gender studies — Blog — FromTheLabBench

Do facts convince?

It’s the classic question, which in science communication is often answered with a qualified “no.” Especially when it comes to the question of whether facts or evidence impact attitudes in the direction of the evidence.

In an interesting study from June 2015, published in Psychology of Women Quarterly, Moss-Racusin and colleagues investigated public comments on three news/blog articles reporting evidence of gender bias among science faculty. The news/blog articles reported on a 2012 scientific study demonstrating that science faculty favor male students over female students who apply to work in their labs.

Moss-Racusin and colleagues found a gender effect in comments posted on the three news/blog articles. They found that men were more likely than women to post negative comments in response to these articles (such as justifications or trivializations of gender bias) whereas women were more likely than men to post positive responses. Positive responses acknowledged the difficult reality of gender bias or called for social change, for example.

“Researchers have called for a scientific approach to the development, testing, and implementation of interventions designed to reduce bias and boost diversity by educating the STEM community about pernicious biases.” – Moss-Racusin et al. 2015

Plenty of research highlights the negative impacts of gender bias on women in STEM fields. Women in STEM are often evaluated as less qualified than men with the same qualifications on paper. Some research has also focused on how to address gender bias in STEM (science, technology, engineering and math) fields. However, as Moss-Racusin and colleagues point out, little research looks at the effect of confronting individuals about this gender bias. While some previous research suggests that clear examples of gender bias might influence attitudes in a positive direction, evidence of gender bias in STEM might also threaten some people’s beliefs about the fair behavior of scientists – or of themselves.

In a gender context, men who were confronted about their own sexism subsequently demonstrated compensatory efforts and reduced usage of sexist language (Mallett & Wagner, 2011). This evidence suggests that highlighting the existence of personal biases can reduce expressions of racial and gender prejudice. […] [However] relatively few studies have examined peoples’ naturalistic responses to scientific demonstrations of bias among others or within institutions. – Moss-Racusin et al. 2015

Moss-Racusin and colleagues analyzed “a large sample of 831 comments responding to prominent journalistic articles discussing social scientific, experimental evidence of bias in STEM.” The comments appeared on articles posted on the New York Times, on a Discover magazine science blog and on the IFL Science Facebook page. The evidence these articles discussed was data from a 2012 paper on STEM gender bias by Moss-Racusin and colleagues.

When possible, the researchers coded each comment for commenter gender and whether the commenter worked in a STEM field. Roughly 40% of the New York Times and Discover blog comments with an identifiable gender came from women, while 60% of the Facebook comments with an identifiable gender came from women. The researchers found 433 negative comments and 754 positive comments, total. 503 comments (61% of total) reflected only positive responses, while 245 comments (29% of total) reflected only negative responses.

On the negative side, the researchers found everything from sexist comments, to comments that justified bias (e.g. ‘‘The successful males I train simply seem to be hungrier and more willing to make the personnel sacrifices required to get ahead of the competition’’), to comments that disagreed with the results of the study.

On the positive side were comments agreeing with the results and constructive comments such as calls for social change and expressions of opinion change (e.g. ‘‘Thought this ‘gender gap’ had stopped… yikes’’ and ‘‘Sometimes it seems like we’re making progress, but then I see things like this and wonder what it takes to overcome seemingly pervasive implicit biases?’’).

Frustratingly, the gender of the commenter made a difference.

[M]any instances of negatively-valenced reactions (such as expressing sexist comments about women, justifying the results with biological explanations, justifying the results with non-biological explanations, justifying bias by arguing that women are perpetrators, disagreeing with results by arguing that gender bias does not exist, and criticizing social science, the study, or researchers) were more likely to be exhibited by men than women. – Moss-Racusin et al. 2015

Roughly 79-88% of male commenters justified gender bias in their comments, while only 12-21% of female commenters did the same. Roughly 65-85% of male commenters disagreed with the results, while only 15-35% of female commenters did the same.

Whether the commenter worked in a STEM field did not make a difference to the nature of his/her comment(s).

Subsequent research has revealed similar findings to Moss-Racusin and colleagues'. In another PNAS study, researchers found via several experiments that "men evaluate research that demonstrates bias against women in STEM less favorably than do women—or, that women evaluate it more favorably." And this was largely among men working in STEM fields - or scientists trained to view evidence objectively.

What do these findings mean? While it is encouraging that in the Moss-Racusin study a majority of the reader comments were positive, the commenter gender effect is concerning. Moss-Racusin and colleagues suggest that men may resist acknowledging gender bias “in order to maintain their own privileged position in the social hierarchy.”

Without the large number of women recognizing gender bias, there would be more instances of negative than positive reactions. This pattern points to the importance of gaining a better understanding of the role of observers’ own gender in shaping responses to demonstrated gender bias. – Moss-Racusin et al. 2015

[E]ven when presented with fairly stark experimental evidence of gender bias, many commenters responded by denying the evidence or justifying the existence of gender bias. This suggests that people may seek to avoid the threats associated with acknowledging that gender bias undermines the fairness of the existing academic system. – Moss-Racusin et al. 2015

No study is without limitations. This study used a non-random sample and only investigated comments on articles related to a single scientific study. However, it still provides evidence that when it comes to interventions for gender bias in science, facts don’t necessarily speak for themselves, especially among men.

If scientific evidence isn’t effective, how can we help men in STEM acknowledge and act upon demonstrations of gender bias while not threatening them into a defensive stance? Storytelling? More personal demonstrations of gender bias? Using fiction to explore realities for women in STEM? Games or simulations that deliver gender inequity information?