Scout Director of Network for LGBT Health Equity Reporting from Intersectionality Working Group meeting of LGBT Population Research Center The Fenway Institute, Boston, MA
Panel 2 – Methodology for Intersectionality Research
Kimberly Balsam starts by laying out some challenges in each stage of the research (sampling, measurement, analysis, interpretation). She talks a bit about two different research projects she’s been finishing, the Rainbow Project and the Rainbow Women’s Health Survey. At least one study is still being published so I won’t talk about the findings as much, but instead focus on her methods. First, each study was mixed methods, combining a few strategies to get to the people they wanted to recruit. Respondent-driven sampling was used (one of the shiny stars of rare population research) as well as targeted sampling. There was an interesting subdiscussion here on interviewing, one researcher talked about a project where they used community members instead of graduate students to do the interviews, then even further, allowed the participants to choose their own interviewers. They found that most of the participants (all Black), chose sex and race matched interviewers. But African and Afro-Carribean interviewers were more likely to request white interviewers. Another research project was brought up where Black men in prison were being interviewed, and while they presented as heterosexual to the heterosexual interviewer, it all changed once the gay researcher got in the room. Everyone agrees, the researcher affects the data, but we don’t have enough information on how.
Next we get to see some advance results of Juan Battle’s national LGBT of color survey, named Social Justice Sexuality project. Like Kimberly, he used mixed methods for his sampling: quota sampling, venue sampling, snowball sampling, respondent driven sampling, the internet and partnering with a “whole host” of community based organizations. He worked for a year and a half to set up the partnerships. Then, they went to every different type of event imaginable, festivals, parties, special events, prides, rodeos, dinner parties, the choir directors caucus at the National Black Baptist Convention, everything they could find. Face it, “mixed methods” hardly describes the breadth of the methods, seems like it’s more like “every method you’ve ever heard of”. The result? Over 5,000 respondents. So what about findings? Well, I won’t say any results because they’re still coming out. But he brought up an interesting methodological point. Usually when you ask about importance of racial identity v. sexuality you get 90% of your sample saying, “no real diff”. So he approached this question differently, he asked “how important is your sexual identity?” in one part of the survey, then in a very different part “how important is your racial identity?” Fascinatingly, suddenly most of the sample said either race or sexuality was more important.
Now onto Mignon Moore, who I gotta give a big shout to because she was one of my dissertation advisors, w00t! She’s done a series of research studies on Black lesbians that are coming out in a new book very soon, Invisible Families: Gay Identities, Relationships, and Motherhood among Black Women. Again, what’s our theme? Mixed methods! Not only in recruitment but in data collection too. She talked about the particular value of surveying all of her sample but supplementing it with in-depth interviews for half, allowing greater exploration of the survey results. She has an interesting (and time saving) take on researching intersectionality: instead of comparing across race or gender categories, analyze the experiences of the people who live at those intersections. She asked questions about primacy of identity like Juan did and that brought up some interesting information: one identity can be used for self-definition, but a wholly different one can be used as a status that helps a person stick with a particular group.
The panel was rounded out by Brian Mustanski, who’s been working on a project with LGBT youth of color in Chicago. He’s really interested in the question of whether resiliency about race can be transferred to become resiliency about homophobia, but of course explores a lot of intersectionality on the journey to that goal. Again our theme, the study used mixed methods. They started with a survey (using the Race Coping Measure), added psychiatric interviews, mental health self reports, then specifically added open ended qualitative questions both at enrollment, and then again at one year followup. His discussion of the analysis was interesting, to integrate the different types of data collection, they ended up using a qualitative coding scheme for the whole project, to try to bring up themes across the different datasets. And he brings us right back to our theme from the first session, it’s all about context. When they asked about race they heard about neighborhoods, when they asked about neighborhoods, they heard about race, so in some cases, context really defines the issue.
Lisa Bowleg sums it up: “To do intersectionality research, you really have to do mixed methods research”. She challenges us to go further, incorporating more disciplines in the research braintrust, go to the economists to see what neighborhoods to study. But mostly, don’t get stuck on methodological limitations. We really are pioneering some interesting research strategies as each of us draws from our different resources to create our own particular mix of methods.
David Chae rounds out the discussion by bringing up the elephant in our room… we need more general population surveys to include measures of sexual orientation and gender identity. Hear hear, we don’t count until we’re counted!