Focus Group Interviews: A New Feminist Method
FRANCES MONTELL
University of California, Santa Barbara
This paper shows how feminist qualitative research can be strengthened
and broadened through the development of feminist focus group inter-
views. The interaction among participants in group interviews provides
a valuable resource for studying issues of gender and sexuality. In addi-
tion, focus groups can be both consciousness-raising and empowering for
the research subjects and for the researcher herself, and allow for a more
egalitarian and less exploitative dynamic than other methods. Drawing
on the author's study of the influence of popular culture on the construc-
tion and maintenance of "normative sexuality" for women, this paper
addresses some of the practical and ethical issues that a researcher must
confront when using group interviews. The author addresses different
concerns in recruiting participants and making the most of the unique
dynamics of a group interview and the kinds of data they produce for
analysis. More than most other methods, group interviews provide femi-
nists vnth the opportunity to conduct research that is consciousness-
raising and empowering, research that does not merely describe what is,
but that participates in shaping what could be.
Introduction
Focus groups have been a standard tool in marketing research for
several decades. While social scientists long ignored them, there has been
a recent surge of interest in their social scientific application (Krueger
1988; Lunt and Livingstone 1996; Morgan and Spanish 1984; Morgan
1988, 1993, 1996). Feminist researchers have occasionally used group
interviews (Callahan 1983; Malhotra 1984; Mies 1983; Nichols-Casebolt
and Spakes 1995), yet few provide detailed discussions of group interviews
from a feminist perspective. In this article I describe how I came to use
focus group interviews and why they are particularly useful for studying
issues of gender and sexuality.
Feminist qualitative research can be strengthened and broadened
through the development of feminist focus group interviews. Group inter-
views allow for a more egalitarian and less exploitative dynamic than
other methods, and the interaction among participants produces a new
and valuable kind of data. In addition, focus groups can be both conscious-
ness-raising and empowering for the research subjects, as well as the
researcher. In an effort to provide practical assistance to researchers at-
1999 NWSA fouRNAL, VOL. 11 No. 1 (SPRING)
Focus GROUP INTERVIEWS 45
tempting to use this method, I also describe issues involved in recruiting
participants and in maximizing the unique dynamics of and the data from
a group interview.
The Available Literature on Focus Groups
Although use of focus groups has mushroomed among social scientists
in the past decade, relatively few scholars specifically address the use of
this method in social science research.' Many academic researchers who
have used group interviews have done so simply because they found it
more convenient to interview several people at once (Morgan 1988, 12).
There is a widespread misconception that group interviews are an easy
"shortcut" to data collection (Agar and MacDonald 1995, 85; and Lunt
and Livingstone 1996, 80), a way to get more data faster than conducting
several individual interviews. In such cases, the researcher is generally
not interested in the unique characteristics of focus group interviews;
consequently this method has not been given systematic attention (Mor-
gan 1988, 12). Researchers who approach focus groups as simply several
ethnographic interviews in one may he disappointed and overlook the
unique and useful data that focus groups can produce (Agar and Mac-
Donald 1995, 85).
The majority of the literature on the use of focus groups comes from the
field of marketing research; however, the wide acceptance of the method
in applied marketing has not carried over to academic marketing research
(Morgan 1996, 132). Despite increased interest in focus groups among
social scientists, researchers still see the need for "considerable borrow-
ing and considerable innovation" (Morgan 1988, 10) and "careful adapta-
tion" to make this marketing tool "appropriate for academic research"
(Holbrook and Jackson 1996, 136).
Academic use of focus groups challenges the model that has developed
in marketing research, including assumptions about the best ways to
select participants, compose the groups, deal with potential problems of
conformity or "groupthink," and use the data (Holbrook and Jackson
1996, 136-7). Unlike the applied marketing researchers, the academic
researcher who uses focus groups is interested in the kinds of data pro-
duced by in-depth interviews, as well as the process of negotiation among
participants. The academic researcher must elicit and analyze group in-
teraction using discourse analysis, the way language structures common
sense beliefs (a level of analysis not found in applied marketing research).
Although the marketing literature on focus groups is in many cases
very detailed, it contains little information on academic or feminist
methodological issues.^ Few feminist researchers who have used group
interviews (e.g., Callahan 1983; Malhotra 1984; Mies 1983) have provided
46 FRANCES MONTELL
in-depth accounts of this method. Shulamit Reinharz's Feminist Methods
in Social Research (1992), the most comprehensive volume on feminist
methods, devotes only tw^o paragraphs to feminist group interviews (222-
3). These paragraphs contain only a very basic description of w^hat a focus
group is and v\^hy one feminist researcher (Callahan 1983) decided to use
them. Reinharz asserts that w^hile feminist group interviews are not
derived from the methodology of marketing research, there is only one
difference hetween the two; feminist researchers are more likely to grant
the interviewees the status of "expert" on the topic of discussion, in
keeping with the feminist principle that women are experts on their own
experience.
Suprisingly, there has heen little development of group interviews as a
feminist method, even though such interviews have much in common
with the consciousness-raising (CR) groups that were a staple of the
women's liberation movement of the late 1960s and 1970s. Marjorie
DeVault asserts that consciousness raising is a method that is at the heart
of the women's movement, the source of feminist methodology (DeVault
1996, 30). Perhaps the prominence of focus groups as a tool of marketing
research has alienated feminist researchers and prevented them from
seeing the connections hetween this method and the concerns of feminist
methodology. In any case, in order to conduct a feminist focus group
interview, researchers will need more information than is currently avail-
able on such topics as recruiting subjects, constructing questions, arrang-
ing and running groups, and dealing with the data in ways that fit with
feminist research principles.
Focus Groups and Feminist Research Principles
Feminist methodology is afieldof inquiry united by membership in the
overlapping research communities of feminism and social science, and
feminist social scientists use the tools of both in a critique aimed at
improving the ways we know society (DeVault 1996, 30). Feminist re-
searchers use a variety of methods, and it is debatable whether there is
such a thing as a "feminist method." Nevertheless, there is consensus
that feminist research is characterized by researchers' striving to adhere
to and achieve certain principles in their research. Many feminist re-
searchers have found that some methods are more conducive to these
principles than others. Cook and Fonow (1986) identify five basic episte-
mological principles that concern feminist researchers: 1) attention to the
significance of gender,- 2) the need to challenge the norm of objectivity and
the rigid separation between the researcher and the researched; 3) the
centrality of consciousness-raising as a methodological tool and "way of
seeing"; 4) an emphasis on the transformation of patriarchal institutions
Focus GROUP INTERVIEWS 47
and the empowerment of women; and 5) concern for the ethical implica-
tions of the research. In the following sections I describe how group
interviews conform to and even advance these principles of feminist
research.^
Studying Sex and Gender
Attitudes and beliefs about gender and sexuality can be very difficult to
study because they are commonly assumed to be simply "natural" at-
tributes of an individual that are unconscious or taken for granted. How-
ever, a constructivist understanding recognizes that gender is not simply
a natural attrihute of individuals nor a social role that we play in certain
circumstances. Rather, gender is primarily a social category that orga-
nizes our perceptions of the world. As Fenstermaker, West, and Zimmer-
man (1991) point out, gender is produced interactionally; it is an ongoing
accomplishment available for production in every interaction (Dull and
West 1991; Fenstermaker, West, and Zimmerman 1991; West and Zim-
merman 1987). It is important to study the taken-for-granted attitudes
and beliefs about gender not as secondary phenomena, but as integral to
the production of the sex/gender system itself.
Because discussion and representation of sex surrounds us, studying
ideas about sexuality might seem easy. However, few people discuss sex
often or intimately. My project on the infiuence of popular culture on the
construction and maintenance of "normative sexuality" required that
research subjects discuss sex in an unusually explicit way. I did not
necessarily want my subjects to describe sexual acts explicitly (although
some did), but I wanted them to state ideas explicitly that usually "go
without saying," to articulate the beliefs and categories that underlie
their conscious attitudes. It is very difficult for people to talk about these
kinds of attitudes and assumptions in an individual interview. In a group
interview, however, the ways participants respond to and interact with
each other can provide richer and more complex data that can reveal
taken-for-granted categories and beliefs.
Focus group interviews can be useful in providing qualitative informa-
tion because they combine many of the advantages of participant observa-
tion with those of in-depth individual interviewing, while avoiding some
of the drawbacks of these other methods (Morgan and Spanish 1984;
Morgan 1988, 15-23). Participant observation allows the researcher to
examine ordinary behavior and group interaction. However it can be a
very inefficient method because of the difficulty of finding and observing
informal interactions around specific topics as they naturally occur. For
example, in my study it would have been difficult to simply observe
"naturally occurring" conversation about sex and about how the media
infiuence ideas about sexuality. Focus group interviews provide the op-
48 FRANCES MONTELL
portunity to observe people's interactions, and they also let the researcher
direct the interactions to the attitudes and experiences of interest (Mor-
gan and Spanish 1984, 259). Although focus groups do not provide access
to interactions as they would ordinarily occur in daily life (Agar and
MacDonald 1995, 81), they do provide a valuable kind of data.
Individual interviews provide researchers with the ability to direct the
interactions; however, this method had other drawbacks for my project. In
order to construct a structured interview protocol on women's ideas about
sex and the influence of the media, I would have needed to conduct many
preliminary interviews to ascertain the categories and issues that are
important to various groups of women. Especially in early interviews, my
questions would necessarily impose my own categories on people's ac-
counts; general and open-ended questions on this topic would be unlikely
to produce useful responses. Such questions require people to consciously
articulate what they usually assume that "everybody knows." A group
interview can also be a better forum for getting at these issues because in
individual interviews each question requires "an answer," while in group
interviews the goal is instead to initiate a conversation. Vague or difficult
questions elicit greater response because a group conversation allows
people to feel more comfortable bringing up different ideas without the
pressure to provide a definitive answer to each question. In a group
interview, participants can build on the responses of others so that a short
or obvious comment does not have to be a dead end as it might be in an
individual interview, but can serve as a spark for another participant's
contribution.
In each group interview that I conducted, for example, I asked such
open-ended questions as "What is sexuality?" and "What is good sex?"
and participants consistently expressed initial confusion over what I
meant. In individual interviews I would have had difficulty getting in-
depth and useful responses to these questions without imposing my own
definitions or subtly directing the answers based on my own expectations.
In each of the groups, however, one or more participants asked questions
and offered tentative answers, and other participants were able to build on
these responses to come to their own understandings of and answers to
these open-ended questions. A group interview also allows certain topics
to be explored in more depth than they might be in individual interviews,
as contradictory ideas prompt women to elaborate on their original state-
ments.
For example, in one group when I asked, "What is good sex?" Janef*
answered that sex can be good because "you're really present with the
person and you're enjoying being with that person, or (else) you're like in
fantasy land, way in fantasy land." I tried to prompt Janet to describe the
differences between "being present" and "being in fantasy land," but she
did not articulate her ideas in a way that clarified her meaning. Had this
Focus GROUP INTERVIEWS 49
been an individual interview I might have simply accepted Janet's catego-
ries and concluded that the primary issue for her was emotional connec-
tion that she seemed to prefer to the disconnection of "fantasy land."
However, in the group interview, another woman, Rebecca, responded to
Janet's statement:
Rebecca: What makes sex good? Well I guess you'd have to think ahout what
makes sex not good, and I don't think I've ever had not good sex.
Eve: Go girl! (all laugh) She's my kind of girl!
Janet: You've never had had sex? My god. Okay . . .
Rehecca: But I guess more narrowly defining sex and sexuality, I think physical
pleasure is prohahly the major defining factor when I think ahout
what makes good sex. Whether it's alone or with somehody else.
Being present with the other person would not he numher one on my
list; physical pleasure would he the numher one thing. (Interview #7)
This led the group to clarify the role of physical pleasure in defining
"good sex," a brief exchange that culminated with Janet saying, "I was
kind of assuming physical pleasure when I was talking about it, but yeah."
Thus the group interaction allowed me to clarify different women's ideas
and assumptions. In an individual interview, when the interviewee says
something that fits with the interviewer's expectations, for example,
about what is and is not important to women regarding sexual experi-
ences, the interviewer does not always probe for other ideas or issues, and
the interviewee is not always able to provide additional information and
clarification. In a group interview, however, there is a greater likelihood
that contradictory ideas will be expressed, and these can serve as prompts
for discussions that address the issues in greater depth.
In a group, if even one person expresses an idea it can prompt a response
from the others, and the information that is produced is more likely to be
framed by the categories and understandings of the interviewees rather
than those of the interviewer. Participants can help each other figure out
what the questions mean to them, and the researcher can examine how
different participants hear possibly vague or ambiguous questions. This is
important in studying sex and gender because these issues are "natural-
ized" to such an extent that it is very difficult to recognize one's own
preconceived notions, much less challenge others' taken-for-granted as-
sumptions. The expansion of the roles available to women in a group
interview, beyond the strict separation between "interviewer" and "inter-
viewee," allows for interactions that are more likely to reveal and even
challenge these taken-for-granted assumptions.
The Researcher/Subject Relationship
Feminists have long advocated the use of in-depth interviews because
they promote a more egalitarian and less objectifying relationship be-
50 FRANCES MONTELL
tween the researcher and subject than quantitative methods. Qualitative
methods potentially allow women to be the "experts" about their own
experiences. However, some feminists question exaggerating the value of
women's experience as the standard of knowledge (Cancian 1992, 632;
Millen 1997). One problem with treating each woman as an "expert" in
this way is that this assumes that each woman is conscious of the forces
that have acted upon her and can articulate her reaction to those forces.
In traditional interviews the interviewer and interviewee bring differ-
ent kinds of authority to the interaction and they may also have very
different goals for the interview and understandings of the topic. The
interviewee may have expertise on the subject at hand but the interviewer
has authority over the research process itself. Ironically, group interviews
may allow the research subjects to be experts to a greater extent than one-
on-one interviews. In a typical one-on-one interview, the interviewer is
actually the expert, the one who frames the issue of study and composes
the questions, while the interviewee is simply an "informant" who pro-
vides information about her particular experience. This different and
unequal status influences the kind of information that is produced. This
does not imply that all interviews are hierarchical and exploitative. In-
deed feminists have often noted that qualitative interviews are far more
egalitarian and collaborative than traditional quantitative research. But
regardless of how open-ended the questions are, an interview is an inter-
action between two people who have different and unequal roles in the
exchange.
"When women interview women, both researcher and subject act on
the basis of understandings about interviewing, and both follow the rules
(or negotiate a shared version of the rules) associated with their respective
roles" (DeVault 1990, 101). One way to change these roles is to incorpo-
rate rather than deny the researcher's personal involvements. Although
other feminists have noted the value of the researcher's personal involve-
ment in interviewing, Marjorie DeVault argues that we need to "move
toward a more disciplined use of the personal" (1990, 104). She calls for
more visibility of the researcher as a resource rather than contaminant in
the product of the research, although precisely how to do this most
effectively remains unresolved (1996, 42). It is very difficult to negotiate
the most productive balance between the researcher's personal experi-
ences and professional role in individual interviews; group interviews are
far more conducive to such a balance.^
A shift in focus from individual knowers to the perspectives of groups
or communities is a theme of feminist epistemology that begins to move
us out of the impasse around personal experience (DeVault 1996,42). This
idea is really not new because "the 'experience' so valued in early feminist
consciousness raising was in fact a collective construction" (DeVault
1996, 42). Because knowledge and meaning are collective rather than
Focus GROUP INTERVIEWS 51
individual productions, focus groups can be an effective method for get-
ting at this socially produced knowledge. "Focus groups give participants
an opportunity to narrate their personal experiences and to test their
interpretations of events and processes with others, and whether con-
firmed or disputed, the result is a polyvocal production, a multiplicity of
voices speaking from a variety of subject positions" (Goss and Leinbach
1996, 116).
Group interviews disrupt the rigid dichotomy between interviewer and
subject, providing the possibility of an equal exchange. Although group
interviews still contain a relationship of "researcher" and "researched,"
and the researcher still asks the questions and frames the issues of the
study, this dynamic is mitigated by the fact that in a successful group
interview the main interaction will be among the participants rather than
between each interviewee and the researcher. Each woman can not only
tell her own story, but she can also question and challenge the other
participants in an effort to gain understanding.
Group interaction can also lead to richer and more complex informa-
tion on certain topics than is possible in individual interviews. Because of
the researcher's authority, and because she requires accurate and "un-
tainted" information, an interviewer is unlikely to directly challenge or
contradict the interviewee. She may point out apparent discrepancies and
ask the interviewee for further elaboration, but she is unlikely to volun-
teer her own opinions or beliefs because such opinions may bias the
interview. In a group interview, however, people can and do challenge and
contradict each other. They ask each other questions, provide examples
from their personal experience, and collectively produce accounts that
would be difficult if not impossible to elicit in individual interviews.
For example, in the second group interview I conducted, Connie stated
that while she had no problem with the idea of masturbation, she had
no personal need for it. Audrey responded skeptically to this, prodding
Connie to elaborate. Connie then described an experience while talking
with an ex-boyfriend on the phone, after which Audrey laughingly ex-
claimed "Oh, honey, that's masturbation!" Connie's further description
of her experiences provided a more nuanced explanation of her ideas about
masturbation. As an interviewer, I would have been unable to challenge
Connie as skeptically as did Audrey, and I would probably not have
elicited as much information. I also would be unlikely to state my own
interpretation of a respondent's experience as definitively as did Audrey.
As Jon Goss and Thomas Leinbach (1996, 122) caution, however, "focus
groups do not eliminate the unequal power relationship between re-
searcher and research subjects, because it is still the researcher who
initiates the exercise and who determines the selection of participants,
the identity of the moderator and the agenda for discussion." Focus groups
"may give greater control over the research process to its subjects, but the
52 FRANCES MONTELL
fact that the social dramas of focus groups are collaborative productions
should not deny the responsibility of the researcher for the analysis and
interpretation of data produced..." (Goss and Leinbach 1996, 122; Millen
1997).
Consciousness-Raising
In describing the difference betw^een traditional and feminist one-on-
one interviews, Bristow and Esper (1984) argue that while a traditional
social science interview is a one-sided "interrogation," a feminist inter-
view is more of a "true dialogue" that "better serve[s] the interest of both
researchers and participants by maximizing the exchange of information
and subsequently, the construction of knowledge" in a way that makes a
feminist interview a consciousness-raising experience (Bristow and Esper
1984, 490). Yet the possibility for the give and take of self-revelation and
questioning among several equal participants makes feminist group inter-
views potentially more consciousness-raising than individual interviews.
Other researchers have noted the consciousness-raising potential of
group interviews. Mies (1983) suggested that feminists shift from indi-
vidual interviews to group discussions not only in order to obtain more
diverse data, but also to help women overcome their structural isolation
and understand the social causes and shared nature of their individual
sufferings. Malhotra (1984) came to a similar conclusion when she in-
volved participants in small group discussions that allowed participants
to influence the methods of the study. These groups also functioned as
support groups for the participants. Callahan (1983) used group interviews
because she believed that women's interactions with each other not only
enhance the flow of ideas and information, but that these interactions are
consciousness-raising in that they encourage women to recognize the
patterns in their shared experience.
The discussions that participants have with each other in a group
interview can be far more consciousness-raising than anything the inter-
viewer could say. For example, the first interview I conducted was with
members of an eating disorder group in which one of the participants was
a young woman, Emily, whose primary experience of sex was that of being
molested. (In a questionnaire I distributed, under "How would you de-
scribe your sexual orientation?" she answered "Bad.") Throughout the
interview Emily talked about sex as a dangerous realm in which women
are very vulnerable, asserting that the only "safe" and therefore good
sexual relationships are those that are based on love and trust. Other
women in the interview, however, expressed a far more casual approach to
sex, arguing that it can be "just an action" and that "it could be just for fun
or something; it doesn't necessarily have to have an emotional tie" (Keri,
Interview #1). Emily later wrote in her journal about how the group
discussion affected her. She wrote:
Focus GROUP INTERVIEWS 53
It was interesting for me to hear how other people's ideas about sex differed
from mine. It made me question my own thoughts and views. This for me is
good, as far as I am concerned . . . sex is pretty awful. I have only experienced
it in a bad light; thus my stand is bitter and extremely fearful. It was good
for me to hear that others have vastly different ideas from mine. (Emily,
Interview #1)
Emily also wrote that she no longer feels that a deeply loving and
committed relationship is necessary for sex to be good, and that satisfying
the "human need to feel good through physical contact" can also be a good
reason to have sex and provide a good sexual experience. I want to empha-
size here that I did not consciously intend to convince participants of
these ideas, and even if I had wanted to it is doubtful that I would have
been effective. The reason that the interview had such a great effect on
Emily's beliefs is that the challenge to her ideas came from several other
ordinary women who spoke from their personal experience, rather than
from the authority of the researcher.
This example also illustrates another aspect of the consciousness-
raising potential of feminist group interviews, and that is the effect on the
researcher herself. Goss and Leinbach's work suggests that "the main
advantage of focus group discussions is that both the researcher and the
research subjects may simultaneously obtain insights and understanding
of particular social situation[s] during the process of research" (Goss and
Leinbach 1996, 117). Bristow and Esper (1984) describe how they con-
ducted informal group discussions with rape survivors in the process of
constructing their structured interview schedule. Through listening to
women whose experiences were both like and unlike their own, the
researchers not only created a better and more comprehensive interview
schedule, they also expanded their own awareness of rape as a social rather
than individual experience (1984, 490-491). The research was thus con-
sciousness-raising for the researchers, as well as the research subjects.*
Because they are similar in some ways to actual consciousness-raising
(CR) groups, feminist group interviews can have the effect of raising the
consciousness of all the participants, including the researcher. However,
significant differences between actual CR groups and feminist focus
groups in many ways reduce the consciousness-raising potential of the
focus groups. CR groups are leaderless groups that meet over an extended
period of time for the purpose of discussing personal experiences (Reinharz
1992, 220-221). "In these meetings, women attempt to articulate a politi-
cal analysis that will facilitate change [The product of consciousness-
raising is] a new way of thinking, relating, naming, or acting" (Reinharz
1992, 221). Focus groups, in contrast, have a leader, the researcher, and
typically meet only once. Participants may or may not ever see each other
again. While the primary purpose of CR groups is to affect the conscious-
ness of the immediate participants, the purpose of feminist focus groups
54 FRANCES MONTELL
is to produce data for analysis and publication in a research report. Any
conscious-raising is a side effect rather than the primary goal. In both CR
groups and feminist focus groups, however, the effects of the process are
impossible to measure immediately. The question for both kinds of groups
is to what extent they can influence and even transform the conditions in
which they take place. What are the effects of raising women's conscious-
ness or of producing a feminist research project? Both can be empowering
and can participate in transforming patriarchal relations, although the
immediate effect is likely to feel quite small and local.
Transformation and Empowerment
Tania Modleski (1991) argues that politically engaged feminist criti-
cism should be performative. Rather than being content to uncover al-
ready existing meanings produced by already constituted subjectivities,
feminist research should be aimed at "bringing into being new meanings
and new subjectivities" (46). Bringing a group of women together to talk
about issues that are important to them creates, even if only temporarily,
connections and solidarity among women that contribute to feminist
consciousness and social action. Group interviews, then, provide one of
the best methods for a feminist cultural critic because they facilitate
active transformations of consciousness. Alan Johnson makes a similar
point when he argues that researchers can "overturn the marketer-con-
sumer model for focus groups" and locate the research process instead in
the "new politics of knowledge," in democratic social relations associated
with the consciousness raising group, the strike committee, or the town
meeting. Such an approach views the research process "as itself a 'trans-
formational' intervention, at once scientific and political" (Johnson 1996,
525). Citing Raymond Padilla (1993), he argues that:
[GJroup discussions raise consciousness and empower the participants pre-
cisely because they "reveal to the investigator and to the subjects themselves
the overt and hidden aspects of problematic experiences in everyday life."
Group discussions can foster a collective identity among the participants
because they can transcend individualism and connect up individual narra-
tives, first to each other, and then to wider social, economic, cultural and
political influences. (1996, 534)
In defining "empowerment," Bhavnani (1989), following Flacks (1983),
defines "power" as "the capacity to influence the condition and terms of
the everyday life of a community or society;" and "empowerment" as
"the realization of this capacity... to create history" (Bhavnani 1989,145,
149). Drawing on Grossberg (1987), I regard a practice as "empowering" to
the extent that it enhances the conditions that enable people to engage in
progressive practices and to live their lives in different and better ways.
Like CR groups, research projects can be empowering to the extent that
Focus GROUP INTERVIEWS 55
they give women access to new information and new ways of thinking
that enable them to question oppressive practices and try out alternatives.
For example, Jon Goss and Thomas Leinbach (1996) note:
For many participants in our research the focus group was the first occasion on
which they had spoken in public or had their opinion about an issue solicited.
This was initially unnerving for some, but also empowering since their exper-
tise and right to speak could be publicly established, and they began to develop
new communication skills. (121)
In addition, research can potentially be empowering when research
subjects have the opportunity to direct the research project itself. Rein-
harz (1992, 181-6) and Cancian (1992, 628) both describe participatory
research projects in which the research subjects were involved in various
stages of the study, including formulating the research questions and
hypotheses, selecting participants, and analyzing the data. Although par-
ticipants can become very involved in influencing the research process, I
agree with those who question the idea that feminist research should be
a truly collaborative process. "Fundamentally, the idea that the research
relationship should be or ever can be equal in any sense is an illusion"
(Millen 1997, 3.3).
It is we who have the time, resources and skills to conduct methodical work,
to make sense of experience and locate individuals in historic and social
contexts.... It is an illusion to think that, in anything short of a participatory
research project, participants can have anything approaching "equal" knowl-
edge to the researcher. (Kelly et al. 1994, 37).
Although involving the research subjects in all phases of the project is
a laudable goal, it can slow down the process considerably. Additionally,
few research participants can afford the time necessary, or even have the
inclination, to participate in an academic research project to such an
extent. Marjorie DeVault (1996, 38-9) questions feminist overemphasis
on problems of power that place "excessive demands" on feminist re-
searchers. But, as Cancian (1992) argues, we will not be able to conduct
truly feminist research and feminist methods will not become widely
accepted until we change the hierarchical, elitist structure of academia.
Although I do not want to overemphasize the idea of "empowerment"
as the reason for involving subjects in research, it is true that by partici-
pating in research people can contribute significantly to the description
and analysis of a social issue that is of great importance to them (Opie
1990), and this can be empowering. For example, in response to my
statement that the goal of my project is to explore the ways that women
define sex, one participant in a group interview agreed that this kind of
research is needed because women have not had much opportunity to
define sex for themselves.
56 FRANCES MONTELL
Ingrid: Yes, because men are the ones that create those . . . out there in the
advertising world, they create those images that are supposed to mean
sex to us. I don't think women have really been able to put tbeir input
out there. We're not supposed to have an opinion on sex. We're sup-
posed to just go along with it.
Angela: We're supposed to be ready.
Ingrid: But be ready, at any moment! (laughter)... No really, no wonder it's so
hard for us to even say what it is to us, because we're not supposed to
have an opinion. (Interview #6)
Along with validating the importance of participation in this research,
Ingrid attrihutes the source of the interview group's inarticulateness
about sex to some of the cultural forces that have silenced women's own
perspectives.
In addition, research can empower those who have been seeking help
for their own personal problems by enabling them to help others, such as
the researcher, other group members, and people who may read the re-
search report. The desire to help others in a similar situation is frequently
the main reason people give for volunteering to participate in research
(Bristow and Esper 1984, 494; Opie 1990, 64). If interviewers are respon-
sive to individual concerns, then interviews can be therapeutic in a way
that is empowering. This effect is heightened in group interviews hecause
of the support such groups can provide. For some participants, reflecting
on and re-evaluating their experience as part of the interview process can
have important personal and political consequences (Opie 1990, 64).
Central to many discussions of feminist methodology is tbe notion that the
purpose of knowledge is to change or transform patriarchy. . . . [K]nowledge
must be elicited and analyzed in a way that can be used by women to alter
oppressive and exploitative conditions in tbeir society. This means that re-
search must be designed to provide a vision of tbe future as well as a structural
picture of tbe present. (Cook and Fonow 1986, 12-13)
While it is difflcult to assess, a project will be empowering to the extent
that a researcher is attentive to the ways her project can provide not only
a critique of conditions as they exist, but also a vision of alternatives for
the future. In my project I not only asked the women in my groups to
critique the images and messages that surround them, I also asked them
to describe alternative images and messages they would like to see and
promote. Also, the women not only provided support for each other in
dealing with particular personal issues, but the ways women described
very different experiences and attitudes sparked the imagination about
possible alternatives.
Focus GROUP INTERVIEWS 57
Ethical Considerations
Discussions of the ethics of a research project frequently focus on the
project's results and the uses to which they are put. But feminist method-
ologists are also concerned ahout the ethical issues involved in the re-
search process itself, particularly in the relationship hetween the re-
searcher and the research suhjects.
It is important that research about oppressed groups attempts to alleviate
repressive and exploitative aspects of the research process itself. Oppressed
and vulnerable persons . , . are frequently tbe objects of social researcb.
Typically, such research is structured so as to reinforce an alienated relation-
ship between "researcher" and "subject." That is, oppressed groups are in-
volved in relationships with others, which benefit those in power to a greater
extent than the oppressed group benefits. (Malhotra 1984, 469)
Feminist researchers rigorously try to avoid perpetuating the exploita-
tion of women through their research, and ethical issues are heightened in
feminist methodology (Reinharz 1992,27). Anticipating the various needs
of participants and reflecting on how people might heneflt from partici-
pating in the research is one way to lessen exploitation. In my own project
I was unahle to offer participants financial remuneration, so I was very
interested in the reasons women participated and the rewards they said
they received. Their reasons included the expectation that the interviews
would he fun or interesting, the desire to have a safe forum to talk ahout
sex, interest in hearing what other women had to say, particular ideas that
they wanted the research to include, and even their wish to help me out.
It was important to consider the women's needs in order to effectively
recruit research suhjects, and to justify the commitment of time and self-
revelation that I was asking of participants. Such considerations are espe-
cially important when conducting interviews with existing groups, where
the researcher is usurping time people have set aside for their own needs.
In her one-on-one interviews with women with hreast cancer. Rasper
(1994) attempted to meet the needs of the interviewees, as well as her own
needs as a researcher hy making the interview less hierarchical. She did
not ask direct questions ahout "sensitive and difficult topics" hut only
pursued such topics as "fears, hody image, sexuality, and the like" if the
interviewee herself raised the issue (Rasper 1994, 270). Rasper's sensitive
approach assumes that discussing difficult topics is exploitative, henefit-
ing the researcher at the expense of the research suhject. However, many
women do not have a safe environment for discussing certain personal
concerns and might he relieved at the opportunity that the researcher
provides. Open discussions of feelings might serve to demystify such
issues as sexuality in relation to cancer and help other women resist the
pressure to conform to standard medical practice, such as hreast recon-
struction surgery. Yet women might also need encouragement to talk
58 FRANCES MONTELL
about these issues. It is difficult to negotiate these issues sensitively in a
one-on-one interview. However, having another woman raise the topic or
hearing another woman's response to questions on this issue could be a
very beneficial experience and encourage an interviewee to share her
experiences more than an interviewer alone could do. I am not implying
that Rasper's research does not provide the opportunity for this kind of
experience, only that a group interview is more likely to promote discus-
sion of sensitive issues and provide more support and validation than a
single interviewer is able to provide.
The potential to give something back to the interview participants
lessens the likelihood that group interviews will be a one-way exchange
and makes the relationship between researcher and researched less ex-
ploitative. However, group interviews may inhibit trust and confidential-
ity, so the researcher must explicitly address these issues and establish
common ground rules.
Conducting Feminist Group Interviews
Recruiting Participants for Group Interviews
Finding and recruiting participants is a vital part of a research project.
Many practical and ethical issues arise before the subject sits down with
the researcher and is ready to provide data. Because sampling issues for
focus groups are similar to those for other kinds of qualitative studies, I
will focus on the purpose of the sample in feminist group interviews. In
quantitative studies, the primary concern in recruiting subjects is obtain-
ing a random sample from which the researcher can generalize to the
population. The primary goal of focus group interviews, however, is to
learn about subjects' experiences and perspectives rather than to test
hypotheses. In selecting participants for focus groups, sample bias rather
than generalizability is of concern (Morgan 1988, 44-5). While a quantita-
tive study requires a random sample to best represent the population, a
focus group study requires participants who can provide the best data.
Dorothy Smith (1987), in proposing a sociology oriented "from the
standpoint of women" and based on "the everyday world as problematic,"
argues that the validity of findings should not refer to how well the
research subjects represent some larger population, but rather to how well
the data describe particular instances of larger social processes. Rather
than a random sample, the participants in focus groups can be thought of
as a collection of individuals whose experiences highlight the social
relations of interest. Recruitment for focus groups does not try to repre-
sent the population in general but instead emphasizes theoretically cho-
sen subgroups who can be expected to provide the best data (Morgan 1988,
44--5). Participants are chosen for their relevant experiences, as well as
Focus GROUP INTERVIEWS 59
their level of self-awareness and comfort in discussing sensitive issues in
front of others. When recruiting subjects for focus group interviews,
concern with sample validity in the quantitative sense might actually
reduce the quality of the data if it produces groups that cannot generate
good discussions. The researcher will also attempt to bring together
participants who are as diverse as possible in terms of categories relevant
to the topic of the study.
In my own study, for example, I wanted participants who represented a
wide age range and varied sexual orientations,' who had given some
thought to sexual issues and would feel comfortable discussing these with
each other. For this reason I began recruiting participants through thera-
pists and facilitators of a wide variety of support groups to whom I gave
information to pass on to interested members of their groups. I assumed
that through such groups I would have a greater chance offindingwomen
who had given some thought to sexual issues and who would feel comfort-
able talking about them in a group setting.
In general, one has several options in recruiting for focus groups. The
first option, which is not very different from recruiting volunteers for
individual interviews, is to recruit individuals and arrange them into
groups. Alternatively, a researcher can ask a potential participant to
recruit friends who might be interested in participating (similar to "snow-
ball" sampling). Another option is for a researcher to interview members
of already existing groups. A researcher need not necessarily choose one of
these strategies over the others. I used all three of these strategies and each
kind of group has its benefits and drawbacks (which I will discuss below
in relation to group dynamics). I focus here on the third option, recruiting
participants from different kinds of support groups.
To recruit participants from already existing groups, the process of
gaining access will vary depending on whether the group is facilitated or
leaderless. Leaderless groups are usually open and easy to attend, but
gaining access to group members for research purposes is more difficult
than with professionally facilitated groups because no one can authorize
participation in the research for the whole group; approval has to be
negotiated by consensus. The researcher seeking to recruit from existing
groups should openly identify herself and her interests before attending
group meetings to avoid the ethical issue of misrepresentation (Gordon
1987). Even though such honesty may prevent the researcher from attend-
ing a group, it is still possible to get a group member or facilitator to assist
in recruiting people in such groups.
In groups that have a leader or facilitator, that person can act as a
gatekeeper and unilaterally allow the research, or can be the researcher's
advocate to group members. Group facilitators are very protective of
group members, but also can be supportive of research that they perceive
as important, especially studies of issues in which they have a personal
60 FRANCES MONTELL
Stake. Again, the most effective approach with group facilitators is for the
researcher to be honest about her identity, the goals of the study, and the
role of the group facilitator in gaining access to group members.
A word of caution is also in order. Group interviews are not a shortcut
to conducting many individual interviews. Unless one already has a group
or groups available for study, recruiting for group interviews can be much
more difficult and time consuming than for individual interviews because
of the difficulty of finding people who will discuss private issues in a more
public setting (rather than just with the interviewer), as well as the need
to coordinate several people's schedules. However, the researcher who
can surmount the difficulties of gaining access to relevant groups, and can
portray the research in ways that will appeal to potential volunteers, will
likely he rewarded with a group of extremely helpful, enthusiastic, self-
aware, and articulate participants in the research process.
Group Interview Dynamics
The dynamics of already existing groups and ad hoc groups can vary
widely, and there are benefits and drawbacks to each. In already existing
groups, participants are familiar and comfortable with each other, and
they can provide the researcher with information about each other's
experiences that each participant may not think to explain. A drawback to
such groups is that participants can he too familiar with each other. A
woman in such a group may not provide much description and explana-
tion of experiences that the others have already heard ahout. In such cases
the researcher needs to pursue information and explanations that may
seem ohvious to the others, and avoid getting swept along into an "Oh yes,
I know what you mean" kind of response. For example, in one interview
I conducted with an alcohol recovery support group, Leonore talked ahout
how traditional messages ahout sex led her to feel so guilty that the only
way she could enjoy sex was under the infiuence of alcohol. During the
ensuing exchange, the participants got so caught up in a particular point
that I was not ahle to pursue explanations of things that were of interest
to me, hut that they all already knew ahout each other. I asked Leonore if
using alcohol to get over the guilt led to prohlems.
Leonore: Oh yeah, definitely. I hadn't had sex without alcohol for ten years.
Beverly: Yeah, that's what led to problems for me was not having sex without
alcohol.
Jill: Me too.
Leonore: And I really think, in order to give myself permission to be--infidel-
ity, was the alcohol. Without the alcohol I wouldn't have been ca-
pable ...
FM: Of being with somebody besides your husband?
Focus GROUP INTERVIEWS 61
Leonore: Besides my husband, right. And I think that's what has happened
more to the generation . . . . I know my daughters all lost their
virginity because they had been drinking.... But I really feel that--
Jill: I definitely agree.
Leonore: It's a theory of my own, but I really feel that--
Jill: No, I agree honey, I agree, totally.
Leonore: That's been a big part. Alcobol and drugs have really changed people's
outlook on sex.
Jill: And as far as being careful, as far as AIDS, everything else. Under the
infiuence it's tbe pleasure of the time. (Interview #3)
The conversation continued on tbe topic of the effect of alcohol on
sexual behavior and I w^as not successful in bringing it back to discussing
the circumstances of Leonore's infidelity (and I forgot to bring the issue up
at a later point in the interview). In this case the participants' enthusiasm
for a particular topic overrode my interest in an event Mrith M^hich they
were probably already familiar. On the other hand, sometimes partici-
pants' familiarity with each other's stories can be an asset to a researcher.
This was the case in another interview with a group of close friends (two
of the participants, Terry and Abby were also lovers). Before each inter-
view I had the participants fill out a questionnaire in which one of the
questions was "What is the most interesting, fun, or outrageous sexual
experience you have heard of?"*During this interview, the topic of this
survey question came up, and Jacquelyn and Abby prompted Terry to tell
certain sexual stories:
Jacquelyn: Oh, Terry, tell your stories! The fun thing--Can I just tell why I like
your stories so much? Because they're all just by yourself. I mean,
all the fun ones, (laughter)
Terry: No no no!
Abby: The Rain Gutter one is not by herselfI
Jacquelyn: Maybe I need to hear The Rain Gutter over again.
Abby: Oh, you haven't heard The Rain Gutter? Oh yeah.
FM: (to Jacquelyn) Well tell the story that you like.
Jacquelyn: Oh, it's just "One Hammock and One Unicycle." But they're not
fair, because they're her stories, so (to Terry) you have to teil them.
I mean you don't have to tell them, obviously, but--
Terry: About sex with myself?!
Abby: Your kneecaps are starting to sweat. Tbis is gonna be a good one. Terry
has this wonderful barometer when she's starting to feel a little uncom-
fortable. She perspires.
Jacquelyn: Anyway, no, I don't think you should have to tell it. I'm sorry.
Terry: No, it's not really--
Abby: Tell the Rain Gutter story, that's a fun one; that's a good one.
Terry: That's actually just the most unusual place. It's obviously not the best
sex I've had.
62 FRANCES MONTELL
Abby: Well she asked about unusual, wild or outrageous. Not the best sex, . , ,
I don't think I want you to tell about the best sex you've ever had, [But]
I would say that that was .,, it was certainly interesting, and hopefully
it was fun, and I would say your mother would probably say it was
outrageous,
Terry: Uh huh. She probably heard it. Sliding down the roof,,,, (Interview #8)
Terry then proceeded to tell hoth stories, and she was prompted and
assisted hy Ahhy and Jacquelyn who encouraged her with their reactions
and provided additional details. This example shows that it is possible to
use participants' familiarity with each other as an asset, hut the inter-
viewer will have to he diligent in following up on topics that the partici-
pants do not discuss fully enough.
Conversely, groups in which participants know each other may feel
uncomfortable revealing certain information, particularly ahout sexual
issues, in front of people they know they will see again unless a high level
of trust has heen estahlished. Thus support groups or friendship groups
where there is already a high level of self-revelation and trust are prefer-
ahle to more casual groups of people who already know each other. In
addition, if a support group's usual facilitator is present, her comments
may carry more weight and greatly influence the discussion. In this
instance, the researcher should speak with the group facilitator ahout
these concerns ahead of time or include her in a different group interview.
The group facilitator may even prefer to participate in a group where she
can he anonymous and can play the same role as other participants. In my
study I interviewed all the members, including the "leaders," of two
groups, hecause these were peer support groups in which the leader was
not seen as having a different status from the other group memhers. But in
the case of another support group in which only the professional facilita-
tor and one group member volunteered to participate in my study, I made
sure to put them in different group interviews.
The dynamics of ad hoc groups composed of individual volunteers are
somewhat different than groups in which the participants already know
each other. In composing such ad hoc groups I have found that the hest
strategy is to maximize the connections among people hy making the
groups as homogeneous as possible in terms of age, ethnicity, sexual
orientation, and relevant experiences. Not only do such groups facilitate
discussion (Morgan 1996, 143), hut homogenous groups also take the
pressure off participants to "represent" their category, if, for example, one
person in the group is the only person more than 40 years old, or the only
lesbian, or the only member of any particular population. Other things
heing equal, anonymous groups that are more homogeneous will he more
likely to foster a feeling of comfort and mutual understanding than anony-
mous groups in which participants are very different from each other.
Participants in ad hoc groups need to introduce themselves more thor-
Focus GROUP INTERVIEWS 63
oughly and explain everytbing more carefully tban tbey would witb
people tbey already know well. Tbis obviously benefits a researcber wbo
will get clearer descriptions and explanations of events and experiences.
Nevertheless, tbe anonymity of sucb groups entails a trade-off: tbe par-
ticipants may feel more comfortable revealing information to people wbo
tbey do not expect to see again, but they will also require more time to
develop trust in the group.
The dynamics of focus group interviews are very different from indi-
vidual interviews. "The hallmark of focus groups is the explicit use of the
group interaction to produce data and insights that would be less acces-
sible without the interaction found in a group" (Morgan 1988, 12, empha-
sis in original). The questions in a group interview should be designed to
promote group discussion. A few open-ended questions (and a short list of
topics that I wanted to make sure the group addressed) worked far better
in fostering lively two-hour discussions than a detailed list of questions
about each individual's experiences and beliefs. I organized tbe interviews
like "brainstorming" sessions, oriented toward getting as many ideas
expressed as possible while also developing some kind of a consensus. For
example, in eliciting information about sexual norms I found it produc-
tive to ask group oriented questions such as "Let's talk about the defini-
tion of 'good sex'. What makes a sexual experience good?" This kind of
question does prompt individuals to talk about their experiences, and it
also facilitates group discussion of the topic.
The researcher may also need to explicitly encourage participants to
respond to and interact with each other. This is especially important if
participants have experience in self-help or support groups, particularly
12-step groups, in which "cross-talk" is not allowed. It can be helpful to
tell the participants that you are interested in how they interact with each
other and in hearing as full a range of opinions and experiences as possible,
that they should feel free to question and respond to, and even to disagree
with, each other. Once participants feel comfortable interacting, how-
ever, they will probably attempt to resolve any points of disagreement on
their own. If one participant says something that contradicts others'
experiences or beliefs, the others are very likely to challenge that state-
ment, or at the very least ask for further explanation until the source of the
disagreement can at least be understood if not eliminated. In feminist
focus groups it is also quite likely that the interviewees have some
interest in the research topic, and more often they have a great interest.
Therefore it can be as important to tbe participants as it is to the re-
searcher that they express themselves openly and challenge statements
with which they disagree. The kind of exchange and negotiation that
occurs when participants disagree or do not understand each other can be
the most informative for the researcher and should be encouraged. The
most I needed to do in this regard was to ask, "Does anyone have a
64 FRANCES MONTELL
different experience or opinion on that issue?" and most of the time I did
not have to prompt them even this much.
Analyzing Focus Group Data Using Discourse Analysis'
Interviews are used differently in discourse analysis than they are in
traditional social science. Traditionally, social scientists use interviews
as a source of information about people's beliefs, attitudes, and behaviors.
What people say in interviews is taken as evidence, with various degrees
of reliability, of a reality outside of the interview itself. Similarly the
researcher who uses focus groups in marketing research does want to
know what each individual really thinks, what products they are likely to
buy, what they really think about a particular advertising campaign. If the
group dynamics prevent an individual from saying what she really thinks,
then the researcher has failed to get accurate and useful data from that
person.
These very issues that make the group interview difficult for getting
consistent individual data make it ideal for exploring public discourse and
taken-for-granted cultural assumptions. The research questions of dis-
course analysis are not about how accurately descriptions fit reality, but
rather about discourse itself; not the individual beliefs and experiences,
but the talk itself is the subject of interest (Potter and Wetherell 1987,
160). Discourse analysis assumes that talk is not neutral but is both
evaluative and performative. Talk performs actions and has consequences.
Discourse analysts examine "how description and explanation are meshed
together and how different kinds of explanations assume different kinds
of objects or supply the social world with varying objects" (Potter and
Wetherell 1987, 52). The interview data are not examined solely for what
they reveal about the speakers' actual beliefs and activities. Instead the
analysis concentrates on what people's talk reveals about the larger cul-
tural discourse, the ways it is possible to talk in that culture, and the ways
particular assumptions and ideas are connected. For example, Robyn
Longhurst notes about her use of focus groups to study pregnant women's
experiences, "When analysing the transcripts of the conversations I fo-
cused primarily on discourse--on meanings, narratives, explanations,
accounts and anecdotes. That does not mean that I think that the material
corporeal state of pregnancy is a simple matter of linguistic practice or of
representation . . . but ideologies and social structures are intertwined
with discourse" (1996, 145).
Group interviews create opportunities for a very productive level of
analysis not available with individual interviews. The researcher can
analyze not only what each participant reveals about herself, she can
observe how people negotiate issues with each other, noting which ideas
Focus GROUP INTERVIEWS 65
the group accepts and which statements spark disagreement. Instead of
simply aggregating individual data, the researcher can directly observe the
extent and nature of agreement and disagreement among participants
(Morgan 1996, 139). Goss and Leinbach (1996, 115) argue that the group
discussion that occurs in a focus group interview "provides valuable
insight into social relations and that the 'stories' produced in the collabo-
rative performance of a focus group better refiect the social nature of
knowledge than a summation of individual narratives extracted in inter-
views" (see also Agar and MacDonald 1995, 80, 82, 84). Further, group
interviews allow the participants themselves to observe and comment on
their differences (Morgan 1996, 139). These points of agreement and
dispute illuminate participants' underlying assumptions and the extent
to which they share a culture of common sense understandings. How did
different people interpret different questions? What staternents did they
all agree with? What ideas were difficult to express or seemed to need
extensive justification? How did the women respond when they were
challenged? The answers to these questions can tell us which ideas have
legitimacy, in that group at least.
For example, in one interview a discussion of gender equity led to a
discussion of gender differences in communication and biological differ-
ences in brain structure. On each point, all the women expressed general
agreement with each other, building on each other's comments rather
than contradicting them. But when Connie asserted, "Women are the
communicators," Beatriz immediately disagreed.
Beatriz: Well, I have a different hit on that right now that came out of doing this
self-defense class because I got really angry at the woman who did the
class, because she did what I call "male-bashing" and part of it had to
do a little bit with wbat you're saying about men not communicating.
[But] tbey just communicate non-verbally, and we define verbal as
communication, and non-verbal is not. . . . And wbat I'm getting to
about tbis self-defense class talking about, "Well women can wear
wbat tbey want, tbey can go wbere tbey want, tbey can do wbat tbey
want, they can wear bigb heels, dress up, and give all tbe non-verbal
signals of 'Hi, I'm ready' and tben say no." And I think tbat's true, but
wby are you doing tbat, and then mixing drugs and alcobol witb tbis
big mix of all tbose non-verbal signals to a non-verbal operating per-
son?
Audrey: But you know, I'll tell you wbat. I was just mugged in August after
years of doing street work all over tbis country jlaugbs) and I was witb
a guy, okay? I was witb a guy and I got jumped by tbree teen-agers and
beaten. We managed to get away and tbey caugbt the kids and tbey're
in jail, but in terms of wbat I'm bearing you saying is like, I wasn't
giving any message about anytbing--
Beatriz: No no no I wasn't talking about tbat.
Audrey: Well I just tbink you have to watch tbat wbole tbing about rape and
66 FRANCES MONTELL
attack, it's a crime of violence and anger, it's not a sexual crime, from
my point of view. I don't think you give a come-hither--
Beatriz: No, I'm talking about frustration and, well, I'm working on this. I feel
that there is some assaultiveness with non-verbal cues that girls
should be told that it just is inappropriate. I agree with "violence is not
sexual" but there's other things going on with--
Audrey:Well you know, one of the things that was a hig learning experience for
me in that whole deal was that I had heen acculturated to feel more
safe when I was with a man--
Beatriz: Ooooh.
Audrey: That you're protected when you're with a man. First of all you're more
protected than when you're alone, hut certainly somehow a guy will
keep you safe. And that was not the case, at all. (Interview #2)
As Audrey goes on to describe her experience and make her argument,
the topic moves to the issue of "safety"; all the women agree that they
need to be aware of these issues but not let them determine how they live
their lives. This movement in conversation topics, from gender differ-
ences in communication, to female responsibility for sexual assault, to an
experience of a non-sexual attack that is taken as evidence that women do
not in any way invite sexual assault, shows the ways that the feminist
idea that "rape is violence, not sex" has become widely accepted (al-
though the precise connection between sex and violence is muddled).
Interestingly, Beatriz was speaking specifically about rape, and maybe
more generally about sexual misunderstandings, but Audrey associated
Beatriz's remarks with a specific instance of non-sexual violence. When
Audrey talks about it as a "learning experience," Beatriz's response indi-
cates that she too assumed that women are safer when they are with men
and that Audrey's experience has challenged this belief.
In tbis interview, the group accepted some statements about men and
about sex differences, but challenged others. In almost all cases, however,
the group eventually came to a consensus after a disagreement, indicating
that they believed that any apparent contradiction hetween each other's
ideas could be resolved. As this example illustrates, this does not mean
that they resolved all issues of disagreement, only that in all cases they
ended discussion of a topic on a note of consensus with the feeling that
they had all come to an agreement.
I am not arguing that researchers who use feminist focus groups are
unconcerned with whether women express themselves honestly or de-
scribe their experiences accurately, only that an accurate refiection of
individual experience is not the focus of a discourse analysis. It is not
individual perspectives but "the group point of view that is the goal of
ethnographic research" (Agar and MacDonald 1995, 81). The point is not
whether the data are likely to be more objective and accurate in either
marketing or ethnographic group interviews, but rather that the goals and
kinds of data obtained are very different for each.
Focus GROUP INTERVIEWS 67
Conclusion:
Group Interviews and Feminist Research Principles
My goal has been to further the development of focus group interviews
as a valuable method for feminist social scientists. This method not only
conforms to feminist principles but offers the possibility of expanding and
utilizing them in new ways. The distinctiveness of feminist methodology
is located in a shared commitment to three goals: 1) to "bring women in,"
that is to find what has been ignored, censored, and suppressed in the
standard focus on men's concerns; 2) to minimize harm, control, and
exploitation in the research process, using research strategies that are
more inclusive and less hierarchical than the standard practice of social
research; and 3) to conduct research that will be of value to women and
will lead to social change or action beneficial to women (DeVault 1996,
32-3). Group interviews provide a new way for feminist researchers to
meet these goals. Focus groups provide the opportunity for studying
issues of gender and sexuality with a more egalitarian relationship be-
tween the researcher and the research subjects, and consciousness-raising
and empowering interaction among participants. Rather than exploiting
participants who may get very little in return, group interviews can
provide support for participants and meet some of their needs.
Focus group interviews provide a way to study attitudes and beliefs,
and more importantly can facilitate the kind of thinking that can lead
people to question their previous assumptions. As Cook and Fonow (1986,
13) point out, "feminist research is . . . not research about women but
research for women to be used in transforming their sexist society."
Group discussions can identify "local theories and popular knowledge"
while group members may generate new knowledge as they attempt to
understand their situation (Gancian 1992, 633). Group interviews can
advance the transformation of patriarchal social institutions through
research (Gook and Fonow 1986, 5) and through their potential for con-
sciousness-raising and empowerment.
An earlier version of this article was presented at the American Sociologi-
cal Association meetings in Washington D.C., August 1995.
Frances Montell recently received her Ph.D. in Sociology from the Uni-
versity of California at Santa Barbara. She is currently conducting evalu-
ations of education programs at Stanford. Her research interests include
gender, sexuality, education, and women's participation in sports. Corre-
spondence should be sent to Montell at The Stanford University Learning
Lab, Bldg. 10, Stanford University, Stanford, CA 94305-2060; montell@
leland.stanford.edu
68 FRANCES MONTELL
Notes
1. Two recent volumes are Morgan 1993, and Vaughn et al. 1996.
2. This paper is concerned specifically with the application of feminist method-
ology to focus groups. For more general information on how to conduct focus
groups, I recommend Morgan and Spanish 1984, Greenbaum 1988, Krueger
1988, and the articles in Morgan 1993, particularly those hy Alhrecht et al.,
Knodel, Krueger, and Zeller.
3. My comparison of focus groups to other methods is specifically in relation to
the principles of feminist methodology. For a more general comparison of
focus groups to other methods see Morgan 1996.
4. All of the names of participants that I use here are pseudonyms.
5. For example, most interviewers try to keep comments ahout their own expe-
riences and ideas to a minimum, using them only to prompt the interviewee.
In analyzing the interview transcripts, researchers usually ignore the parts of
the interview in which they themselves speak hecause the interviewer is
technically not a subject of the study and thus it is difficult to analyze this
information in a systematic way. I do not have the answer to this problem
here, I only point out that there is less need for such interviewer comments in
a group interview because each participant can have this same prompting
effect on the others.
6. When I began my research, like many feminists, I took it for granted that sex
is a vitally important part of the human experience (Montell 1997). I had no
particular critique ahout the importance we place on sex in our society,
particularly for women, and how deeply conservative the connection between
"love" and sex can he. Through listening to the women in my study discuss
their ideas with each other, my own consciousness was raised ahout these
issues. I also had my consciousness raised ahout my unthinking acceptance of
ageist ideas ahout older women's sexuality and the idea that women who grew
up before the sexual revolution are conservative and "repressed." Individual
interviews would probably not bave had as great an effect on my own con-
sciousness hecause it would have heen easier for me to isolate any surprising
or challenging statements as coming from a few unusual individuals. When
such ideas emerged as the product of consensus among a group of women, it
was impossible for me to marginalize the ideas.
7. I also would have liked greater diversity in race and ethnicity, but I was unable
to accomplish this.
8. I used the questionnaire as hoth a recruitment tool and as a way of getting
certain consistent information ahout the participants. I included some ques-
tions that I thought would he "fun" to answer, as well as questions that
provided standard demographic data such as age, occupation, sexual orienta-
tion, etc.
Focus GROUP INTERVIEWS 69
9. Discourse analysis is not the only, nor even the predominant, method of
analysis for focus groups. An important factor in considering this kind of
analysis is that it requires detailed transcripts from tape recordings of the
interviews. I also found it impossihle to distinguish speakers from each other
without videotaping the interviews as well. If a videotape is used, it is impera-
tive to inform participants beforeband. Altbough this may intimidate some
potential participants, it is possible to reassure them that it is necessary to the
project, that it will only be used for analyzing the data, and that their confiden-
tiality will be absolutely respected. Some participants may still prefer to sit
out of tbe frame or witb tbeir back to tbe camera, but I did not bave any
participants back out of tbe project because of the videotaping. See Bertrand et
al. (1992) for a detailed description of alternative methods of recording the
group interviews and analyzing the data. The methods they describe are based
on, and are particularly useful for, international family planning and bealtb
researcb projects.
References
Agar, Michael, and James MacDonald. 1995. "Focus Groups and Ethnography."
Human Organization 54(l):78-86.
Alhrecht, Terrance L., Gerianne M. Johnson, and Joseph B. Walther. 1993. "Un-
derstanding Communication Processes in Focus Groups." In Successful Focus
Groups: Advancing the State of the Art., ed. David L. Morgan, 51-64. Newbury
Park, CA: Sage.
Bertrand, Jane T., Judith E. Brown, and Victoria M. Ward. 1992. "Techniques for
Analyzing Focus Group Data." Evaluation Review 16(2)(April): 198-209.
Bhavnani, Kum-Kum. 1989. "What's Power Got to Do With It? Empowerment
and Social Research." In Deconstructing Social Psychology, eds. Ian Parker
and John Shotter, 141-152. London: Routledge.
Bristow, Ann R., and Jody A. Esper. 1984. "A Feminist Research Ethos." Human-
ity and Society 8(4):489-496.
Callahan, Joan Thompson. 1983. "Upward Mohility from the Inside: A Phenom-
enological Study of Female Psychologists from Working Class Backgrounds."
Ph.D. dissertation, Massachusetts School of Professional Psychology.
Cancian, Francesca. 1992. "Feminist Science: Methodologies that Challenge In-
equality." Gender and Society 6(4):623-642.
. 1989. "Participatory Research and Working Women: Democratizing the
Production of Knowledge." Paper presented at the annual meeting of the
American Sociological Association.
Cook, Judith A., and Mary Margaret Fonow. 1986. "Knowledge and Women's
Interests: Issues of Epistemology and Methodology in Feminist Sociological
Research." Sociological Inquiry 56:2-29.
Devault, Marjorie L. 1996. "Talking Back to Sociology: Distinctive Contributions
of Feminist Methodology." Annual Review of Sociology 22:29-50.
. 1990. "Talking and Listening from Women's Standpoint: Feminist Strat-
egies for Interviewing and Analysis" Social Problems 37(1):96-116.
Dull, Diane, and Candace West. 1991. "Accounting for Cosmetic Surgery: The
Accomplishment of Gender." Social Problems 38(l):54-69.
70 FRANCES MONTELL
Fenstermaker, Sarah, Candace West, and Don H. Zimmerman. 1991. "Gender
Inequality: New Conceptual Terrain." In Gender, Family, and Economy: The
Triple Overlap, ed. Rae L. Blumberg, 289-307. Newbury Park, CA: Sage.
Gordon, David F. 1987. "Getting Close by Staying Distant: Fieldwork with
Proselytizing Groups." Qualitative Sociology 10(3):267-287.
Goss, Jon D., and Thomas R. Leinbach. 1996. "Focus Groups as Alternative
Research Practice: Experience with Transmigrants in Indonesia." Area 28(2):115-
123.
Greenbaum, Thomas L. 1988. The Practical Handbook and Guide to Focus
Group Research. Lexington, MA: D.C. Heath and Co.
Grossberg, Lawrence. 1987. "Critical Theory and the Politics of Empirical Re-
search." In Mass Communication Review Yearbook, eds. Michael Gurevitch
and Mark R. Levy, 86-106. London: Sage.
Holbrook, Beverley, and Peter Jackson. 1996. "Shopping Around: Focus Group
Research in North London." Area 28(2): 136-142.
Johnson, Alan. 1996. "'It's Good to Talk': The Focus Group and the Sociological
Imagination." Sociological Review 4-4-{3 Aug):517-538.
Kasper, Anne S. 1994. "A Feminist, Qualitative Methodology: A Study of Women
with Breast Cancer." Qualitative Sociology 17(3):263-281.
Kelly, L., S. Burton, and L. Regan. 1994. "Researching Women's Lives or Studying
Women's Oppression? Reflections on what Constitutes Feminist Research."
In Researching Women's Lives from a Feminist Perspective, eds. Mary May-
nard and June Purvis, 27-48. London: Taylor and Francis.
Knodel, John. 1993. "The Design and Analysis of Focus Group Studies: A Practi-
cal Approach." In Successful Focus Groups: Advancing the State of the Art,
ed. David L. Morgan, 35-50. Newbury Park, CA: Sage.
Krueger, Richard A. 1993. "Quality Control in Focus Group Research." In Suc-
cessful Focus Groups: Advancing the State of the Art, ed. David L. Morgan,
65-85. Newbury Park, CA: Sage.
Krueger, Richard A. 1988. Focus Groups: A Practical Guide for Applied Research.
Newbury Park, CA: Sage.
Longhurst, Robyn. 1996. "Refocusing Groups: Pregnant Women's Geographical
Experiences of Hamilton, New Zealand/Aotearoa." Area 28(2):143-149.
Lunt, Peter, and Sonia Livingstone. 1996. "Rethinking the Focus Group in Media
and Communications Research." Journal of Communication 46(2):79-98.
Malhotra, Valerie Ann. 1984. "Research as Critical Reflection: A Study of Self,
Time, and Communicative Competency." Humanity and Society 8(4):468-
477.
Mies, Maria. 1983. "Towards a Methodology for Feminist Research." In Theories
of Women's Studies, eds. Gloria Bowles and Renate Duelli Klein, 117-139.
Boston: Routledge.
Millen, Dianne. 1997. "Some Methodological and Epistemological Issues Raised
by Doing Feminist Research on Non-Feminist Women." Sociological Re-
search Qnline, available at http://www.socresonline.org.Uk/socresonline/2/
3/3.html.
Modleski, Tania. 1991. Feminism Without Women: Culture and Criticism in a
"Postfeminist" Age. New York: Routledge.
Montell, Frances. 1997. "Women and Sex Talk: Popular Culture and Normative
Sexuality." Ph.D. dissertation. University of California, Santa Barbara.
Focus GROUP INTERVIEWS 71
Morgan, David L. 1996. "Focus Groups." Annual Review of Sociology 22:129-52.
, ed. 1993. Successful Focus Groups: Advancing the State of the Art.
Newbury Park, CA: Sage.
-. 1988. Focus Groups as Qualitative Research. Newbury Park, CA: Sage.
Morgan, David L., and Richard A. Krueger. 1993. "When to Use Focus Groups and
Why." In Successful Focus Groups: Advancing the State of the Art, ed. David
L. Morgan, 3-19. Newbury Park, CA: Sage.
Morgan, David L., and Margaret T. Spanish. 1984. "Focus Groups: A New Tool for
Qualitative Research." Qualitative Sociology 7(3):253-270.
Nichols-Casebolt, Ann, and Patricia Spakes. 1995. "Policy Research and the
Voices of Women." Social Work Research 19:49-55.
Opie, Anne. 1992. "Qualitative Research, Appropriation of the 'Qther' and Em-
powerment." Feminist Review 40:52-69.
Padilla, Raymond V. 1993. "Using Dialogical Research Methods in Group Inter-
views." In Successful Focus Groups: Advancing the State of the Art, ed. David
L. Morgan, 153-166. Newbury Park, CA: Sage.
Potter, John, and Margaret Wetherell. 1987. Discourse and Social Psychology:
Beyond Attitudes and Behavior. London: Sage.
Reinharz, Shulamit. 1992. Feminist Methods in Social Research. New York:
Qxford University Press.
Smith, Dorothy E. 1987. The Everyday World as Problematic: A Feminist Sociol-
ogy. Boston: Northeastern University Press.
Vaughn, Sharon, Jeanne Shay Schumm, and Jane Sinagub. 1996. Focus Group
Interviews in Education and Psychology. Thousand Oaks, CA: Sage.
West, Candace, and Don H. Zimmerman. 1987. "Doing Gender." Gender and
Society l{2):125- 151.
Zeller, Richard A. 1993. "Focus Group Research on Sensitive Topics: Setting the
Agenda Without Setting the Agenda." In Successful Focus Groups: Advancing
the State of the Art, ed. David L. Morgan, 167-183. Newbury Park, CA: Sage.