Innovative sampling and participant recruitment in sexuality research

Abstract

Although making significant contributions to empirical and theoretical understanding of sexuality in society, qualitative research on lesbian, gay, bisexual, and transgender (LGBT) individuals encounters problems in participant recruitment and sampling. Participants for research on LGBT individuals are often recruited through counseling services and snowball sampling with the result that samples tend not to be representative of the general population—a significant problem given that it is a common practice for scholars to generalize from multiple sets of qualitative research. In this essay, I draw on a research project that used innovative recruitment procedures to argue for a plurality of recruitment procedures in future research on sexualities and personal relationships.

Keywords

Bisexuality LGBT populations marginalized groups qualitative research sampling

Qualitative research has had a profound influence on our understanding of sexualities in contemporary cultures (e.g., Anderson, 2010; O’Neill, 2001; Plummer, 1995). The focus of qualitative research on contextual meaning and lived experience illuminates the lives of lesbian, gay, bisexual, and transgendered (LGBT) individuals and helps to develop our understanding of the nature and reproduction of sexuality as a mode of power (e.g., Rubin, 2012). Much of this contribution has come from a feminist perspective (e.g., DeVault, 1996; Ramazanoglu & Holland, 2002), which can be understood as a methodological approach that privileges analysis of gender and sexuality that is deeply rooted in the lived experiences of these marginalized groups (see Cook & Fonow, 1986).

Feminists have recognized the importance of quantitative research (Thompson, 1992), particularly in terms of documenting inequalities (Sprague & Zimmerman, 1993). However, Cancian (1992) argues that there are particular techniques of feminist methodology that especially lend themselves to a qualitative approach—with both feminist and qualitative approaches focused on the social meaning and lived experience of oppressed people’s lives. Indeed, these perspectives have tended to locate their importance in the conceptual analysis that develops from rich data and thick description (Ramazanoglu & Holland, 2002). Questioning the existence of “objective facts,” feminist qualitative research enables inductive theory building and provides explanations for the occurrence of trends related to gender and sexuality (Delamont, 2003).

However, the inclusion of sexuality as a central component of a feminist approach has been controversial. Although much feminist research examines both gender and sexuality (e.g., O’Neill, 2001), Rubin (2012) contends that sexuality has historically been excluded from consideration because it was deemed to detract from the focus on gender. Diamond (in press) argues that this erasure is still present in contemporary relationships research, and Manning (2013) highlights the particular troubles faced by academics who use qualitative methods in researching sexuality and personal relationships—with academics needing to defend both the sexuality and qualitative components of the research.

One of the central tensions between qualitative and quantitative research is the issue of generalizability—a particularly pertinent one for research on sexualities, given the difficulties of recruiting representative samples of LGBT individuals (Meyer & Wilson, 2009; Savin-Williams, 2001). Although qualitative scholars’ rejection of generalizability ostensibly seems to bypass these problems, the common practice of citing two or three qualitative studies from similar demographic groups and assuming they have relevance for other research indicates that participant recruitment and sampling is still of central importance in qualitative research. Thus, in this essay, I explore these problems by drawing on a current research project on bisexual men to call for innovative recruitment techniques in future research.

Researching bisexuality

Recruiting LGBT individuals as research participants is often fraught with difficulty, not least because publically identifying as such can result in stigma and discrimination (Meyer & Wilson, 2009). Savin-Williams (2001) highlighted the difficulties in recruiting sexual minority populations that make attaining a representative sample particularly difficult: Many sexual minority individuals are not out to other people, while others reject dominant sexual identity categories (including LGBT), and some only recognize their desires later in life. Furthermore, some LGBT individuals will only feel comfortable coming out to particular people or to support groups or counselors (Walby, 2010). Thus, it can be difficult to reach sexual minorities who are not part of support networks or LGBT communities.

Given the historical discrimination from both gay and straight communities (Weiss, 2011), recruiting bisexual participants is even harder. Bisexuality is often erased from lesbian and gay communities (Klein, 1993). Hartman (2011) suggests that the difficulty of locating a sizeable sample of a marginalized group (bisexuals), within an already-marginalized group (LGBT people), is like finding a needle in a haystack. When qualitative researchers have been able to recruit bisexuals in sizeable numbers, they have had to rely on snowball sampling and other forms of nonprobability sampling (e.g., Burleson, 2005). Thus, rather than recruit a group of bisexuals from a diversity of ethnic, class, and geographical backgrounds, academics have had to rely on existing networks of bisexuals and those within particular social milieus. This mirrors the critique of research on sexual minorities more generally (Savin-Williams, 2001), with the issue being exacerbated because of the stigmatized position of bisexuality within many of these social settings (Hartman, 2011).

A significant body of multiparadigmatic research sought to ameliorate this issue by grouping together the experiences of bisexuals with other minority sexual identity categories (see Worthen, 2013a). Although this increased the sample size to enable statistical analysis (or, indeed, to recruit “enough” participants for in-depth interview), it did not resolve the broader issue as it lost the specificity of bisexuality and spoke instead to a broader LGBT (and frequently only LGB) experience.

These forms of participant recruitment are problematic because of the common practice of citing several qualitative studies as evidence for a broader social trend. Thus, if qualitative researchers, as a group, use similar recruitment methods, this can have the result that the body of literature can privilege particular narratives or experiences (particularly of victimization), excluding those from people who are harder to reach. Multiple research projects using the same recruitment methods, even from a diverse range of social demographics, will thus present a skewed version of bisexuals’ experiences. It is for these reasons that when colleagues and I sought to examine the experiences of openly bisexual men, as part of our broader research agenda regarding the changing nature of masculinities in British and American cultures, we decided that new forms of participant recruitment were required.

The project design

When designing our research examining how self-identified openly bisexual men experience their identity and their social and personal relationships, a key aim was to recruit a broader sample than much of the existing research on bisexuals (e.g., Burleson, 2005). The then recent literature was still documenting the prevalence of biphobia and stigma toward bisexuals (e.g., Barker, Bowes-Catton, Iantaffi, Cassidy, & Brewer, 2008) and not reflecting the liberalization of attitudes toward homosexuality found in our own and others’ research (Anderson, 2009; Keleher & Smith, 2012; McCormack, 2012). We desired to investigate whether this was the result of problematic participant recruitment or an effect of the differences between being gay and bisexual. Accordingly, we sought to avoid recruiting bisexual men from traditional routes, including snowball sampling. This meant that we would not recruit men directly from counseling services, support groups for LGBT individuals, or bisexual organizations.

With these traditional routes of data collection rejected, we had to be innovative in recruiting participants. The central concern was being able to recruit a sufficient number from the general population: Given that 1.4% of the population is estimated to be bisexual (Gates, 2011), this required advertising to a potentially very large number of people. We first turned to the Internet, given that it has become a primary way of linking with LGB people (Harper et al., 2009). We tried recruiting participants through a range of social networking sites but found this to be limited in success, with people either unwilling or unable to meet in the time frame available. Many participants were unreliable and others were primarily seeking sex (cf., Worthen, 2013b).

We thus resolved that it was necessary to recruit men from dense public spaces and that we needed to recruit participants from places where thousands of people could be solicited. Accordingly, we took to the crowded city streets and recruited bisexual passersby. While bisexual men recruited from these areas might use LGBT themed webpages, visit gay clubs or even see a counselor, we were not selecting them from these venues. We called for bisexual men to be interviewed for academic research, and we interviewed those who were eligible immediately. We restricted our sample to men who self-identified as bisexual and who were open about their sexual identity to at least one family member or friend.

We also had age brackets to enable a cohort analysis (see McCormack, Anderson, & Adams, in press). We determined these categories such that the participants experienced their adolescence in particular social contexts of high cultural homophobia of the late 1980s, decreasing homophobia in the mid-1990s, and more positive attitudes toward homosexuality of the late 2000s (Loftus, 2001; Keleher & Smith, 2012). Thus, we categorized three age cohorts for analysis with men aged 36–42, 25–35, and 18–23.

Collecting data in three cities, we varied locations to include a range of areas (e.g., commercial, tourist, etc.). By doing this, we recruited 90 bisexual men who came from a range of ethnic and class backgrounds, and the great majority was not part of gay social networks, therapy, or bisexual organizations (see McCormack, Adams, & Anderson, 2013). While a minority of participants were part of these networks (something we asked during interview), the key difference from earlier research is that we were not restricted to members of these groups.

Apart from recruitment, our research design was rather traditional. We split participants into three age cohorts to examine for intercohort differences, and we used semi-structured interviews. We undertook open and selective coding of interview transcripts to identify the patterns within the stories (Urquhart, 2013), and emerging themes were related back to the original transcripts, and their internal coherence checked (Braun & Clarke, 2006). This cohort-difference analysis was a key part of examining for the influence of a liberalization of attitudes toward homosexuality. The innovative participant recruitment enabled us to examine these issues with a more representative sample of bisexual men. Given the homogeneity of recruitment strategies in sexualities research, our approach contributes to there being a plurality of recruitment processes and this, I argue, will help provide feminist qualitative scholars with a better overall understanding of the operation of sexuality in society.

Evaluating the approach

In evaluating the research process, my colleagues and I determined that it was a successful recruitment technique (see McCormack et al., 2013). The participant recruitment strategies enabled 90 bisexual men to be interviewed in the period of time available for the study, with a large diversity of bisexual men. Participants ranged from an 18-year-old, who had only recently come out, to men who had been out for 30 years; some frequented the gay scene while others had never been; two thirds of participants were Black or Latino, and only a small minority had used counseling services. There was also a clear range in wealth and level of education.

A key benefit was that the approach made the research accessible to a large number of people. There were a large number of inquiries from people about whether they fit the criteria, how long the interviews would take, and how we defined bisexuality. Several participants would not have responded to other forms of recruitment (particularly posters and emails) and some of the younger participants stated that seeing us “in the flesh” dispelled potential fears they had in talking with us. And although some potential participants may have decided against approaching us because of the public nature of our recruitment, this corresponds with the aim to exclude closeted individuals.

This form of participant recruitment also established an alignment between the goals of the project and the methodological framework. Given that is desirable for a qualitative methodology to, in some way, reflect the social realities of participants, recruiting bisexual men from crowds recognizes the diversity of experiences of bisexual men that transcends LGBT populations and counseling services. This diversity of bisexual men would not have occurred if traditional forms of recruitment had been used. This is not to say no potential weaknesses of this approach exist, but it is to recognize that more diversity in recruitment processes and participants’ backgrounds will only strengthen qualitative research on sexualities.

Concluding thoughts

Within the body of literature on how to develop a strong methodological framework, there is a growing call for methodological innovation (e.g., Taylor & Coffey, 2008). Although funding bodies in the United Kingdom have shown an interest in fostering such innovations (Gwyther & Possamai-Inesedy, 2009), traditional thinking and the small group of academics that constitute the panels for funding bodies mean that innovation only occurs in particular structured ways that rarely transgress dominant methodological paradigms (Taylor & Coffey, 2008). The expansion of the role of ethics review panels has also had a chilling effect on the development of new methodological processes (Haggerty, 2004)—an effect that may be exacerbated on research on sexualities because the topic is frequently determined to be “sensitive” or “dirty work” (Irvine, in press).

I have drawn on my experience of innovation in participant recruitment not as a call for all academics to “take to the streets” (McCormack et al., 2013) but as a call for further developments in how participants are recruited. It would not, for example, be suitable to recruit victims of domestic violence in the ways described above; yet the persistence of sampling issues in qualitative research on sexualities demands consideration. Given the need for research methods to be sensitive to the social context of any research project, innovation in recruitment processes will need to be developed by experts in the particular field. But I argue that what is vital as we seek to develop feminist qualitative research on sexualities is that we ensure that there is a plurality of recruitment processes. In a research context where qualitative research is generalized to subsections of society, and where the majority of qualitative research excludes particular narratives and experiences, a narrow set of recruitment procedures can have negative consequences. Although new recruitment strategies will require particular scrutiny, they should also be welcomed by the academy as contributing to a more comprehensive understanding of bisexuality, LGBT individuals, and sexuality more broadly conceived.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Anderson

(2009). Inclusive masculinity. London, UK: Routledge.

Anderson

(2010). “At least with cheating there is an attempt at monogamy”: Cheating and monogamism among undergraduate heterosexual men. Journal of Social and Personal Relationships, 27, 851–872.

Barker

Bowes-Catton

Iantaffi

Cassidy

Brewer

(2008). British bisexuality: A snapshot of bisexual representations and identities in the United Kingdom. Journal of Bisexuality, 8, 141–162.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101.

Burleson

W. E.

(2005). Bi America. New York, NY: Harrington Park.

Cancian

(1992). Feminist stances: Methodologies that challenge inequality. Gender & Society, 6, 623–642.

Cook

J. A.

Fonow

M. M.

(1986). Knowledge and women’s interests: Issues of epistemology and methodology in feminist sociological research. Sociological Inquiry, 56, 2–29.

Delamont

(2003). Feminist sociology. London, UK: Sage.

DeVault

M. L.

(1996). Talking back to sociology: Distinctive contributions of feminist methodology. Annual Review of Sociology, 22, 29–50.

10.

Diamond

L. M.

(in press). Where’s the sex in relationship research? In Stewart

R. S.

(Ed.), Let’s talk about sex. Sydney, Australia: Cape Breton University.

11.

Gates

G. J.

(2011). How many people are lesbian, gay, bisexual, and transgender? (Census & LGBT Demographic Studies). Retrieved from University of California, Los Angeles, Williams Institute website: http://williamsinstitute.law.ucla.edu/wp-content/uploads/Gates-How-Many-People-LGBT-Apr-2011.pdf

12.

Gwyther

Possamai-Inesedy

(2009). Methodologies a la carte: An examination of emerging qualitative methodologies in social research. International Journal of Social Research Methodology, 12, 99–115.

13.

Haggerty

K. D.

(2004). Ethics creep: Governing social science research in the name of ethics. Qualitative Sociology, 27, 391–414.

14.

Harper

G. W.

Bruce

Serrano

Jamil

O. B.

(2009). The role of the internet in the sexual identity development of gay and bisexual male adolescents. In Hammack

P. L.

Cohler

B. J.

(Eds.), The story of sexual identity (pp. 297–326). New York, NY: Oxford University.

15.

Hartman

J. E.

(2011). Finding a needle in a haystack: Methods for sampling in the bisexual community. Journal of Bisexuality, 11, 64–74.

16.

Irvine

(in press). Is sexuality research “dirty work?” Institutionalized stigma in the production of sexual knowledge. Sexualities.

17.

Keleher

Smith

E. R.

(2012). Growing support for gay and lesbian equality since 1990. Journal of Homosexuality, 59, 1307–1326.

18.

Klein

(1993). The bisexual option. New York, NY: Haworth Press.

19.

Loftus

(2001). America’s liberalization in attitudes toward homosexuality, 1973 to 1998. American Sociological Review, 66, 762–782.

20.

Manning

(2013). Interpretive theorizing in the seductive world of sexuality and interpersonal communication: Getting guerilla with studies of sexting and purity rings. International Journal of Communication, 7, 2507–2520. Retrieved from http://ijoc.org/index.php/ijoc/article/view/2250/1023

21.

McCormack

(2012). The declining significance of homophobia: How teenage boys are redefining masculinity and heterosexuality. New York, NY: Oxford University.

22.

McCormack

Adams

Anderson

(2013). Taking to the streets: The benefits of spontaneous methodological innovation in participant recruitment. Qualitative Research, 13, 228–241.

23.

McCormack

Anderson

Adams

(in press). Cohort effect on the coming out experiences of bisexual males. Sociology.

24.

Meyer

I. H.

Wilson

P.A.

(2009). Sampling lesbian, gay, and bisexual populations. Journal of Counseling Psychology, 5, 23–31.

25.

O’Neill

(2001). Prostitution and feminism: Towards a politics of feeling. London, UK: Polity.

26.

Plummer

(1995). Telling sexual stories. London, UK: Routledge.

27.

Ramazanoglu

Holland

(2002). Feminist methodology. London, UK: Sage.

28.

Rubin

(2012). Deviations: A Gayle Rubin reader. Durham, NC: Duke University.

29.

Savin-Williams

R. C.

(2001). A critique of research on sexual-minority youths. Journal of Adolescence, 24, 5–13.

30.

Sprague

Zimmerman

(1993). Overcoming dualisms: A feminist agenda for sociological methodology. In England

(Ed.) Theory on gender/feminism on theory (pp. 255–280). New York, NY: Aldine.

31.

Taylor

Coffey

(2008). Innovation in qualitative research methods: Possibilities and challenges. Cardiff, UK: Cardiff University School of Social Sciences.

32.

Thompson

(1992). Feminist methodology for family studies. Journal of Marriage and the Family, 54, 3–18.

33.

Urquhart

(2013). Grounded theory for qualitative research. London, UK: Sage.

34.

Walby

(2010). Interviews as encounters: Issues of sexuality and reflexivity when men interview men about commercial same-sex relations. Qualitative Research, 10, 639–657.

35.

Weiss

(2011). GL versus BT: The archaeology of biphobia and transphobia within the U.S. gay and lesbian community. Journal of Bisexuality, 11, 498–502.

36.

Worthen

M. G. F.

(2013a). An argument for separate analyses of attitudes toward lesbian, gay, bisexual men, bisexual women, MtF and FtM transgender individuals. Sex Roles, 68, 703–723.

37.

Worthen

M. G. F.

(2013b). An invitation to use craigslist ads to recruit respondents from stigmatized groups for qualitative interviews. Qualitative Research. doi:10.1177/1468794113481791