of
the
Experiential authority
Just as in much quantitative research writing, the author
disappears from view when writing ethnography. We are
told what members of a group say and do, and they are
theonly people directlyvisible in the text. The author pro-
videsa narrative in which he or she is no longer to be seen.
As a result, an impression is conveyed that the findings
presented are what any reasonable, similarly placed
researcher would have found. As readers, we have to
accept that this is what the ethnographer saw and heard
while working as a participant observer or whatever. The
personal subjectivity of the author/ethnographer is essen-
tially played down by this strategy. The possibility that
the fieldworker may have his or her own biases or may
lihve become too involved with the people being studied
issuppressed. To this end, when writing up the results of
their ethnographic work, authors play up their academic
credentials and qualifications, their previous experience,
and so on. All this enhances the degree to which the
author's account can be relied upon. The author/ethno-
grapher can then appear as a reliable witness.
A further element of experiential authority is that,
when describing their methods, ethnographers invariably
make a great deal of the intensiveness of the research that
they carried out-theyspent so many months in the field,
had conversations and interviews with countless indi-
viduals, worked hard to establish rapport, and so on.
These features are also added to by drawing the reader's
attention to such hardships as the inconvenience of the
fieldwork-the danger, the poor food, the disruptive
effect on normal life, the feelings of isolation and loneli-
ness, and so on. Also worth mentioning are the extensive
quotations from conversations and interviews that invari-
ably form part of the ethnographic report. These are also
obviously important ingredients of the author's use of evid-
ence to support points. However, they are a mechanism
for establishing the credibility of the report in that they
demonstrate the author's ability to encourage people to
talk and so demonstrate that he or she achieved rapport
with them. The copious descriptive details-s-of places,
patterns of behaviour, contexts, and so on-can also be
viewed as a means of piling on the sense of the author
being an ideally placed witness for all the findings that
have been uncovered.
Typical forms
The author oftenwrites about typical forms of institutions
or of patterns of behaviour. What is happening here is
that the author is generalizing about a number of recur-
ring features of the group in question to create a typical
form tl1at that feature takes. He or she may use examples
based on particular incidents or people, but basically the
emphasis is upon the general. For example, in Taylor's
(1993) conclusion to her ethnographic research on female
drug users, which was cited several times in Chapter 17,
we encounter findings such as these: 'Yet the control exer-
cised over women through the threat to remove their chil-
dren highlights a major factor differentiating female and
male drug users. Unlike male drug users, female drug
users, like many other women, have two careers: one in
the public sphere and one in the private, domestic sphere'
(Taylor 1993: 154). This is meant to portray drug users in
general, so that individuals are important only in so far as
they represent such general tendencies.
The native's point of view
The point has been made several times that one of the dis-
tinguishing features of much qualitative research is the
commitment to seeing through the eyes of the people
being studied. This is an important feature for qualitative
researchers because it is part of a strategy of getting at the
meaning of social reality from the perspective of those
beingstudied. However, it also represents an important ele-
ment in creating a sense of authoritativeness on the part of
the ethnographer. After all, claiming that he or she takes
the native's point of view and sees through his or her eyes
means that he or she is in an excellent position to speak
authoritatively about the group in question. The very fact
that the ethnographer has taken the native's point of view
testifies to the fact that he or she is well placed to write
definitively about the group in question. Realist tales fre-
quently include numerous references to the steps taken
by the ethnographer to get close to the people studied and
his or her success in this regard. Thus, for her research on
female drug users, Taylor (1993: 16) writes:
Events I witnessed or took part in ranged from the
very routine (sitting around drinking coffee and eating
junk food) to accompanying various women on visits
to DSS [Department of Social Security] offices or to
the HIV clinic; I accompanied them when they were
in court, and even went flat-hunting with one woman.
I went shopping with some, helping them choose
clothes for their children and presents for their
friends. I visited them in their homes, rehabilitation
centres, and maternity wards, sat with them through




I
I
withdrawals, watched them using drugs, and
accompanied them when they went 'scoring'
(buying drugs). (Taylor 1993: 16)
Similarly, referring to his study of a factory in a small
Welsh community, Delbridge (1998: 19) writes:
I stood out a sore. thumb .. , My actual
participation fn the tasks which faced the workers
helped to break down the barriers and several people
approached me over the weeks and told me that
when they actually saw me sitting there alongside
them day after day they began to have some respect
for what I was doing. It was important to be able to
develop some shared ground.
He goes on to say:
the relationships developed over long hours working
on the shop floor, chatting over lunch, moaning about
the weather, and so on. Inthe close-knit village
community, Isoon got involved in long conversations
about families, mine and theirs, which was a most
unusual topic in the social world from which I had
come.... the common ground we found in our family
lives cemented relationships and founded them on
something other than a student/subject basis.
(Delbridge 1998: 20)
These passages are very effective in demonstratin
ho,,:, the ethnographer was able gradually to be tr g
formed from an outsider to an insider with similar ex
pen.
ences and concerns. As such, his credibility as som
eone
who can speak authoritatively about these wOrkers and
their lives is enhanced.
Interpretative omnipotence
When writing up an ethnography, the author rarely pre.
sents possible alternative interpretations of an event or
pattern of behaviour. Instead, the phenomenon in ques-
tion is presented as having a single meaning or signi-
ficance, which the fieldworker alone has cracked. Indeed
the provided is carefull marshalled to
the singular interpretation that is placed on the event
or pattern of behaviour. We are presented with an
inevitability. It seems obvious or inevitable that someone
would draw the inferences that the author has drawn
when faced with such clear-cut evidence.
These four characteristics of realist tales imply that
what the researcher did as a researcher is only one part
of creating a sense of having figured out the nature of
a culture. It is also very much to do with how the
researcher represents what he or she did through writing
about ethnography. For the postmodernist position, any
realist tale is merely one 'spin'-that is one version, that
can be or has been formulated in relation to the culture in
question.
Checklist
Issues to consider for writing up a piece of research
o Have you clearly specified your research questions?
o Have you clearly indicated how the literature you have read relates to your research questions?
o Isyour discussion of the literature critical and organized so that it is not just a summary of what you
have read?
o Have you clearly outlined your research design and your research methods, including:
o why you chose a particular research design?
o why you chose a particular research method?
o how you selected your research participants?






strating
etrans.
experj.
)meone
ersand
elypre-
'ventor
.n ques.
,rsign],
Indeed,
suppon
e event
vith an
)meone
drawn
ilythat
'nepart
iture of
ow the
writing
on,any
m,that
.lturein
-?
,.
t you
o whethertherewereanyissues to do with cooperation (e.g. response rates)?
o why youimplemented your research in,aparticularway(e.g. howthe interviewquestions
relateto your research questions, why youobserved participants in particular situations, why
your focus groupguide asked the questions in a particularwayand order)?
o if your research requiredaccess to anorganization, howandonwhat basis was agreement for
access forthcoming?
o steps youtook to ensure that your research wasethicallyresponsible;
o howyou analysed your data?
o anydifficulties youencountered in the implementation of your research approach.
oHave youpresented your datain amannerthat relates to your research questions?
oDoes your discussion of your findings showhowthey relateto your research questions?
oDoes your discussion of your findings showhowthey shed light on the literaturethat you presented?
:-:
o Are of the datathat you offerfully supported with tables,' figures, or segments ;
from transcripts?
o If youhavepresented tablesand/ or figures, arethey properlylabelled with a title andnumber?
oIf you have presented tables and/or figures, arethey commented uponin your discussion?
oDoyour conclusions clearlyallowthe readerto establish what your research contributes to the
literature?
o Have youexplained the limitationsof your study?
o Do yourconclusions consist solely of a summary of your findings? If they do, rewritethem!
o Do yourconclusions make clearthe answers to your research questions?
o Does your presentation of the findings andthe discussion allowaclearargument andnarrative to be
presented to the reader?
oHaveyoubrokenupthe text in each chapterwith appropriate subheadings?
o Does your writingavoidsexist, racist. anddisablist language?
o Have youincluded all appendices that you might needto provide(e.g. interviewschedule, letters
requesting access, communications with research participants)?
o Have youchecked that your list of references includes all the itemsreferred to in your text?
oHaveyouchecked that your list of references followsprecisely the stylethat your institution
requires?
oHave youfollowedyour supervisor's suggestions whenheor shehas commented onyour draft
chapters?
oHaveyou got peopleother thanyour supervisor to readyour draft chapters for you?
o Haveyouchecked to ensure that there isnot excessive use of jargon?
o Doyouprovideclearsignposts in the course of writing, sothat readers areclearabout what to
expectnext andwhy it isthere?
oHave youensured that your institution'srequirements for submittingprojects arefullymet in terms
of such issues asword length(sothat it isneithertoo longnor too short)andwhetheran abstract and
tableof contents arerequired?



o Haveyouensured that you do not quoteexcessively whenpresenting the literature?
o Haveyoufully acknowledged the work of otherssothat youcannot beaccused of plagiarism?
o Isthere a goodcorrespondencebetweenthe titleofyour project anditscontents?
o Haveyouacknowledged the helpof otherswherethis isappropriate(e.g, your supervisor, people
who mayhavehelpedwithinterviews,peoplewho readyourdrafts)?
. .' .
Key points
Good writing isprobably just asimportant asgoodresearch practice. Indeed, it isprobablybetter
thoughtofasapartofgoodresearchpractice.
,
Clear structure and statement of yourresearch ",uestions are important components of writingup
research.
Besensitiveto the waysin which writers seektopersuade usoftheir points of view.
The study of rhetoric andwriting strategies generallyteaches usthat the writings ofscientists and
social scientistsdo more than simply reportfindings. They aredesigned toconvinceandto
persuade.
Theemphasisonrhetoricisnotmeanttoimplythatthereisnoexternalsocialreality;itmerely
suggeststhatour understandingof thatrealityisprofoundly influencedby thewaysitisrepresented
bywriters.
While postmodernismhasexerted a particular influenceon this last point, writers workingwithin
othertraditionshavealsocontributedtoit.
Thebasicstructureofandthewritingstrategiesemployedinmostquantitativeandqualitative
researcharticlesare broadly similar.
We needto get awayfrom the idea that rhetoric andthe desiretopersuade othersof the validityof
our work are somehowbad things. They are not. We all want toget our pointsacrossandto
persuadeour readersthatwe havegot things right. The questionis-do we do itwell?Dowemake
the bestpossiblecase? We all havetopersuadeothersthatwe havegot the rightangleonthings;
the trickistodo itwell.So,when you writeanessayor dissertation,do bear in mind thesignificance
ofyourwritingstrategy.
Questions for review
Whyisitimportanttoconsider thewaysinwhichsocialresearch iswritten up?
Writing up your research
Whyisitimportanttobeclearaboutyourmainargumentwhenwritingupyourfindings?
Writing up quantitative research
Read anarticlebasedonquantitativeresearchinaBritishsociologyjournal.Howfardoesitexhibit
thesamecharacteristicsasKelley and DeGraafs (1997)article?



I
lie
er
up
nd
mted
in
tyof
lake
~
ance
.ibit
What is meant by rhetorical strategy?Why might rhetorical strategiesbe important in relation to the
writing-upofsocialresearch?

Do Kelley and De Graafemploy an empiricist repertoire?
Writingup qualitativeresearch
Read an articlebased on quantitativeresearch in a British sociologyjournal. How far doesit exhibit
thesamecharacteristicsasBeardsworthand Keil's (1992) article?
How far isthe structure of Beardsworthand Keil's article different from Kelley and De Graafs?
-,
Writingup mixed methods research
Read an articlebased on quantitativeresearch in a British sociologyjournal. How far doesit exhibit
thesamecharacteristicsastheonebyPoortingaet al.?
Postmodernismand its.implications forwriting
Why haspostmodernismproduced agrowth of interest in writing socialresearch?
Whatisreflexivity?
Writingethnography
How far is it true to saythat ethnographic writing istypically imbued with realism?
Whatformsofethnographicwritingotherthanrealisttalescanbefound?
Whatarethemaincharacteristicsofrealisttales?
Online Resource Centre
http://www.oxfordtextbooks.co.uk/orc/brymansrm3e/
Visit the Online Resource Centre that accompanies thisbook to enrich yourunderstanding of writing
upsocial research. Consult weblinks, test yourself using multiple choice questions, and gain further
guidance andinspiration from the StudentResearcher's Toolkit.