Content uploaded by Cynthia Cannon Poindexter
Author content
All content in this area was uploaded by Cynthia Cannon Poindexter
Content may be subject to copyright.
Qualitative Social Work
Vol. 1(1): 59–78 Copyright ©2002 Sage Publications London, Thousand Oaks, CA and New Delhi
1473-3250[200203]1:1;59–78;021745
Meaning from Methods
Re-presenting Narratives of an HIV-
affected
1
Caregiver
Cynthia Cannon Poindexter
Boston University, USA
ABSTRACT
This article outlines the methodological process followed in
examining a portion of an interview in which an older
woman tells of two incidents where she felt effects of
associative HIV-related stigma. Through the process of apply-
ing different techniques of narrative analysis, the author
learned research methods and deepened interpretations
of the text. Data management techniques both reflect
assumptions and augment understanding. In narrative
analysis, the structural whole can best be understood by first
examining the architectural detail. This report demonstrates
how meaning can emerge from method, theory from
transcription, and richness from rigor. It also argues for the
utility of narrative analysis to social work research, teaching,
and practice.
KEY WORDS:
caregiving
HIV/AIDS
methodology
narrative
ARTICLE
59
05 Poindexter (JG/d) 18/1/02 9:38 am Page 59
This article describes the process I followed as a beginning social work quali-
tative researcher in examining two stories about associative HIV-related stigma
in an interview with an HIV-affected caregiver.
2
A stepwise exercise involving
five technical approaches to narrative analysis not only expanded my knowledge
of the utility of various methods of data re-presentation, it led to different
interpretations and increasing understanding of the text. The purpose of this
article is to demonstrate how meaning can emerge from method, theory from
transcription, and richness from rigor in narrative studies. The potential utility
of these analyses to social work research, teaching, and micro and macro social
work practice is discussed.
Narrative researchers believe that storytelling is a naturally-occurring
form of talk which helps humans make sense of disruptive past events through
a cohesive presentation. Narrative is both process and result (Benner, 1994;
Mishler,1986a;Polkinghorne,1988;Riessman,1990).Stories (bounded accounts
with plot lines, characters, chronological sequence, and detailed reports of sig-
nificant episodes and turning points) are ubiquitous in research interviewing
(Mishler, 1986a). Narrative researchers believe that interviewers are likely to
elicit stories when they listen closely and do not over-structure the interview,
interrupt respondents unnecessarily, or suppress natural forms of expression
(Mishler, 1986a,b; Riessman, 1993). Methods of analyzing stories embedded in
interviews involve a focus on language, significance, and context rather than
standardized codes for content. To begin to examine stories, one chooses a
portion of ‘text’, an extended sequenced response of a research participant to
an interview question in which past events are recapitulated (Mishler, 1986a,b).
The overarching purpose of attending closely to stories is to more fully
understand what the speaker wished to convey through word choice, phrasing,
tone, pace, and emphasis as the account is performed for the listener. Speakers
are attempting to not only describe events, but their perceptions, conclusions,
and feelings about those events. ‘Experience’ is both a concrete incident and
how one interprets it. In other words, a detailed analysis of words and their
expression can more closely attune the researcher to the respondents’ intended
meanings.
An illustrative case from interviews conducted with an older HIV-affected
family caregiver demonstrates what various narrative methods contribute to
research analysis. Examination of a conversation with Ann (pseudonym), a
woman who had provided personal care to her adult son as he died in her home
from AIDS, led to a closer analysis of two juxtaposed bounded stories where
HIV-related stigma was significant (as per Stuhlmiller and Thorsen,1997). I use
these accounts, separated in structure yet connected in themes, to illustrate the
problems of transcription and treatment of qualitative data.
Method of transcription reflects one’s research goals and theoretical
perspectives. Ochs (1979) illustrated that decisions regarding transcription (the
60 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 60
primary mode of data re-presentation) both mirror and influence theory.
Mishler (1991) also views transcription of taped interviews as more than a
routine technical procedure; it is an interpretive process in itself, an important
decision regarding how to present and ‘re-present’ an interviewee’s story. Re-
presenting text is not merely a step on the way to real work; it is a phase of
the real work. Deciding on the form of transcription and re-presentation of
data is a methodological issue and a larger research problem because it is inter-
pretive. Methodological decisions lead to critical reflection on theory. Research
findings,of course,are shaped in some way by the researcher:‘Framing an indi-
vidual or an event; there are only choices, no faithful copies’ (Myerhoff, 1992:
296). In this article I demonstrate how the data,re-presented in various formats,
generated successive interpretations that I had not gleaned from previous models.
METHODS
The interview excerpt examined here is part of a much longer text: a taped
and transcribed interview with a white 69-year-old social worker. Ann took
care of her adult son Lars (pseudonym) as he died in her home from AIDS six
years before the 1998 study. She was part of a small convenience sample of
HIV-affected caregivers who had participated in two unstructured interviews
with me (see Poindexter, 2001, for the full report of this study). During data
collection I used frameworks of naturalistic inquiry (Lincoln and Guba, 1985)
and grounded theory (Charmaz, 1990; Glaser and Strauss, 1967; Strauss and
Corbin, 1990).
Returning to the transcripts to identify specific stories, I was drawn to
two brief back-to-back accounts explaining how Ann or Lars had experienced
HIV-related stigma in their lives. I was attracted to this section for methodo-
logical and theoretical reasons. I was intrigued by the fact that these clearly
bounded and juxtaposed stories appeared so cleanly and clearly in her account,
and I thought that they would be useful in learning narrative methods of analy-
sis. In addition, HIV stigma is a construct on which I have focused in previous
work. Although stigma and discrimination were not overarching or major
themes in Ann’s interviews,her energetic telling of these incidents was impres-
sive, and I wanted to understand them better.
Following Riessman’s (1993) method under her personal supervision, I
used a rough interview transcription to identify stories and their boundaries.
An extended narrative emerged, beginning with my summary concerning Ann
not being treated badly because of HIV, progressing through her accounts of
two incidents of stigma, and closing with: ‘Those are the only two instances,
everyone else was just wonderful.’ (See Figure 1 for full transcript of the stigma
stories and lead-in paragraph)
3
Story one concerns Lars’ experience with rela-
tives in Sweden and story two pertains to something that happened to Ann in
Poindexter Meaning from Methods ■ 61
05 Poindexter (JG/d) 18/1/02 9:38 am Page 61
Vermont. She differentiates between them as ‘not me’ and ‘me,’ indicating that
the first happened to Lars and the second to her personally.
After identifying the boundaries of the stigma stories, I transcribed this
section of text repeatedly, trying several models of analysis.The first re-presen-
tation – the usual rough transcription used in qualitative social work research
– focused only on verbal content, with no attention to interviewer input or
complexities in Ann’s utterances. The second level of transcription was based
on Labov’s (1982) research on the elements of a coherent story. Examining
responses to a question regarding violent traumatic injuries, Labov identified
key narrative structures: the abstract (summary); orientation (who, what, when,
where); complicating action (events, crises, problems); evaluation (significance
and meaning); result (outcome, resolution); and coda (relating the story to the
present). The third transcription used Gee’s (1985, 1986, 1991) approach to
studying the architecture of speech.Gee developed a structural presentation that
arranges text in poetic units,such as idea units,lines,stanzas,strophes,and parts.
Gee argues that researchers need to understand how people are making sense;
linguistic units,sequencing,pace,tone,and phrasing are significant for coherence
and congruence.The fourth incarnation of the stories was inspired by Mishler’s
(1986a) work on co-construction of narrative,which led to modifying the Gee
format to include interviewer contributions to Ann’s stories (i.e. paralinguistic
and linguistic cues). The final re-presentation consists of poetry crafted from
Ann’s words (Miles and Huberman, 1994: 110). At each stage of transcription,
more and different impressions surfaced regarding Ann’s experience of stigma.
MEANING FROM METHODS
Rather than discuss the content of these two stories (see Poindexter, unpub-
lished,for an analysis of her experience of stigma),here I examine only the five
transcription and re-presentation methods I used and how those treatments
influenced my interpretation and understanding of the data. Techniques are
discussed in the order in which I practiced them. At each point I recount how
insights occurred to me incrementally as each subsequent method took me
deeper into Ann’s world.
Rough Transcription is Limited to Content
The first transcription (Figure 1) presents the entirety of Ann’s stigma stories
and the preceding paragraph in a rough content-driven re-presentation. Previ-
ously Ann had characterized Lars and the family as fully disclosing the HIV
diagnosis and having no experience with discrimination. She had also said how
‘wonderfully’ people had treated Lars throughout his illness. When the stigma
stories were examined in traditional simple format (Figure 1),they appeared to
be triggered by my reflection of Ann’s experience with ostracism as reported
62 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 62
Poindexter Meaning from Methods ■ 63
INTRODUCTION
People were very loving to Lars. There was nobody who wasn’t. It was mainly
what I had to do. Lars didn’t want a lot people that he didn’t want. And I had
to be really the lion at the gate. And sometimes I tried to handle it so that people
didn’t feel hurt. He didn’t want people, for example, my younger brother, who
drinks a lot, and when he was drinking he made terrible homophobic remarks
and stuff and I didn’t want him coming, even though he was no longer drink-
ing but he had really hurt Lars’s feelings and I had to sort of say ‘well,he’s really
too ill to have a visitor, and he’d appreciate a letter’ and stuff like that. And he
monitored his own phone calls and stuff like that. He’d listen to who it was and
answer it if he wanted to, so that was a role, that I kind of checked with him
ahead of time, that was okay.
‘NOT ME:’THE SWEDEN STORY
[I: So you didn’t experience anybody treating you badly because of this?] Well,
no one I knew, with one exception, and that really wasn’t me, and that was a
different situation. My husband’s Swedish so his family lives in Sweden and are
very close, my sister-in-law is a doctor and we’ve always been close, we corres-
pond and what not,and she’s married to a doctor. Lars was going to,this is when
he was still well enough, but when he got very ill he couldn’t go there, but he
went over to see his Swedish cousins and wonderful Eric and his family and
whom he’d not met, and my brother-in-law, who’s this doctor, wrote me letter
saying that he would not allow Lars, that he’d talked to doctors and they would
not allow Lars to come and he’d talked to his children and there are 3 children
and only one of them was willing to have Lars after talking like that. And he
didn’t write Lars that, he didn’t tell his wife, he just wrote me. And I wrote a
letter back, I was furious of course, but I wrote back and said, something along
the lines that I do understand your fear and so forth, and so on. And of course
he picked up the irony in that and he was irate and wrote me back. And anyway,
when Lars arrived over there he was staying with Eric’s family, which is another
branch of the family, and he was very surprised because he wasn’t hearing from
the sister-in-law and Eric and everything, so, finally he was very curious, and he
asked me about it, and finally he did hear that they were meeting not at her
house but in town and all that, and she had found out of course and she was
very upset too. And she fortunately had a place in the country, and so she, and
I flew, Willis, my younger son and I flew over also, we all went to that place
together and we didn’t get to see the other family at all, and the other members
of the family didn’t want to see Lars but Rose,my sister-in-law got a letter from
Lars and she never told me what the contents was, but she said it was the most
continues
Figure 1 ROUGH CONTENT TRANSCRIPT
05 Poindexter (JG/d) 18/1/02 9:38 am Page 63
so far in the interview: ‘So you didn’t experience anybody treating you badly
because of this.’ I was responding to her earlier statement: ‘People were very
loving to Lars,I mean there was nobody who wasn’t’and had come to the con-
clusion, at this point in the interview, that Ann had no HIV stigma to report.
64 ■ Qualitative Social Work 1(1)
beautiful letter she’d ever had. So I know that he took it in stride. So, it’s a big
problem for me because I really don’t want to see that guy again, where he,he’s
crazy. [I: And he’s a physician.] He’s a physician, yeah. But he went beyond that
and he just began damning that whole family, really. So I keep up with her and
all, but I want to see her and I’m sure she’s getting old, she’s older than I am
though she’s probably healthier. So I’m really wracking my brain trying to think
how I can get back and see her without seeing him and so forth and so on. I
think I have to just be direct about it,and say that it’s just too uncomfortable for
me, can we meet somewhere that’s not uncomfortable. And I’m surprised that
he drew in other cousins, I mean Lars’s cousins, because you know, they’re nice
folks, just going along with what dad says I guess. The youngest boy that was
their problem child but he was very sweet to Lars the last time he saw him, so
that was one thing.
‘ME:’THE VERMONT STORY
The other incident that happened to me, was not someone who knew me, but I
went back, my father was a headmaster of a boys’ school in Vermont and I went
there for the 50th reunion of one of my cousins. But anyway, the minister there,
the school had a minister come for every morning service,and afterwards we were
walking back to have breakfast, it was early Sunday morning and I was with the
minister’s wife and she’s,she had,we were talking about ministry stuff because my
father’s in the ministry and my, her husband had known of him and stuff, and she
was saying they used to have a church down here in Massachusetts and they retired
up here in Vermont,and then she said out of the blue she said,‘and it’s getting so
crowded up here,I thought the AIDS epidemic would take care of that’,she said,
and you know. And we’d just gone to, just celebrated communion service and
talked about things to do with communion service and all that so I just said to
her,‘You know, what you just said offends me, I have a son with AIDS’ and she
was just taken aback. Fortunately for me, there was 2 of the people at the 50th
who were gay men, one of whom, both of whom had HIV, whom I naturally
gravitated to, hadn’t seen since forever and so forth. So I went and I said ‘I can’t
quite believe what I heard’ and I wanted someone to validate. [I: So you could
share it.] Yes, validate because it was just mind-blowing to me that this had hap-
pened. So you never know where that kind of stuff will crop up,you’ve probably
experienced some. [I: But it’s enraging every single time.] It is. [I: And hard to
explain.] So, those are the only 2 instances, everyone else was just wonderful.
05 Poindexter (JG/d) 18/1/02 9:38 am Page 64
From the rough transcript it appeared to me that she recounted these two stigma
incidents to correct my mistaken impression. It was clear that she was still some-
what upset by these events, although they had happened before her son’s death
six years prior to her relating them in this interview. It also was evident from
this transcription that her anger stemmed from her desire to protect Lars and
from her sense of outrage that HIV-positive persons suffered the effects of
stigma. I saw that she had arranged the stories by whether they had or had not
happened directly to her,but I did not have any idea about why she chose those
dichotomies in her answer to my probe. Before considering other ways of tran-
scribing and interpreting these excerpts, I treated – in presentations and an
article – Ann’s encounters with HIV stigma as a desire to live in a world of
social justice (see Poindexter, 2001).
Examining the stories in simple transcriptions was useful because it told
me that she had been affected by HIV stigma and that she fought against it.
However, this method was not as evocative in illuminating the complexities of
the respondents’ experience as other models of analysis later proved to be. At
the time that I was using content-only transcriptions,I was unaware of the limi-
tations of this method.
Labov’s Functional Model Highlights Evaluation
The transcription fragment from story two, shown in Figure 2, is based on
Labov’s principle that stories consist of distinct parts with unique functions.
Through the process of truncating the clauses into abstracts, orientation, com-
plicating action, evaluation, result, and coda, it became evident that the Labov
model was not suitable for Ann’s complex stories,which do not progress orderly
through those elements but which jump back and forth between them. This
realization helped me to understand the limits of the model. However, using
Labov’s method was helpful at this point in the research process because it
illustrated that Ann’s stories are complicated.They vacillate between action and
orientation and between action and evaluation; much of the action is internal;
and evaluative statements are peppered throughout. I saw that it is not always
the external events that are important in a story; it is the narrator’s interpre-
tation that makes the story salient. Ann’s thoughts and judgments, like morals
of a story, teach why the experience of stigma was significant. Examples are
line 18, the end of the Vermont incident:‘it was just mind-blowing to me that
this had happened’ and line 19, which could be the moral of both stories: ‘So
you never know where that kind of stuff will crop up.’The evaluations contained
in those lines were so striking to me at the time that I chimed in to agree with
her judgments. When she followed her opinion with ‘You’ve probably
experienced some-’she broke off when I interrupted with an evaluation of my
own: ‘But it’s enraging every single time,and hard to explain.’She affirms with
‘Yeah. It is.’
Poindexter Meaning from Methods ■ 65
05 Poindexter (JG/d) 18/1/02 9:38 am Page 65
The primary advantage of using Labov’s labels was that I realized that
she was passing judgment on these two stigma-related events. One drawback
of Labov’s model is that not all storytellers construct their accounts in a tight
sequence. Another disadvantage is that the researcher’s utterances are not
included in the transcription, so I did not realize that we were partners in con-
structing the moral of the story until I moved to Mishler’s model.
Gee’s Structural Model Uncovers Emphasis
Figure 3 shows a portion of the Sweden story in a stanza format based on James
Gee’s work. In order to present data in Gee’s arrangement, one cannot merely
regroup a simple transcription into idea units,lines,and stanzas; one must listen
carefully to the tape numerous times to capture emphasis,breathing,pitch glides,
and pauses.When I began using Gee’s model, I began to discover complexities
in Ann’s narrative that had not been evident through a rough transcription or
Labov’s techniques. Listening repeatedly and closely to the tape showed that my
emphasis of the word ‘you’in the summative comment in line 1 (‘So
YOU didn’t
experience anybody treating
YOU BADLY because of this’) had probably prompted
the categorization of the stories according to whether they had happened to
Lars or her: ‘not me’ and ‘me.’(I meant ‘you’ there as plural to refer to the care
dyad, but Ann evidently heard it as singular.) If the text had been left in the
traditional transcription format, my linguistic emphasis and how it influenced
Ann’s telling would have remained hidden to me.
The Gee treatment of the text led me to the realization of how often
Ann used personal pronouns emphatically. An example of that occurs in lines
66 ■ Qualitative Social Work 1(1)
Fortunately for me,
there was, 2 of the people at the 50th who were gay men,
one of whom, both of whom had HIV,
and whom I naturally gravitated to,
I hadn’t seen them since they were school boys.
So I went and I said ‘I can’t quite believe what I heard’
and I wanted someone to validate . . . [I: So you could share it.]
Yes, validate
because it was just mind-blowing to me that this had happened.
So you never know where that kind of stuff will crop up,
you’ve probably experienced some – [I: But it’s enraging every single time.]
Yeah. It is. [I: And hard to explain.]
Figure 2 LABOV FORMAT, EVALUATION
05 Poindexter (JG/d) 18/1/02 9:38 am Page 66
Poindexter Meaning from Methods ■ 67
[I: So YOU didn’t experience anybody treating YOU BADLY because of this.]
Well, no, no one that I
KNEW, / exc – with one exception,
and that really wasn’t
ME, / but, my, {p} and that was, that was a DIFFERENT
situation, that, um –
{p} My, my
HUSBAND was Swedish/ they’re Swedish/ and so his family lives in
Sweden/
and I’m very, very close to my sister-in-law/ who’s a doctor,
and, um, we’ve always been close, /we correspond and what not, but –
and
SHE’
S
married to a doctor =
and {p} Lars was going to go, /this is when he was still
WELL enough,
but remember he got very ill/ and he couldn’t go there,
but, but, but he went over to see his Swedish cousins/ including wonderful Eric
and
HIS family/ and whom he’d not met,
and um {p} my brother-in-law who’s this doctor/ wrote
ME a letter saying that
he would not allow Lars/
he’d talked to doctors/ and they would not allow Lars to come
AND he had talked to his CHILDREN /and there are 3 children /and {p} only one
of them was willing to have Lars after talking like that.
And um {p} he didn’t write
LARS that,
he didn’t tell his
WIFE {p}
he just wrote
ME.
And I {small giggle in voice} wrote a letter back and said – / I was
FURIOUS of
course, / but I wrote back
and said um {p} something along the lines that I understand your –,I
DO under-
stand your
FEAR and so forth, and so on.
And of course he picked up my,the
IRONY in that/ and he,he just launched into
a
TIRADE and wrote me back and so –
GEE TRANSCRIPTION KEY
CAPS Vocal emphasis
? Rising intonation or pitch glide
. Falling intonation or pitch glide
1. Utterance broken off
= Successive utterances with no gap
[ ] Interviewer utterances
{ } Author explanations
{p} Short pause
{P} Longer pause
/ Separates idea units, or phrase with one pitch glide
hard return Indicates a line, or one topic
blank line Separates stanzas, or paragraphs
Figure 3 GEE FORMAT
05 Poindexter (JG/d) 18/1/02 9:38 am Page 67
14 and 21 when she says that her brother-in-law wrote ‘ME,’ emphasizing that
he confronted her rather than telling his wife or Lars. There followed a snow-
balling of insight. Once my attention had been drawn to Ann’s use of pro-
nouns to contrast,I realized that throughout these stories she not only separated
the incidents as happening to ‘me’ or ‘not me,’ she divided people into ‘them’
and ‘us’: those who are HIV-affected and/or accepting, and those who mistreat
persons with HIV and their loved ones. Noticing this use of tensions prompted
the discovery of other distinctions in these excerpts: strangers versus familiar
people, safe versus unsafe people,ignorant persons versus people who ought to
know better, being prepared versus unprepared to defend against stigma, and
her response versus Lars’response. Furthermore,the slowing of the tape to listen
closely to her speech provided clues about what was most important in these
accounts. I became aware for the first time,due to Gee’s accentuation of pitch
glides and pace, that Ann talked rapidly and with few pauses, often mumbling;
changed speed and tone, depending on the topic. She used an urgent rate of
speech to build emphasis or to make a point explicit, and a halting delivery to
show hesitancy or tentativeness.
In addition, my parsing the text into stanzas helped to delineate Ann’s
topical and situational shifts,such as background on the Swedish relatives,expla-
nation of Lars’ desire to go, rejection by the uncle, and the outcome. Her
accounts had seemed jumbled to me previously, but with the stanza format I
saw the structure of her speech. The Gee transcription also showed that she
spoke in fragments,sometimes breaking off her own phrases,assuming I under-
stood what was unspoken. I had previously speculated that our shared roles of
social workers and caregivers might have contributed to her trust that I would
follow her thoughts even if they were not fully expressed; however, not until I
noticed my own utterances in the next phase of transcription did I have a clear
idea about why she would sometimes not complete sentences.
The strength of Gee’s methods in this project was that it led me to dis-
cover the subtle differences in emphasis, word usage, and structure. However,
Gee’s model shares a weakness with Labov’s: there is no inclusion of the
researcher’s part in the construction of the story.
Gee’s Model Modified with Mishler’s Highlights Co-construction
Figure 4 is a transcription of the introductory text which appears before the
stigma stories. It augments Gee’s stanza analysis according to Mishler’s work on
joint construction in narrative and significance of researcher utterances.
Although I had used Gee’s model to transcribe my questions and probes, I had
not been aware of the frequency of my encouraging and affirming utterances
threaded throughout Ann’s telling. Modification of Gee’s model occurred as
a result of a meeting of a narrative studies group, facilitated by Mishler.
While listening to portions of the audiotaped interview as they read the Gee
68 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 68
transcription, colleagues were struck by the frequency, intensity, and emphasis
of my utterances and comments. Following this experience I produced a Gee
transcription with my paralinguistic cues included. Only then did my partici-
pation in the interview become evident to me. The awareness of the co-
construction of Ann’s narration changed my view of her stories. It became clear
that I was signaling Ann that I was attending and thought I understood what
she wished to convey. The effects of this are evident from what was not said:
she often did not elaborate and at times did not complete her sentences. I saw
Poindexter Meaning from Methods ■ 69
I mean, people were v- were VERY loving to Lars =
I mean there was nobody who wasn’t =
and it was
MAINLY that I/ what I had to do really was that Lars did not want/
{p} a lot people that he didn’t
WANT = [I: yeah]
I mean, so, and I had to be really the lion at the gate [I: umhum]
and sometimes {p} I mean try, try to handle it/ so that people didn’t feel hurt,
he didn’t
WANT people – [I: umhum]
for example, my
YOUNGER brother/, who used to drink a lot,
um,and when he was drinking he made
TERRIBLE homophobic remarks and stuff
and, {p} and I didn’t want him coming,
um. even though, um he no longer was
DRINKING/ and, uh, but he had really
hurt Lars’s
FEELINGS [I: yes]
and, you know, so um, uh, I had to first say well, I thi-/ he’s really too
ILL to, to
have a visitor/ and I think he’d appreciate a long
LETTER and stuff like that.
[I: umhum]
GEE TRANSCRIPTION KEY
CAPS Vocal emphasis
? Rising intonation or pitch glide
. Falling intonation or pitch glide
1. Utterance broken off
= Successive utterances with no gap
[ ] Interviewer utterances
{ } Author explanations
{p} Short pause
{P} Longer pause
/ Separates idea units, or phrase with one pitch glide
hard return Indicates a line, or one topic
blank line Separates stanzas, or paragraphs
Figure 4 GEE FORMAT WITH MISHLER’S CO-CONSTRUCTION
05 Poindexter (JG/d) 18/1/02 9:38 am Page 69
that Ann’s lack of explanation was due in part to my pre-emptive ‘umhum’s’.
An example is line 7 when she says ‘he didn’t
WANT people-’and broke off then
at my ‘umhum,’ not saying what Lars didn’t want. I had anticipated that the end
of that sentence would be that he did not want people visiting him, because I
have experienced how people who are very ill only want to see whom they
want to see. Without hearing my interpretations, others who read Ann’s tran-
script thought that she was not completing thoughts. I, however, was during
the interviews completing her thoughts in my mind and communicating that
fact to her through my vocal responses. I had no explanation of why Ann
seemed to interrupt herself until I used this model and heard and saw myself
interrupting her. I felt that adding Mishler’s orientation to co-construction
greatly enhanced the usefulness of Gee’s model.
Experimental Poetry Evokes Emotion
Working with Gee’s method of dividing utterances into lines and stanzas led
me to a final step: crafting poems from Ann’s stories. I had previously experi-
mented with this form of re-presenting research data (see Poindexter,1997a, b,
c) before I had become familiar with Gee’s belief that natural storytelling occurs
in poetic structures, read the work of scholars who were crafting poetry from
the words of research interviewees (Richardson, 1992, 1994), or realized that
some anthropologists were writing poetry about their fieldwork (Prattis, 1985;
Richardson, 1998). I returned to this method of data re-presentation because
of my own emotional connection with the material.The poems seemed to me
to be embedded in the stories, just as the stories were embedded in the inter-
view, and when I extracted them, I felt a deeper sense of empathy and reson-
ance with Ann’s experiences.
The poems in Figure 5 are not traditional transcriptions; they are
rearrangements of Ann’s words into a literary form.The poems set Ann’s experi-
ences into a coherent,abridged form, acknowledging her expression and words
while honoring the morals of her stories and the power of her experience.
Richardson (1997) characterizes poems stemming from research transcripts as
mini-narratives. Indeed,this treatment of text seems to boil her stories into core
meanings, rendering their essence. ‘The letters,’ taken from the Sweden story,
highlights her repetition of ‘write’ and ‘wrote,’ as well as making more vital the
flurry of letters that ensued. The poem entitled ‘Stuff will crop up’ pares down
the story into the evaluation of the Vermont incident, spotlighting Ann’s shock
and judgment.
I have several times used poetry crafted from research interviews to end
presentations and workshops, as well as in conference poster sessions, and have
found that listeners and readers tend to be moved by their simplicity and power.
As Richardson (1993) states, the intent of the research poem is both aesthetic
and empathic;it can communicate the respondent’s emotional world effectively
70 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 70
and efficiently. In developing a poem, the researcher selects talk and reforms it
into a nontraditional re-presentation. The resulting poem may bring points to
the fore,clarify and make the account more compelling,create a different effect,
and engage the reader and listener, and tell us something about lived experi-
ence which we did not previously understand.
Qualitative researchers struggling to use their work for advocacy and for
Poindexter Meaning from Methods ■ 71
THE LETTERS
My brother-in-law
wrote me a letter saying
that he would not allow Lars to come.
He didn’t write Lars that
he didn’t tell his wife,
he just wrote me.
I was furious of course –
I wrote a letter back.
He just launched into a tirade
and wrote me back.
We didn’t get to see this family at all –
the other members of the family didn’t want to see Lars.
But my sister-in-law says that she got a letter from Lars –
it was the most beautiful letter she’d ever had.
STUFF WILL CROP UP
The minister’s wife said out of the blue
‘it’s getting so crowded up here,
I thought the AIDS epidemic would take care of that’
I just said to her,
‘what you said to me offends me –
I have a son with AIDS.’
I can’t quite believe what I heard –
It was just mind-blowing to me
that this had happened.
You never know where that kind of stuff will crop up.
Figure 5 POETRY
05 Poindexter (JG/d) 18/1/02 9:38 am Page 71
raising public awareness may want to experiment with poetic re-presentations.
The disadvantage of using research poetry is that its usefulness as a form of data
re-presentation is debated and controversial. The advantage is that core narra-
tives and strong emotions can be communicated with an economy of words.
IMPLICATIONS
From this progressive transcription exercise it became evident that different
conclusions can be reached through various techniques of re-presenting data,
linking method to meaning and adding rigor to narrative analysis. My
interpretation of Ann’s intended messages changed and expanded depending
on the transcription method. It is, of course, impossible to guess whether I
would have gained the same insights if I had proceeded in a different order of
methods,or if I had immediately adopted a Gee/Mishler approach. In this case
the iterative process produced for me incremental insights. From the rough
transcription I harvested sequence,content,and conclusions,which is how most
social work research reports are done and which heretofore had been the only
method I had used. However, the fuller significance is not completely access-
ible from a rough transcription. It takes other methods to uncover deeper and
broader interpretations. Expanding analysis through examining story elements
as in Labov’s model, vocal emphasis and stanza structure as in Gee’s, co-
construction as in Mishler’s,and core messages as in poetry accumulated under-
standing of Ann’s experiences with HIV stigma.
I am not recommending that researchers use all of these transcriptions
methods for all projects.I am suggesting that it is perhaps worthwhile to experi-
ment with ways to re-present data in order to find one which fits the researcher’s
philosophy and objectives. I now use a rough transcription only when I wish
to report pure content excerpted from a long interview or several interviews.
I may borrow from Labov’s model to identify summary,orientation,action, and
evaluation statements in a stretch of talk.For examining distinct bounded stories,
I now prefer a Gee stanza format with interviewer utterances included, as per
Mishler, as demonstrated in Figure 4. To unpack these particular stigma stories
from Ann’s interview,I used this more complete,more complex hybrid method
that allowed me to exhibit the poetic structure while recognizing co-construc-
tion (see Poindexter,unpublished).I sometimes use poetry formed from respon-
dents’ words in classes and training sessions, making the core narrative and the
eloquence of expression clear while having an emotional impact on the audi-
ence and thus helping listeners to remember the respondent’s message and
experience.
Narration and storytelling are not new notions. Telling tales has been a
dominant form of human communication since the dawn of language. Listen-
ing to stories to reach understanding is also not new. Anthropologists have
72 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 72
documented folk tales and oral history by spending time with tribal chiefs;
psychotherapists hear dreams and events in narrative to understand inner life;
theologians dissect ancient stories to find messages and lessons; gerontological
researchers document life review tales of older persons; historians gather data
by interviewing people who experienced past events and periods; and sociolo-
gists study norms by listening for the meta-narratives which tellers may take for
granted. Social workers, however, have not paid close attention to the import-
ance of language, form, and expression from persons whose voices have been
muted and who feel ready or even compelled to tell their stories. Narrative
study could illuminate and augment social work research,teaching,and practice.
Implications for Research
Making mistakes in hearing or transcribing is one of three validity errors in
qualitative methods (Easton et al., 2000). Narrative analysis, a systematic way of
listening to interviews and re-presenting qualitative data, is one way of bring-
ing greater rigor to qualitative research because it highlights the interviewer’s
co-construction of the data, facilitates a more detailed transcript, and deepens
the analysis. Investigators need to allow stories in various forms to emerge in
interviews, because of the richness of information and understanding that can
be harvested from storytelling as data. In order for researchers to receive and
hear individual stories, the stories must be invited, encouraged, and welcomed,
and research respondents must be treated as equal partners in constructing reality
(Chase,1995). If participatory action research is our goal,then listening closely
to stories must be the first step.Furthermore,examining the interviews in greater
detail can reveal the interviewer’s assumptions and blind spots.In order to ensure
that we as researchers are interpreting others’ lives in the most ethical and
respectful way possible, we can strive to re-present text as accurately as is feas-
ible and to focus on the respondent’s own expressions and perspectives. To be
more true to the speaker whose reality we wish to understand, we must tran-
scribe text purposefully, carefully, as completely as possible, and clearly reveal
the theoretical underpinnings of transcription and re-presentational decisions.
Scholars who are striving to glean theoretical and experiential meanings from
interviews can shift paradigms and perspectives to increase the yield from a
research interview.
Implications for Teaching and Field Instruction
Stories and narrative analysis are useful in a variety of educational situations:
classrooms, field settings, agencies, and communities. I have invited Ann to
present her caregiving experiences on a conference panel, and have used por-
tions of her interview in continuing education workshops and in an HIV class.
Putting a face on an overwhelming social problem such as HIV stigma helps
students and colleagues develop empathy,a prerequisite for effective social work.
Educators can present written case studies taken from research and clinical
Poindexter Meaning from Methods ■ 73
05 Poindexter (JG/d) 18/1/02 9:38 am Page 73
interviews,invite guest speakers to tell their stories, play taped interviews, treat
process recordings as a form of narrative, or present stories which have arisen
from inquiries and practice. Teachers discourage learners from drawing con-
clusions from scant data,and that is as it should be.However,new social workers
without much field experience can be assisted in development of engagement
and insight if individual stories are included in their curriculum and learners
are encouraged to interpret them carefully. Practitioners already in the field also
respond to stories as illustrations of theory and interventions.
Students of research can be taught to listen for stories in interviews,tran-
scribe them carefully, and present the same text through various techniques to
discover the best rendering of the data. Teachers of research should be well
versed in various transcription techniques so that they can encourage learners
to experiment and explore. Transcripts in various formats can be presented in
classrooms to teach methods of data presentation.Studying and applying various
theorists in transcribing data contributed greatly to my own learning as a prac-
titioner-researcher. Not only did I learn research methods by re-transcribing
and re-presenting interview excerpts, I learned more about the stigma experi-
ences of HIV caregivers.
Implications for Counseling
In therapy, counseling, crisis intervention, and case management, practitioners
can consider taping and analyzing individual, family, and group sessions and
examine them for stories, looking for the intersection of content and form and
what it can reveal. Mattingly (1991, 1998) and White and Epston (1990) show
counselors how to enter life stories by listening closely to the relationship of
individual stories to ‘master narratives’of culture,and then using the therapeutic
partnership to help re-author a person’s interpretation into a more hopeful and
self-affirming story.Storytelling can be the method through which a practitioner
helps someone craft a new life story with more empowering themes (Capps
and Ochs, 1995). Counselors can tape and listen to sessions; pay attention to
words, structure, and paralinguistic features; point out patterns; and work on
making dominant the sense of control and hope. The clinician can coauthor
significance and conclusions by witnessing revisions and extensions of the
person’s story, acknowledging and celebrating the rewritten meta-tale. A
counselor working with Ann, for example, might help her to see her desire to
protect her son as a strength, gently reframe her failure to predict these events
as normal expectations for nondiscrimination, and acknowledge her successful
resistance to stigma.
Implications for Policy Development and Community Organizing
Although one person’s story cannot exemplify everyone’s, honoring one story
at a time as indicative of possible general realities is an important source of
insight. Managers and policy makers can aggregate stories that arise in research
74 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 74
studies and counseling sessions (individual, family, and group) to begin to get
a clearer picture of how HIV affects caregivers. Hidden, oppressed, or dis-
enfranchised individuals and populations should be supported in raising con-
cerns in their own voices. Richardson (1997) suggests that by attending to
personal stories, scholars can help ‘empower individuals, contribute to liberat-
ing civic discourse, and support transformative social projects’ (p. 33).
Researchers can through narrative studies help to raise voices of persons silenced
or oppressed. When an individual who is disenfranchised or hidden reads
another’s story, that person can feel validated and less isolated, part of a shared
consciousness and an alternative collective story,which can move that individual
toward self efficacy, social action, and ultimately to societal transformation. In
that respect,highlighting stories can produce an alternative to oppressive master
narratives, a ‘collective story’ or ‘liberation narrative’ (Richardson, 1997: 34) that
empowers. Ann’s individual story is part of the collective story of older HIV-
affected caregivers exposed to stigma in various ways,and finding forms of resist-
ance and ways to fight back.
In conclusion, the craft of re-presenting data influences theoretical
insights and substantive conclusions, deepening the theoretical and experiential
meaning researchers can reap from interviews. The sequential transcriptions
created a continuum of insights. Listening to stories requires really hearing them
and the way in which they are told. Transcription is not a mechanical event,
but an important step in the process of ‘meaning-making’ and co-construction
of interpretive findings. Data management techniques both reflect our assump-
tions and deepen understanding. It is best to vacillate between detail and themes,
because the larger narrative can best be understood by first attending to the lin-
guistic form, utterances, pauses, and breaths. From the architectural fragments
we can extrapolate the narrator’s overall vision. Pairing facts with the narrator’s
evaluation enlightens us. The careful transcription and analysis which narrative
methods require can help us hone our social work research, education, and
interventions.
Acknowledgements
The 1998 recruitment and interview process was funded by a faculty development grant
through the Boston University School of Social Work.The subsequent study and appli-
cation of narrative methods in 2000 was funded by the John A. Hartford Foundation’s
Geriatric Social Work Faculty Scholars Program.The author thanks her narrative studies
mentors: Dr Elliot Mishler of Harvard Medical School Department of Psychiatry and
Dr Catherine Kohler Riessman of Boston University School of Social Work and Depart-
ment of Sociology. She also acknowledges the helpful comments of the members of
the Hartford Scholar’s Program and thanks Professor Trudy Duffy of the Boston Uni-
versity School of Social Work for helpful editing. She is also indebted to Dr Barbara
Berkman of Columbia University School of Social Work and Dr Laura Robbins of
Poindexter Meaning from Methods ■ 75
05 Poindexter (JG/d) 18/1/02 9:38 am Page 75
Cornell Medical School for their encouragement to write about this methodological
dilemma. Nancy Capobianco Boyer of Boston University School of Social Work pro-
vided valuable technical support, without which this analysis would not have been
possible.
Notes
1 ‘HIV-affected’ refers to family members who have responsibilities of caregiving for
an adult or child who has HIV Disease, the spectrum of illnesses caused by the
Human Immunodeficiency Virus, the end stage of which is often called ‘AIDS’
(Acquired Immune Deficiency Syndrome).
2 Goffman (1959) defined stigma as pejorative labeling and defined associative stigma
as stigma applied to associates of the stigma carrier. Herek and Glunt (1988) coined
‘AIDS-related stigma’ to describe the specific intense linking of stigma to persons
with HIV and their associates. Because this phenomenon occurs across the spec-
trum of HIV disease and does not apply only to receiving an AIDS diagnosis, I call
it ‘HIV-related stigma’ or ‘HIV stigma’ throughout this article.
3
Full transcripts of the stigma stories in all Labov and Gee formats are available from
the author.
References
Benner, P. (1994) ‘The Tradition and Skill of Interpretive Phenomenology in Studying
Health, Illness, and Caring Practices’, in P. Benner (ed.) Interpretive Phenomenology:
Embodiment,Caring,and Ethics in Health and Illness,pp. 99–127.Thousand Oaks,CA:
Sage.
Capps,L. and Ochs,E. (1995) Constructing Panic:The Discourse of Agoraphobia.Cambridge,
MA: Harvard University Press.
Charmaz, K. (1990) ‘“Discovering” Chronic Illness: Using Grounded Theory’, Social
Science and Medicine 30(11): 1161–72.
Chase,S. E. (1995) ‘Taking Narrative Seriously: Consequences for Method and Theory
in Interview Studies’, in R. Josselson and A. Leiblich (eds) Interpreting Experience:
The Narrative Study of Lives,Vol. 3, pp. 1–26. Newbury Park, CA: Sage.
Easton,K. L., McComish, J. F. and Greenberg,R. (2000) ‘Avoiding Common Pitfalls in
Qualitative Data Collection and Transcription’, Qualitative Health Research 10(5):
703–7.
Gee, J. (1985) ‘The Narrativization of Experience in the Oral Style’,Journal of Education
167(1): 9–35.
Gee, J. (1986) ‘Units in the Production of Narrative Discourse’, Discourse Processes 9(4):
391–422.
Gee, J. (1991) ‘A Linguistic Approach to Narrative’, Journal of Narrative and Life History
1(1): 15–39.
Glaser, B. and Strauss, A. (1967) The Discovery of Grounded Theory. Chicago, IL: Aldine.
Goffman, E. (1959) The Presentation of Self in Everyday Life. Englewood Cliffs, NJ:
Prentice-Hall.
Herek, G. M. and Glunt, E. K. (1988) ‘An Epidemic of Stigma: Public Reactions to
AIDS’, American Psychologist 43(11): 886–91.
76 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 76
Labov, W. (1982) ‘Speech Actions and Reactions in Personal Narrative’, in D. Tannen
(ed.) Analyzing Discourse:Text and Talk. Washington, DC: Georgetown University
Press.
Lincoln,Y. and Guba, E. (1985) Naturalistic Inquiry. Beverly Hills, CA: Sage.
Mattingly,C. (1991) ‘The Narrative Nature of Clinical Reasoning’,The American Journal
of Occupational Therapy 45(11): 998–1005.
Mattingly, C. (1998) Healing Dramas and Clinical Plots:The Narrative Structure of Experi-
ence. Cambridge: Cambridge University Press.
Miles, M. B. and Huberman A. M. (1994) Qualitative Data Analysis,2nd edn. Newbury
Park, CA: Sage.
Mishler, E. G. (1986a) Research Interviewing: Context and Narrative. Cambridge, MA:
Harvard University Press.
Mishler,E.G.(1986b) ‘The Analysis of Interview-narratives’,in T.R.Sarbin (ed.) Narra-
tive Psychology:The Storied Nature of Human Conduct, pp. 233–55. New York: Praeger.
Mishler,E. G. (1991) ‘Re-presenting Discourse: The Rhetoric of Transcription’,Journal
of Narrative and Life History 1(4): 255–80.
Myerhoff, B. (1992) Remembered Lives:The Work of Ritual, Storytelling,and Growing Older.
Ann Arbor, MI: University of Michigan Press.
Ochs,E. (1979) ‘Transcription as Theory’,in E. Ochs and B. B.Schieffelin (eds) Develop-
mental Pragmatics, pp. 43–72. New York: Academic Press.
Poindexter, C. (1997a) ‘Stigma and Support as Experienced by HIV-affected Older
Minority Caregivers’, unpublished dissertation, University of Illinois at Chicago.
Poindexter, C. (1997b) ‘Poetry as Data Analysis: Listening to the Narratives of Older
Minority HIV-affected Caregivers’, Reflections: Narratives of Professional Helping 4(3):
22–5.
Poindexter, C. (1997c) ‘Ms. Carol’ and ‘Ms. Dorothy (Research Poetry)’, Affilia: Journal
of Women and Social Work (Summer): 486–9.
Poindexter,C. (2001) ‘“I’m Still Blessed”: The Assets and Needs of HIV-affected Care-
givers over Fifty’, Families in Society 82(5): 525–36.
Poindexter, C. (unpublished) ‘The Lion at the Gate: An HIV-affected Elder Caregiver
Resists Stigma’, unpublished paper (copies can be obtained from author, email:
ccp@bu.edu).
Polkinghorne, D. E. (1988) Narrative Knowing and the Human Sciences. New York: State
University of New York Press.
Prattis, J. I. (ed.) (1985) Reflections:The Anthropological Muse. Washington, DC: Ameri-
can Anthropological Association.
Richardson.L.(1992) ‘The Consequences of Poetic Representation:Writing the Other,
Rewriting the Self’, in C. Ellis, and M. G. Flaherty (eds) Investigating Subjectivity:
Research on Lived Experience, pp. 125–37. Newbury Park, CA: Sage.
Richardson, L. (1993) ‘Poetics, Dramatics, and Transgressive Validity: The Case of the
Skipped Line’, The Sociological Quarterly 34(4): 695–710.
Richardson, L. (1994) ‘Nine Poems: Marriage and the Family’, Journal of Contemporary
Ethnography 23(1): 3–13.
Richardson, L. (1997) Fields of Play: Constructing an Academic Life. New Brunswick, NJ:
Rutgers University Press.
Poindexter Meaning from Methods ■ 77
05 Poindexter (JG/d) 18/1/02 9:38 am Page 77
Richardson, M. (1998) ‘Poetics in the Field and on the Page’, Qualitative Inquiry 4(4):
451–62.
Riessman, C. K. (1990) ‘Strategic Uses of Narrative in the Presentation of Self and
Illness: A Research Note’, Social Science Medicine 30(11): 1195–200.
Riessman, C. K. (1993) Narrative Analysis. Newbury Park, CA: Sage.
Strauss, A. and Corbin, J. (1990) Basics of Qualitative Research. Newbury Park,CA: Sage.
Stuhlmiller, C. M. and Thorsen, R. (1997) ‘Narrative Picturing: A New Strategy for
Qualitative Data Collection’, Qualitative Health Research 7(1): 140–9.
White,M. and Epston,D. (1990) Narrative Means to Therapeutic Ends. New York: Norton.
Cynthia Cannon Poindexter, MSW, PhD, is an Assistant Professor at the
Boston University Graduate School of Social Work (BUSSW). Her current
research concerns older HIV-related caregivers,an HIV training evaluation,and
participation of HIV-infected volunteers and employees in the AIDS service
system. Address: Boston University School of Social Work, 264 Bay State Rd,
Boston, MA 02215, USA. [email: ccp@bu.edu]
78 ■ Qualitative Social Work 1(1)
05 Poindexter (JG/d) 18/1/02 9:38 am Page 78
