Eric Patridge (the president of oSTEM) worked with Ramón Barthelemy and Susan Rankin to reanalyze data from Campus Pride's 2010 State of Higher Education for LGBT People [PDF]. That survey had 279 responses from LGBTQ faculty members, 130 of which worked in STEM fields. Patridge and his coauthors separated out 83 faculty in the social sciences, leaving 47 STEM faculty members. Similarly to our survey, the Campus Pride study asked participants to rate how "out" they were on a scale of 1 to 5, and also asked how comfortable they were on campus as a whole, in their departments, and specifically in classroom settings.
The result is surprising, especially in comparison to our own data: of the STEM faculty who answered the Campus Pride survey, those who who rated their "outness" level as 4 or 5 were much more likely to say they were uncomfortable in their department.
Figure 1 from Patridge et al. (2014): Faculty members who are out to their colleagues are much more likely to describe themselves as "uncomfortable" in their departments.
(Patridge discusses the new paper, and this result in particular, in this interview with the Yale Daily News.)
In the most comparable analysis from our own study, we found that participants who described their workplaces as "welcoming," or said they were "treated the same" as their straight colleges, were much more likely to be out to their colleagues—exactly the opposite of the relationship Patridge et al. found between begin out and feeling comfortable. This is the graph from our preliminary analysis, showing a result that held up in the analysis we've now submitted for publication:
So how did we get such dramatically different results? The faculty participating in the 2010 Campus Pride study are a much smaller sample than the one we have for Queer in STEM, but this seems like a bigger difference than can just be explained by sampling effects.
I e-mailed Patridge shortly after we heard about the paper—he was good enough to send a PDF copy, when it turned out that the University of Minnesota doesn't subscribe to the journal—and he and I batted around some possible explanations:
The times they are a-changin'. Between 2010, when the Campus Pride survey ran, and 2013, when Queer in STEM launched, there's been a tremendous amount of concrete legal progress for LGBTQ Americans—Congress ended the U.S. military's "Don't Ask, Don't Tell" policy, the U.S. Supreme Court struck down critical portions of the Defense of Marriage Act, and marriage equality became legal in (depending on how you count them) something like half of the states where it is now the law. Maybe this ongoing, tectonic shift in the legal rights of queer Americans is reflected in a very different relationship between begin out and being comfortable in the workplace. But, again, it's hard to understand how even these big societal changes could completely reverse that relationship.
Life is different in the ivory tower. Patridge et al. only analyzed survey responses from academic faculty, while our survey includes responses from folks in the private sector and at non-degree-granting institutions. Maybe that accounts for the difference? But academics are still the largest single portion of our participants (69%), and they're no more or less likely to be out to their colleagues than non-academics. So it seems unlikely that the non-academics are completely altering our results.
Age isn't just a number. Again, Patridge et al. only analyzed survey responses from faculty, while Queer in STEM participants are at career stages from graduate school to retirement—with the largest group being grad students and postdocs, and under the age of 40. Maybe the relationship between being out and being comfortable differs at later career stages in academia? Among Queer in STEM participants, outness ratings in all contexts differ significantly with age cohort, with older participants being less out—and, indeed, only 14% of the faculty in the Campus Pride data set are out in professional settings, compared to 38% of Queer in STEM participants who gave their outness to colleagues ratings of 4 or more. However, even if we consider only Queer in STEM participants in academic workplaces who aren't students, we still find that participants who call their workplaces welcoming are more out to their colleagues.
Apart from those three possible explanations, we're left with the possibility that Queer in STEM participants represent a fundamentally different population from the folks who responded to the Campus Pride survey. This is a difficult hypothesis to test, because there isn't any obvious way we can imagine in which our sampling strategy differs that radically from the one used by Campus Pride. Campus Pride drew on its own membership list, but also recruited participants using online social networks, professional association contacts, and referrals by other participants—very similar to what we did. And the fact remains that there may be some additional factor that provides the explanation, but wasn't measured by either Campus Pride or Queer in STEM!
Considerations like this are potentially frustrating, but they're also important to keep in mind as we interpret the results of our analysis. How representative are the experiences described by Queer in STEM participants—of LGBTQA people working in STEM? Of LGBTQA professionals in general? Of LGBTQA Americans?
We do know that Queer in STEM participants are generally very open about their identities in personal contexts, and given some clever recent research that suggests that this isn't the case for gay men across the U.S., we have to consider that Queer in STEM participants may not be very representative of LGBTQA Americans as a whole. Even without that direct evidence, we might already expect that people who have graduate degrees in STEM fields, who are working in research and technology careers may have very different needs and concerns than folks who are harder to reach.
Patridge E, RS Barthelemy, & SR Rankin. 2014. Factors impacting the academic climate for LGBQ STEM faculty. Journal of Women and Minorities in Science and Engineering. 20(1):75-98. doi: 10.1615/JWomenMinorScienEng.2014007429.