Qualitative Data Collection Methods

midcoast ride

InterviewingandPsychosocialCaseStudyAnalysisPDF.pdf

Home >Law homework help >Criminal homework help >Qualitative Data Collection Methods

In-Depth Interviewing and Psychosocial

Case Study Analysis

In: The SAGE Handbook of Criminological Research Methods

By: David Gadd

Edited by: David Gadd, Susanne Karstedt & Steven F. Messner

Pub. Date: 2012

Access Date: November 17, 2019

Publishing Company: SAGE Publications Ltd

City: London

Print ISBN: 9781849201759

Online ISBN: 9781446268285

DOI: https://dx.doi.org/10.4135/9781446268285

Print pages: 36-48

This PDF has been generated from SAGE Research Methods. Please note that the pagination of the

online version will vary from the pagination of the print book.

https://dx.doi.org/10.4135/9781446268285

In-Depth Interviewing and Psychosocial Case Study Analysis

DavidGadd

Psychosocial Criminology: Something Old, Something New

Although the psychosocial approach has emerged relatively recently as a specialism within criminology, it

is possible to chart a history of delinquency research that addressed the relationships between inner-world

psychological and outer-world sociological issues dating back to the 1920s. Often, this work is dismissed

for its limited empirical basis and Freudian orthodoxy (Radzinowicz and King, 1977). In some cases this

criticism is justified (Hamblin Smith, 1922; Glover, 1960), but it should not cause us to forget the endeavours

of a number of early twentieth-century writers to put psychoanalytic concepts to the empirical test. Take, for

example, E.W. Burgess' (1930) analysis of the defensiveness of Clifford Shaw's (1930) Jack-Roller; John

Bowlby's (1946) case notes on ‘juvenile thieves’, variously traumatised by experiences of war, estrangement,

illness, institutional care, abuse or neglect; or Mark Benney's (1936) critical autobiography of working-class

life and how it contributed to his life as a thief. The best of this work used empirical evidence to engage with

the wider social scientific debates that have always mattered to sociological criminologists–crime, its causes

and consequences, and their relation to poverty, politics and social control–while at the same time taking the

complexity of criminal subjectivities, their potentially conflicted nature and the possibility of motives that were

not always consciously accessible to the individuals concerned, very seriously.

By way of contrast, contemporary criminology is more exclusively sociological in focus, concerned with

the cultures, structures and modes of social control to which we are all subject, but not so much with the

peculiarities of how these intersect with individual biographies, or the meaning making involved when they

impinge upon, or are resisted by, particular individuals. Criminological psychology has done little to rectify this,

risk-focused crime prevention merely compounding the problem with a model of the offender that reduces him

or her to the aggregate of ‘factors’ liable to instil a ‘propensity’ for offending. All of this is rather unfortunate

since some of the most interesting developments in social theory over the last 20 years have come from

those willing to explore a rapprochement between sociological work attendant to the intersections between

power and knowledge and the kind of psychodynamic thinking that the different strands of psychoanalysis

are variously concerned with (Du Gay et al., 2000; Henriques et al., 1984). Across the social sciences these

theoretical developments have given impetus to a new field of ‘psychosocial studies’ which seeks to explore

how the psychological and social constitute each other, without collapsing one into the other. The website of

the UK's fledgling Psychosocial Network, for example, defines the focus of psychosocial studies as the ‘the

tensions between and mutual constitution of the social and the psychic’, while the website of the International

Research Group of Psycho-Societal Analysis draws attention to the ‘dialectic between inner experience

and the social conditions of people's lives, paying attention to the conflictual and dynamic nature of both

psychological and social processes and their mutual effects on social action and change’. This is not the place

SAGE

SAGE Research Methods

Page 2 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

to chart the debates that gave rise to these formulations (Clarke, 2006; Frosh, 2003; Layton, 2008), but two

features of it are worthy of note. First, that in order to posit human subjects with depth and complexity, those

working within psychosocial studies have tended to find it necessary to assume unconscious motivations

inaccessible to the individual but nevertheless having effects on their behaviour and their relationships with

other people. Second, that in the process of addressing the irreducible relationships between the psychic

and the social, psychosocial theorists and practitioners have often turned to criminological subject matter

for answers. For them, the problems of crime and violence, racism and fundamentalism, abuse and how

it is remembered, fear and anxiety, retributive sensibilities, what lies behind them and what it is like to be

subject to them, require approaches that engage with issues of social structure, power and discourse, without

requiring us–as academics, practitioners, and researchers–to assume that particular people inevitably think

or feel in particular ways simply because of the things they have done, the groups they belong to, or because

of the times and places in which they have lived. Breaking with these assumptions is particularly difficult for

academics schooled in criminology, not least because so much of our discipline is rooted in governmental

projects that presume that certain segments of the population, especially the poor, share a common outlook

because of their structural positioning; that is, that people offend because they have experienced ‘strain’, have

inherited poor self-control from the ‘problem families’ in which they were raised, or because they have been

labelled as deviant by law enforcers or other agents of social control (Jefferson, 2002). Capturing agency,

without resort to rational choice models, is something criminologists have yet to prove very adept at, some

notable exceptions notwithstanding (Katz, 1988; Sykes and Matza, 1957).

Producing Rich Enough Data

At the time of writing, readers looking to begin their enquiries with a new body of psychosocial criminology

imbued with a more agentic model of the subject have only a small body of literature to contend with. Until

relatively recently the psycho-social criminological literature was largely theoretical (Anderson and Quinney,

1999; Brown, 2003; Hood-Williams, 2001), confined in its more empirical instances to the analysis of case

material produced initially, not for academic research, but in the course of criminal justice processes and/

or in the course of journalistic enquiry (Evans, 2003; Hollway and Jefferson, 1998; Jefferson, 1997). In this

context, Hollway and Jefferson's (2000) account of how they sought to develop an adequately psychosocial

approach to researching the fear of crime, Doing Qualitative Research Differently, established itself as the

pioneering text with respect to methodology, although there are other psychosocial methodology texts worth

engaging with beyond the criminological domain (Clarke et al., 2008a; Hollway, 1989; Kvale, 1996; Wengraf,

2001). Within this wider field of psycho-social studies there is a marked preference for material that is

richly textual, the advent of portable tape recorders and, subsequently, digital recording devices, making it

possible for a new generation of researchers to not only test out hypotheses about subjective inner-world

reactions in a way that was near impossible when psychoanalysis was first conceived, but also to evidence

their arguments about them through the production of highly detailed interview transcripts. Doing Qualitative

Research Differently capitalised on this moment. It was the first book in 40 years to make a sustained case

for the production of criminological data informed by psychoanalytic hypotheses. It was also a text that

SAGE

SAGE Research Methods

Page 3 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

resuscitated what had become a long dormant, but nevertheless invaluable, tradition of life-story work within

criminology, the methodological tenets of which had never been properly articulated (cf. Carlen, 1980; Shaw,

1930; Soothill, 1999). More than anything else, however, it was a book that underlined the importance of

attending as carefully as possible not only to what people say about themselves, but also how they say it,

what they struggle to say and what they cannot or do not (quite) say. What mattered most to Hollway and

Jefferson was the ability of their approach to produce analyses that did justice to the complexity of the people

they were researching: their guardedness and their bold assertions; their principles and their prejudices; the

tensions between their values and their desires; the contradictions between what they claim and their actual

experiences. Adapting ideas from the psychoanalytic work of Melanie Klein, Hollway and Jefferson adopted a

model of subjectivity that has anxiety at its core. Essentially, we are all anxious beings and much of what we

say and do is motivated by a desire to defend against the feelings of vulnerability this anxiety arouses in us.

For them, incomplete, contradictory or confused sentences are just as important as wordperfect responses,

since both can read as evidence of the character structure of inherently defended subjects. What was most

critical was that complexities in the data were treated as evidence of the nature and depth of the people

under study, and not written out of the analysis as mere idiosyncratic noise of no criminological interest. For

Hollway and Jefferson it is only when both the unique and the culturally common features of people's lives are

adequately grasped that they become truly recognisable, enabling us to see which aspects of our experiences

are akin to our research participants' as well as what it is that makes us different.

In writing their own methodology book, Hollway and Jefferson took inspiration from a then emerging body

of German sociological research being conducted on Holocaust survivors and service personnel who had

served the Nazis during the Second World War (Bar-On, 2004; Breckner, 1998; Rosenthal, 1993; Rosenthal

and Bar-On, 1992). Noting how the methods used in that research elicited rich material from interviewees who

had good reason to deceive both themselves and the researchers, Hollway and Jefferson (2000) spotted the

potential of narrative methods to grapple with the more commonplace phenomenon of human defensiveness.

Troubled by a British policy debate that assumed that people ought to be able to calibrate their fearfulness

according to what is known about their risk of victimisation, Hollway and Jefferson elaborated on the dangers

of research that failed to take the issue of subjectivity seriously. In the 1980s people who ticked the same

boxes in the British Crime Survey were assumed both to be ‘telling it like it is’ and to have meant the same

thing, with no checks being made to test the validity of these unquestioned presumptions. Arguing that the

problem was not so much the public per se, but about a failure to anticipate in the design of research

instruments how the public experience crime and fear, Hollway and Jefferson began to develop an alternative

methodological approach that would encourage individuals to convey the complexity of their experiences and

preserve as much of the uniqueness of what was disclosed in the course of the data analysis.

Through the use of a narrative-focused, experientially responsive approach to interviewing, Hollway and

Jefferson attempted to ‘transform’ themselves ‘from the highly visible asker[s] of … questions to the almost

invisible, facilitating catalyst’ of their interviewees' ‘stories’ (ibid. 36). The essence of their approach was to

encourage people to tell their own stories and then to ask questions that invited participants to elaborate

further about their own particular experiences. Rather than asking people to score how safe they felt walking

SAGE

SAGE Research Methods

Page 4 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

around their neighbourhoods at night, Hollway and Jefferson encouraged their interviewees to recall whatever

came to mind–even if seemingly irrelevant to the topic–when invited to talk about how crime had impacted on

their lives since they had been living wherever they were. As the following extract reveals, participants often

rose admirably to this invitation, introducing aspects of their experience that would not have been elicited

using closed questions, and narrating those experiences with the kind of meaningful disclosures that have to

be taken account of if one wants to understand why some people are more fearful than others.

Extract 1: Interview Conducted by Tony Jefferson with ‘Arthur’,

October 1995

TJ: Can you tell me about how crime has impacted on your life since you've been living here?

A: Well the only thing that you've got to be wary about really is er, like at the moment I've er, I've

got a disability because–on account of this injury at work I had to retire and now I'm having to walk

about with a stick. (light laugh) [TJ: Yeah.] And I dread–you know, I'm a bit nervous about going out

at night. [TJ: Um hum.] Er, you never know who's at back of you. Whereas when I were fit it wouldn't

have bothered me. [TJ: Mmm.] But there is a lot of er, things happen on this estate that er, you know,

last few years it's got really worse. [TJ: Um hum.] I don't know whether that answers your question

but, it don't em, it don't bother me really but only thing you worry about is if you go away, you never

know when you'ree gonna come home er, who's been in your house.

TJ: Can I just take those things one at a time that you mentioned?

As the extract reveals, the strength of this approach was in its capacity to generate the nuanced material

needed to make sense of how particular individuals feel about crime and the disjunction between this and

what, in certain circumstances, they might be prepared to say about it. Tony Jefferson's invitation to Arthur

to tell him how crime had impacted on his life enabled Arthur to explain that he is not so much afraid but

‘nervous’, and not necessarily on account of having been victimised. On the one hand, Arthur was ‘bothered’

about being out at night and leaving his house unattended now, but he had not always been so back in the

days before he was injured at work, lost his ‘fitness’, and retired; changes in his life which he considered to

have coincided with when the ‘estate … got really worse’, whatever that meant. On the other, we learn that

Arthur was either not unambiguously bothered or, perhaps more likely if his ‘light laughter’ is read as ‘nervous

laughter’, that he did not want the interviewer to think he was, hence his retraction, ‘it don't bother me really'.

Making sure that such ambiguities and tensions are retained as data to be made sense of within the analysis

is critical for researchers trying to undertake psychosocial analysis. Hollway and Jefferson, in particular, refer

to this as the Gestalt principle (an idea they borrow from the German methodologies discussed earlier),

meaning that the whole of any one case–including how it is organised, presented and structured–is always

greater than the sum of its individual parts.

In terms of how the Free Association Narrative Interview Method works, the dialogue between Tony Jefferson

SAGE

SAGE Research Methods

Page 5 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

and Arthur illustrates well the nature of the ‘free’ or spontaneous ‘associations’ that are being sought. Having

facilitated Arthur's responses to the opening question through the use of a string of very minimal cues–‘Yeah’,

‘Mmm’ ‘Um hum’, Jefferson's next prompt prepared Arthur to go back through his initial reactions, ‘one at a

time’. The full version of the transcript (available via the Economic and Social Data Service's website listed

at the end of this chapter) reveals that Jefferson achieved this through a combination of techniques that

became the principal tenets of the Free Association Narrative Interview Method: presenting elements of what

Arthur had disclosed back to him (his ‘disability’ feeling a ‘bit nervous about going out at night’ the ‘things’

that ‘happen on the estate’) in the order Arthur had revealed them; asking Arthur to say a bit more about

these; reflecting back some of Arthur's reactions in order to encourage him to elaborate further; and by posing

further narrative-eliciting questions that used Arthur's words and phrases as prompts.

Importantly, for practitioners of the Free Association Narrative Interview Method, these follow-up questions

are never ‘why?’ questions. In this method, ‘why?’ questions are explicitly avoided because in asking

participants to explain things they may not be able to explicate, ‘why?’ questions have a tendency to elicit

clichéd rationalisations, further obscuring the uniqueness of the interviewees' biographical experiences from

view. Often interviewees do not know very much about the problem under study–at least when expressed

in the policy-relevant terms through which research projects are increasingly conceived–and are hence

uncertain both as to what caused it and what should be done about it. As Adorno (1950) once perceptively

noted, and as a swathe of psychological research in the social identity tradition now confirms (Branscombe,

1999), when pressured to give an opinion, respondents who are otherwise unsure in their viewpoints can

feel inclined to express authoritarian attitudes so as to avoid appearing like naïve outsiders (see also the

chapter by Roberts et al., this volume). The same can be true of people's understandings of aspects of

their own personal experiences. As Hydén (2008: 125) remarks: ‘The significance of the difference between

“having been through a lot” and “knowing about” what happened is that of reflection and giving meaning to

experience.’

In criminological research it is not uncommon to encounter people who have not benefited from this kind of

knowledge-enhancing reflection, who do not know why they did what they did, nor why terrible things have

happened to them. In such circumstances, it is rarely productive to pressurise interviewees into providing

an explanation. At the same time, it is not always easy to find narrative-eliciting questions to ask such

interviewees, especially when the focus of the research is on behaviour that is illicit, attitudes that are taboo,

or experiences that are potentially shameful. Not everyone who agrees to participate in this kind of research

finds it easy to tell–or construct afresh–their stories. Most need the interviewer to help facilitate and many

need to be convinced that they are genuinely allowed to introduce any aspects of their experience they deem

relevant. As Hollway and Jefferson explain:

Interviewees' preparedness to open out intimate material also reflects the building up of an

expectation that stories are what the researcher wants–that they are interesting, relevant and

valued. This expectation has to be actively built because the normal expectation is otherwise; that

is the interviewer will come round with a batch of questions, for which one-word or short replies are

SAGE

SAGE Research Methods

Page 6 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

required (Hollway and Jefferson 2000: 44).

The beauty of the Free Association Narrative Interview Method is that with skill and practice even short

one-word replies can be used to elicit accounts that might otherwise have proved inaccessible to research.

Consider, for example, the extract from an interview I conducted with a former far-right activist that follows.

The numerical values in square brackets signify the length of paused silences in seconds.

Extract 2: Interview Conducted by David Gadd with ‘Frank’, April 2004

DG: I was wondering if you could start by sort of telling me the story of your life in Stoke-on-Trent,

obviously start as far back as you can remember and take as much time as you need. What I'll do

is, I'll take some notes so that I can ask you some questions as we go along.

F: Well, I was born in [mining estate], come off a family of seven, five brothers and one sister. [16

second pause] Basically I don't really know what you want like. That's the thing about it. I mean a lot

has gone on in me life like. What specific points, like?

DG: Well if you, I mean, perhaps you could tell me the story of growing up in [mining estate] to start

with.

F: Growing up, well, we had nothing, never had nothing like, you know coming off a big family (DG:

Hm, hm). Me dad wasn't too kind to us.

DG: He wasn't kind?

F: No not really. [3 second pause] Always fighting at school, you know. Cos me dad said so (DG:

Hm, hm). [25 second pause]. What basically do you want to know, like?

DG: Well what you've told me so far is that, you told me, you had nothing and your dad wasn't kind

and you were into fighting at school, got into fighting at school. So maybe we could start with that. If

you could kind of tell me how having nothing kind of impacted upon your upbringing.

Even in this highly reticent retort, littered as it was with painfully long pauses, the interviewee, Frank, provided

several leads through which further narration could be pursued. From it, new narrative questions could be

formulated about the estate where Frank was born, his family, his siblings, or even about the whole ‘lot’ that

had ‘gone on’ in his life. I initially responded with a narrative question about life on Frank's estate, then with

some reflection (‘He wasn't kind?’), some minimal encouragement (‘hm, hm’), toleration of long pauses, and

then by offering back, as a narrative-evoking question what appeared to be one of the more meaning-laden

of Frank's initial disclosures, namely that his family ‘never had nothing, like’. My response was not perfect.

The reflection might have been more precise had it been phrased: ‘Your dad wasn't too kind?’; and one of

the questions clumsily exposed Frank to my disarray, ‘Well if you, I mean, perhaps …’. But in combination the

techniques did ultimately elicit many more in-depth disclosures from Frank, a man who was not accustomed

SAGE

SAGE Research Methods

Page 7 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

to talking about himself in this way, but who nevertheless had a very important story to tell (Gadd, 2006).

Of course practitioners of the Free Association Narrative Interview Method also run the risk of being

overwhelmed by the extensiveness of what is revealed when interviewees begin to ‘freely associate’.

This may be especially likely to happen in research with participants who have been waiting to tell

somebody–perhaps anybody–about things they have done or have happened to them, safe in the knowledge

that assurances of confidentiality will be honoured. It is not uncommon for novice researchers to approach

this challenge with the view that it does not matter if they do not hear and remember what the interviewee

tells them because they can always listen to the recording later. There is an element of truth to this. The Free

Association Narrative Interview Method works by returning to participants for a second follow-up interview,

informed by an analysis of the first interview–a process to which this chapter returns. Nevertheless, it is a

mistake to assume that the first interview is not compromised when the interviewer fails to demonstrate to

the interviewee–by accurately utilising the their words and phrases to pose further questions–that they have

listened attentively to what has been disclosed. Re-expressing what the interviewee has disclosed in words

that were not their own can make them feel misconstrued, or, worse still, prejudged for their particular mode

of expression; and this, in turn may make them much more guarded than they would ordinarily be in terms of

what they are prepared to say and how they are prepared to say it.

Learning to take notes while retaining a level of eye contact with the interviewee is one of the more difficult

skills practitioners of the Free Association Narrative Interview Method must develop if they are to gain the

interviewee's confidence. Being able to keep eye contact at the same time as taking notes is often crucial in

the opening moments of the interview because, if captured accurately, what is revealed when the interviewee

does begin to open up can be utilised to structure the rest of the interview. Consider, for example, how many

lines of enquiry were opened up by the following response to an invitation to Alan–a 39 year-old man on

probation for a racially aggravated assault–to tell his life story.

Extract 3: Interview Conducted by David Gadd with ‘Alan’, May 2004

A: I'm the youngest boy out of ten children. I got a younger sister. And everybody else is older than

me. I've got five brothers, four sisters. Em, my mum was from Stoke on Trent, and me dad was from

South Wales. So I'm half Welsh, half English. Em. Earliest memories is infant school, which is now

closed, em moved on to … Middle School, and then … High School. Em, I got into football at an

early age. I went to me first. I'm a big Stoke City fan. Went to me first football match when I was

ten. And I, I have been crazy about Stoke City ever since and that has played an integral part of

me life. Em … Went off the rails, early teens. From being quite a bright lad I just got no interest in

school whatsoever. I was bunking off to work on the markets, em ice cream vans and things like

that to earn some cash. [DG: Hm, hm] I got into the punk rock movement in me early teens. Em,

followed me through until I left school, basically … I left school at 15. I vastly under achieved. Got

into drinking again while I was still at school. Going to local pubs and that. Started getting into trouble

just after I left school. Em basically, em drink related incidents and football hooliganism. … Eh … Got

SAGE

SAGE Research Methods

Page 8 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

sentenced, got sent to detention centre, when I was 19, for 3 months … That was for fighting on the

pitch at [away game]. 1984. Em, come out of there, drifted from one job to another. … Em lost me

driving licence through drink driving. Signed meself up in college and got my City in Guilds in painting

and decorating. That was me line of work then for, until like 6 years ago. Em I've suffered mental

health problems some very severe. Em, one in 2000 lasted about three to four months and again in

2002 which lasted a similar time. Em I've suffered from depression again since me teens, em I've

dabbled with drugs. Er, oh, alcohol has always been a big problem in me life. Em about me two spots

of mental illness, em we're not talking just depression there. We're talking severe psychotic bouts.

[DG: Right.] A: Em probably caused by long term cannabis and alcohol abuse. Not 100% certain on

that we're not, but it's the probable cause of it. Em, … I'm sorry mate, I'm blanking here, I'm, I could

do with more questioning.

This extensive disclosure is testimony to the richness of human experience that can be captured with the right

methodological approach. In it Alan provided enough material for the interviewer to map out a provisional list

of questions that exceeded what was possible within the confines of a ninety-minute interview. Responding to

Alan's request for more questioning, while being guided by the ordering of his disclosures, a Free Association

Narrative interviewer could respond by asking Alan to ‘tell the stories of’: growing up in a large family of five

brothers and four sisters; his schooling; getting into football; going to his first match aged ten; ‘going off the

rails’ and ‘losing interest’ in school; working on the markets, ice cream vans and ‘things like that’; getting into

punk rock; drinking, drug use, and/or football hooliganism; being arrested and detained; drifting from one job

to another; getting his ‘City and Guilds’; his experiences of psychosis and depression; or more specifically

his ‘two spots’ of mental illness and their possible connection to drink and drug abuse. With a forthcoming

interviewee like Alan, the interviewer was almost sure to elicit detailed narrations that could be probed even

further, as long as the questions posed invited more stories and accurately utilised his meaning frame.

Box 3.1 Key Tenets of the Free Association Narrative Interview Method

• Before the interview commences prepare the interviewee for the kind of research

encounter that will follow.

• As the interview commences reassure the interviewee that what they have said

has been interesting and that you would like to hear more of their stories.

• Ask narrative-focussed questions wherever possible.

• Avoid ‘closed’ questions and ‘why?’ questions.

• Do not interrupt the interviewee when they are talking or fill their pauses and

silences when they are thinking.

• Keep notes while remaining attentive.

• Facilitate with minimalist cues. This can mean simply nodding, saying ‘Hm, hm’,

SAGE

SAGE Research Methods

Page 9 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

‘right’, or encouraging the interviewee to ‘go on’.

• Reflect back the interviewee's sentences or phrases in order to facilitate further

disclosures while showing that one has listened properly to what has already been

divulged.

• Use the interviewee's frames of meaning to pose new narrative questions.

• As far as possible, ask questions in an order that mirrors the way the interviewee

has ordered their disclosures.

Preparing Follow-Up Interviews

While questions left over from the first interview can be pursued in the second interview, this is not the

primary purpose of the follow-up interview. For those using the Free Association Narrative Interview Method

the analysis starts when the first interview ends. What is gathered in the first interview is used to formulate

provisional hypotheses about the interviewee which are tested in the second; ‘to interrogate critically what

was said, to pick up on the contradictions, inconsistencies, avoidances and changes of tone’ (Hollway and

Jefferson 2000: 43); and to think about what more one would like to know and how best to ask about it.

Both for reasons of continuity and to minimise the risk of losing contact with the participant, most people

using this method tend to arrange follow-up interviews within two weeks of the first interview. The available

time between interviews can, therefore, put constraints on how thoroughly one can prepare for the second

interviews. It is worth remembering this when scheduling both research encounters. If it is possible to

generate a full transcript of the first interview ahead of preparing the questions for the second, this is worth

doing. If not, listening to the recording several times may be the best substitute. In order to make sure

the dynamics of the interview remain in view, transcripts need to be as full as possible, incorporating half

finished words and sentences, noting changes of tone and/or accent, gesturing and other non-verbal and

non-linguistic modes of expression.

There are no hard and fast rules for working on the data produced in the first interview, not least because what

one picks out will always be partially informed by the focus of the research being undertaken. That said, this

is not the kind of research that is best undertaken completely alone. Often, how the interviewee comes across

at interview, on tape and through the transcript will vary, and this should be the subject of reflection among

the research team during the preparation for the follow-up interview. It is often helpful for the interviewer to

develop their own working hypothesis, hunches and queries, informed both by the substance of the interview

and any other information they picked up about the interviewee, and then to share this with a supervisor, co-

researcher, or interested group of scholars who are willing to either listen to the recording, read the transcript,

or both. At this stage it is not necessary to resolve differences of opinion with regard to the interpretation of

the first interview, and some of those using this approach have found that such differences can be fruitfully

harnessed in print as a means of conveying both the complexity of the people studied and the risks inherent in

SAGE

SAGE Research Methods

Page 10 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

reducing them to the sum of their attitudes (Clarke, 2008b). What does matter, however, is that the interviewer

develops a sense of what might be distinctive, revelatory or perplexing about the interviewee. Sometimes

these puzzling qualities are connected to unusual things the interviewee has said or claims they repeatedly

make without contextualising or explaining them. But, at a more everyday level, what can also be puzzling is

why people tell their story one way and not another. Although not an exhaustive list, I have found the following

facets of interviewees' accounts particularly worth exploring in follow-up interviews:

Avoidances. Does the interviewee avoid answering particular questions, appear reluctant to talk

about particular events or periods in their life, or change the subject for no apparent reason? Is

there a way of drawing their attention to what appears to have been avoided without causing

offence?

Conflations. Are there points in the interview where several events are inextricably conflated,

making their ordering difficult to decipher? For example, is the recollection of something in the

recent past being told as if it followed the logic of something that happened much earlier, or vice

versa?

Inconsistencies. Are there inconsistencies or contradictions in the stories told? Are two or more

different accounts of the same event offered? Are assertions made that are not supported by

the examples given? Does the chronology of events seem plausible, or are there events that are

unlikely to have happened when the interviewee claims they did? Is the story consistent with the

historical record as well as internally consistent? How does the story told measure up against

what other interviewees have said about the same event or phenomenon?

Delivery style. Are there sudden changes of pace or tone during the interview? Is there a pattern

to the pauses or hesitations? Is a third person perspective introduced when a first person account

might have been expected? Are some events depicted as things that ‘just happened’, as a product

of good or bad fortune, when other are construed as entirely under the interviewee's control and

of their own making? Does the interviewee ask the interviewer to endorse their viewpoint, become

inexplicably defensive, or seek some additional recognition or validation beyond the affirmative

facilitation provided? Was this recognition forthcoming, and if so–or if not–with what effect?

Unsubstantiated generalisations and projection. Are there certain descriptive terms that are

repeatedly applied to a range of quite different events or relationships? Does the interviewee

accuse others of doing things that they have done? Are insinuations made–whether about

significant others, the government, or people like the interviewer–that merit further exploration?

Reconciling the need to follow-up issues like these with the need to pose more topic-focused questions that

the project or its funders require can be difficult. Anticipating how the second interview may unfold and being

able to prioritise is therefore critical, not least because interviewees often wish to introduce new stories or

revise their first interview accounts in the second interview. Space needs to be made for this, not only out of

courtesy to the interviewee, but also because such revisions can be revelatory in and of themselves.

SAGE

SAGE Research Methods

Page 11 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

Pen Portraits and the Challenge of Interpretation

Because of their commitment to the Gestalt principle–the idea that the whole of the person being interviewed

is greater than the sum of any of their particular attitudes, experiences or other disclosures–those using the

Free Association Narrative Interview Method tend to favour modes of analysis that endeavour to preserve as

much the data's complexity as possible until an analysis that can be presented to a wider audience has been

produced. A case study approach is often used, not only because of the depth of the material garnered about

interviewees' experiences, but also because there are almost always

• multiple dimensions to these experiences that are difficult to disentangle from each other;

• a range of hypothesis that ought to be considered in relation to these experiences, and;

• conceptual difficulties to overcome with respect to deciding what is distinctively personal and/or

biographical about the person's life and what is a relatively common product of pervasive cultural or

social attitudes (Yin, 2009).

Hollway and Jefferson's response to these challenges has been to construct textual ‘pen portraits’ of their

interviewees that bring their case material alive to readers unfamiliar with the original sources. Essentially

these pen portraits are descriptive pictures that bring together all the information to hand about a particular

individual–including any ‘inconsistencies, contradictions and puzzles’ evident within the interview

transcripts–enabling ‘subsequent interpretations’ to be ‘assessed’ against them (Hollway and Johnson, 2000:

70).

Critics of this approach have argued that such pen portraits are more ‘constructed’ than Hollway and Jefferson

concede; constructed with a particular hypothesis–that we are all ‘defended subjects’, motivated by anxieties

of which we were not fully conscious–fully in mind. For some such pen portraits present us with too ‘definitive’

a story, leading almost ‘automatically’ to revelations about the ‘real character’ of the people under study

(Wetherell, 2005: 171). This is indeed a real risk of the method, although one that can be countered if one

is both aware of it and upfront about the hypotheses that have informed the data analysis. It certainly helps

that today's psychoanalytic interpreters–unlike those of the early twentieth century–can and do make their

data available through archives so they can be interrogated by others, and furthermore, that some cases

are analysed not only by academic groups (as they are in the approaches advocated by Oevermann et al.,

1987 and Wengraf, 2001, discussed later) but in the course of public debate; that is, through special issues

of journals and conference sessions specifically devoted to single cases.

Moreover, in the contexts where I have conducted research using the Free Association Narrative Interview

Method–in research with domestic violence perpetrators and racially motivated offenders–the converse risk is

just as great. Too often the orthodox criminological position relies upon a one-dimensional model of offender

motivation–for example, the idea that racially motivated offenders offend simply because of their ‘biases’ or

that domestic abusers are driven exclusively by a will to power–that oversimplifies the complex humanity

of the people whose behaviour it purports to understand. Of course, none of us can avoid simplification,

SAGE

SAGE Research Methods

Page 12 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

but when it comes to writing about offenders, academic work that does not do full justice to the humanity

of the people it studies runs the additional risk of compounding the very demonising discourses folk devils

and moral panics are made of. Showing that more complex models of criminal motivation can be sustained

with the right kind of evidence has been the aim of all my research using the Free Association Narrative

Interview Method and the reason why I have pursued modes of analysis that are sensitive to ambiguities of

meaning and the interpretive challenges these pose. The following extract, abbreviated from a pen portrait of

‘Alan’ published in Gadd (2009) illustrates this point, while at the same time revealing more about whom this

particular interviewee–introduced earlier–turned out to be.

Psychosocial Case Analysis

Before one begins to develop an analysis of data like this it is often helpful to refine one's research questions

further, focusing the interpretive work required more precisely. For example, it was when I began to wonder

why so many people accused of racial aggravated assaults resist becoming labelled as racist more than they

resist being known as offenders (Burney and Rose, 2002), that I started to revisit

Box 3.2 Abbreviated Pen Portrait of Alan

Alan was a 39-year-old white English man of muscular build, on probation for the racially

aggravated assault of an Asian taxi driver. He claimed not to be able to recall the assault

for which he was convicted because he was mentally ill at the time. Alan accepted that he

might have spat in the face of the taxi driver in the ‘heat’ of the moment–‘I'm terrible for

it. I think it's like the ultimate insult’–but doubted he was guilty of the racially aggravated

element of his conviction. He could not imagine using the words ‘Paki bastard’ he was

alleged to have said.

Alan's disavowal of the racial dimension to the assault appeared to be consistent with his

the non-racist beliefs. When, as a schoolboy, he worked on the markets, he ‘got on like

house on fire’ with the ‘Indian or Pakistani’ traders. As a teenage football hooligan, he

remembered two ‘coloured’ guys in their gang and that ‘they never got no racist abuse off

us’. He used to wear a ‘Rock Against Racism’ badge and would take issue with younger

hooligans who purported to have connections with militant far-right groups. Nowadays he

was appalled by post-9/11 Islamaphobia. This kind of racism had, in his view, done ‘a lot of

damage to the Muslim community … which they don't deserve’. Alan even claimed to be

‘good friends’ with the other Asian taxi drivers who worked for the ‘same firm’ as the man

he had assaulted. He had taken one to a local soccer match and ‘introduced him’ to all of

SAGE

SAGE Research Methods

Page 13 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

his ‘friends down the match’. Three of his niece's children, all of whom he loves ‘to bits’,

have a father who is Pakistani (and also a taxi driver). While there had been some family

animosity over this niece's choice of husband, it was, according to Alan, ‘nothing racial’.

The animosity, Alan said, had more to do with ‘how they treat their women’ and the strong

suspicion that the father, with whom ‘everybody gets on’, has a second family in Pakistan.

As for the taxi driver who had been assaulted, Alan had hired him on two previous

occasions before the alleged assault. On both of those two previous occasions the taxi

driver, according to Alan, had made sexually derogatory remarks about women. The taxi

driver's offensive remarks on the first occasion were with reference to two young girls

walking past. In his first account of this journey, Alan claimed the taxi driver had said: ‘I

wouldn't mind giving it her up the arse’, to which Alan said he replied: ‘I'm not like that.

I don't enjoy people speaking like that about women.’ In the second interview, however,

Alan recalled the taxi driver's remarks as ‘I bet you like having it up her arse mate’ to which

Alan said he had replied: ‘I'm not like that … I don't enjoy people speaking like that about

women … You don't know me’.

Returning from a night out, at the time of the second taxi journey, Alan had ‘had a few to

drink’, when the driver allegedly asked him: ‘Where you been mate? You been fucking?’

Alan said he replied:

No offence mate, but this is twice you've made derogatory offence like remarks

about women and now you're asking me if I been shagging … You're nothing but

a pervert.

For this reason, Alan thought that during the third taxi journey, when he was alleged to

have spat at the driver, he would have called him a ‘pervert’ not a ‘Paki’: ‘because that's

what I genuinely thought of him’.

Asked whether he could recall other times when he felt offended by ‘sexist or sexually

derogatory comments’, Alan remembered a boy at school with ‘dark-looking skin’, now

serving a life sentence in prison for rape. Alan knew some of the victims and said he would

‘joyfully’ act as the rapist's executioner. Another instance when questions of sexuality were

touched upon by Alan arose in relation to his explanation of the ‘male bonding’ dimension

of the football hooliganism he had found so ‘addictive’ since his teens. ‘It was’ he said,

always nice if … you was on the coach and you were a bit scared … and some

older lad would say, ‘stick with us kid… You'll be alright’. It's like a … a male

bonding type of thing. I'm not homosexual, or nothing like that.

SAGE

SAGE Research Methods

Page 14 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

This male bonding often involved the bloodiest of violence, Alan's fights with rival fans and

police officers exposing him to some serious injuries, numerous bans from the club, and

a stint in a detention centre in his late teens. It also involved overt sexism: ‘It's something

different when ten thousand lads are singing to a pretty model parading round the pitch,

“Get your tits out for the lads”. That's a bit of a laugh.’

The bravado this male-bonding entailed contrasted sharply with the pain Alan felt in

relation to his own relationships with women. When his first fiancée discovered that Alan

had been unfaithful, she ended the relationship. Alan was so ‘devastated’ that he took

an ‘overdose’, ‘slit his wrist’, and became ‘really down depressed’. His ‘stormy’ ten-year

marriage to another woman ended when he discovered she was having an affair with a

neighbour, whereupon Alan hit his wife hard enough to put ‘her arm in a sling’ and ‘spat in

her face’. His father dying the same week, Alan was ‘cut up’ with grief. He started ‘drinking

heavily’ and began ‘trawling’ the pubs and clubs ‘looking for another partnership’. The

assault on the taxi driver occurred around this time, in the midst of mental health problems

that were subsequently diagnosed as ‘psychotic bouts’.

the pen portraits of racially motivated offenders, we, as a research team, had constructed several years

earlier (Gadd et al., 2005). Remembering that when I had interviewed Alan I had quite liked him and did

not experience him as particularly racist or spiteful, I began to wonder how it was possible that his victim

might have experienced him otherwise. In hindsight, I suspect I identified closely with Alan and his vicarious

accounts of violent hooliganism at the time of the interviews, and only became fully conscious of this when I

began to anticipate how his account would be received by people from minority ethnic groups and/or others

who might see themselves as potential victims of men like him. I looked again at the transcripts and began

to contrast the evidence suggesting that Alan was not racist with those elements of his disclosures which

appeared, at least potentially, to have some connection to racialisable hatred. Alan did not have particularly

racist attitudes, disliked the racism he witnessed in others, and was not regarded as ‘a racist’ by his probation

officer. Other researchers, working within different traditions, might have concluded on this basis that Alan

was most likely not guilty of the racially aggravated element of the assault for which he had been convicted,

his own perspective affirmed through triangulation with the researcher's assessment of it and that of a

knowledgeable and critical third party. To settle on this interpretation would, however, have meant losing sight

of that fact that the police, the courts and the taxi driver had all perceived things rather differently. Was it not

equally possible, given his self-confessed propensity for violence, proclivity for spitting, drinking and mental

health problems, that there was another side to Alan, a more unpredictable, frightening and abusive side than

the one both I and his probation officer had seen? In order to answer this question I returned to the pen

portrait armed with theories and hypothesis drawn from the literatures on hate crime and racism. Thereafter,

two further contradictions began to stand out in the transcripts:

The contradiction in Alan's attitudes towards women. Alan's self-professed ‘respect’ for women

SAGE

SAGE Research Methods

Page 15 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

sat uncomfortably with the chauvinism he indulged in with other football hooligans and the

disrespectful behaviour his ex-partners had experienced from him.

The contradiction in how he recalled what the taxi driver had said on the first journey. The first

account of this journey entailed a story about what the taxi driver said he would not mind doing;

the second entailed a story in which the taxi driver appeared to be making insinuations about what

he thought Alan would have liked to do to two female strangers.

Could a fuller understanding of the role of these contradictions in Alan's thinking help explain why Alan

had wanted to deliver the ‘ultimate insult’–spitting–to the taxi driver? Had this act somehow enabled him to

temporarily avoid the feelings of loneliness and grief he was trying to satiate with alcohol while ‘trawling’ for

another relationship? Was there a projective dimension to the generalisation Alan made about the way in

which Asian men, like the taxi driver and his nieces' husband, ‘treat their women’? Was there a connection

to be made between Alan's own assertions about his sexuality and what he presumed, probably erroneously,

the taxi driver was insinuating about him? Returning to the transcripts with these questions, I was able to

produce an analysis that showed how it was possible for someone who was not overtly racist, like Alan, to

become susceptible, in the ‘heat’ of the moment, to a racialised form of hatred of which he was not ordinarily

conscious, and hence to make a much wider argument about why racial aggravated crime remains such an

elusive phenomenon for those trying to tackle it in, despite a swathe of reform to the law and the institutions

of criminal justice in recent years.

Readers interested in the exact substance of this analysis should consult Gadd (2009). What is important

to underline here, however, is that psychosocial analysis requires the researcher to commit to an ongoing

dialogue with the original data in all its complexity, their own and other's people's queries about it, potential

explanations, and a continuous sensitivity to the tensions within and between inner world experiences (like

love and hate) and outer-world phenomenon (like racist attacks). Working in this way it becomes possible to

achieve more holistic interpretations than are often achieved by other approaches to criminological research

studies, including the kinds of qualitative research that construct a common, but often quite meaningless,

mindset out of many different people's responses. Often, however, it is only possible to achieve a more

nuanced analysis by cultivating, over time, a heightened level of reflexivity in the research process, something

which is much harder to do than those sceptical about the worth of positing ‘defended subjects’ usually

realise. This, as Wengraf (2001: 144, emphases in original) surmises, is because

[i]n an interview, there are two anxious defended subjects, not one. In the interview, the researcher

also must be assumed to be at least potentially ‘motivated to not know’ certain things that would

be upsetting for him or her, and thus subtly or obviously influencing the production of some or all

of the text of the interview. In addition, the researcher's anxieties do not vanish after the end of the

interview. They may even be exacerbated.

Defence and anxiety are features of the researcher-position right through data-analysis and into the writing up

and publication. The operation of such anxiety in the stages of analysis and interpretation cannot be avoided,

but its shape and implications need to be detected and combated.

SAGE

SAGE Research Methods

Page 16 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

Current Controversies and New Horizons

Wengraf's own view is that such anxieties are best combated through panel discussions. Advancing the

objective hermeneutical approach advocated by Ulrich Oevermann and colleagues (1987), those using the

method preferred by Wengraf, bring together groups of people–often colleagues, doctoral students and co-

researchers–to examine extracts from the transcript and to hypothesise about what kinds of meanings may

lie latent within the spoken words. For Wengraf, bolstering the diversity within the panel is the best protection

against the hindrances of the researcher's defensiveness: ‘the more intercultural and cross cultural the

panel, the more the “sleeping assumptions” of any given researcher are likely to be disturbed and raised

to consciousness’ (Wengraf, 2001: 145). But it is also the case that whether or not people are able to

overcome their defensiveness can depend as much upon how the issues are presented to them, who is

doing the presenting, and, critically, their perception of whether that person holds them in high esteem, as the

actual substance of challenge being made. In panel discussions many of these moderating prerequisites are

often lacking. Consequently, when panel members interrogate the role of the interviewer they run the risk of

entrenching the researcher's defensiveness further (Froggett and Wengraf, 2004).

In criminology we are only just starting to come to terms with the ways in which the relationships between

interviewers and interviewees have shaped what we think we know about crime and criminals, Loraine

Gelsthorpe's (2007) intricate analysis of the relationship between Clifford Shaw and Stanley unearthing an

untold story of such dynamics that changed the direction of our discipline for the course of century (ibid. 2007).

Rarely are criminologists privileged enough to get to know their subjects as well as Shaw got to know Stanley,

even if those of us who do a lot of biographical and narrative work often leave our research encounters with

the perception that our research participants are more like us than we had anticipated. How, when and why

we identify with some research subjects and not others ought to be reflected upon much more frequently and

in much greater depth. As I have documented in my research about domestic violence, it is all too tempting

to pass over those research encounters where one has struggled to identify with the interviewee as simply

‘bad interviews’, rather than ponder what it was about the research relationship that mired the interviewee's

participation (Gadd, 2004). Shoshanna Garfield and her colleagues (2010), by contrast, have illustrated, how

identification with what research participants say can blind one to the meanings behind the words spoken.

Garfield's analysis exposes how her unconscious (over) identification with an interviewee who complained

about how little his partner did around the house caused her to overlook what she already knew about him,

namely that he had blinded his wife by hitting her in the face with a rolling pin.

What both my own work and that of Garfield et al. reveals therefore is just how emotionally and intellectually

demanding engaging with the kind of ‘toxic’ material the Free Association Narrative Interview Method can

elicit from violence perpetrators often is; how acute the desire ‘to have company in the hard task of bearing

witness’ to this material can be (ibid. 166); and how unnerving it is to confront the ‘wounded parts’ of oneself

that often underscore interviewer–interviewee identifications when they come to the fore in data analysis,

or even in the course of presenting this kind of research to wider audiences (ibid: 167). The opportunity to

SAGE

SAGE Research Methods

Page 17 of 21 In-Depth Interviewing and Psychosocial Case Study Analysis

bear witness to the toxicity of what some criminological research participants say and the distress, pain and

trauma that often lie behind it is, of course, both what attracts some researchers to using the Free Association

Narrative Interview Method in research with offenders and also what puts others off it. It is a method that often

enlarges one's own emotional repertoire while putting intense demands on it. These demands tend to persist

long after the interviews have ended and the writing up has commenced, for they also–as my experiences of

revisiting the case of Alan reveals–provide the nagging doubts that inspire intellectually driven analyses. To

put the matter starkly, in research addressed to why violence perpetrators do what they do, criminologists are

often faced with a moral dilemma: to conduct research which avoids eliciting subject matter that is distressing

and complicated but, at the same time, has its limits with respect to the extent to which it can realistically hope

to challenge preconceptions about populations that are feared, despised or little understood; or to conduct

research that is prepared to engage with the distress of its subject population as a matter of course, less

sure what it means to secure informed consent from participants whose spontaneous disclosures may lead

in any direction, takes longer to digest the meaning of the data it produces, and more cautious in terms of

what it promises with respect to the generalisabilty of its findings. To the extent that psychosocial research is

offered in the service of rehumanising some of the most demonised sections of our populations, this second

approach remains for me, the more promising; but only if those of us engaged in it are prepared to separate

our data from the layers of interpretation we subject to it so that both can be interrogated critically by others. I

hope, therefore, that this chapter is a contribution to preparing those looking to acquire the skills and foresight

needed to embark on these tasks.