Thursday, March 26, 2015

Gender Imbalance in Discussions of Best Research Practices

Over the last couple of weeks there have been some really excellent blog posts about gender representation in discussions of best research practices. The first was a shared Email correspondence between Simine Vazire and Lee Jussim. The second was a report of gender imbalance in discussions of best research practices by Alison Ledgerwood, Elizabeth Haines, and Kate Ratliff. Before then (May 2014), Sanjay Srivastava wrote about a probable diversity problem in the best practices debate. Go read these posts! I'll be here when you return.

Welcome back!

Let me first preface this blog post by saying that I am not an expert on gender issues. That said, diversity and representation issues are important to me (and probably to society). These issues are also important to our ongoing discussion of best research practices: A discussion about how we should do research would be better if it included all the kinds of people who conduct said research. Now for my observations:

(1) Gender imbalance is real in discussions of best research practices
When I think back to conversations I've had about gender representation, I find that many of these conversations get derailed at the outset because people are not convinced that imbalances based on gender actually exist in the first place. The argument usually goes that we might not remember the actual gender distribution or there might be some other non-gender related reason for a lack of representation.  Thankfully, Ledgerwood et al., provide very clear data suggesting that a gender imbalance is real and observable in discussions of best research practices online, in journals, or at conferences. These data suggest that our memories of these discussions as imbalanced are not wrong. And, because the imbalance occurs across domains, the data suggest that it may very well be something related to gender that is causing the imbalance.

I will admit that, based on these data alone, we are not able to know the causes of gender imbalance (e.g., individual v. institutional), nor the effects. But these data do suggest that imbalance exists. Perhaps having a best practices conversation with equal representation won't look any different than the one happening now--it's important to find out.

(2) Imbalance matters
There are many sound/empirically-supported reasons why a gender imbalance in discussions of best research practices matter to our science. Much of this was reviewed marvelously in the Ledgerwood et al., piece. Some of my favorite reasons include that male over-representation might allow the tone of the discussion to be more combative/offensive, where less consensus can be reached (e.g., "Nut up or shut up."; Dovidio et al., 1988), and that diversity of viewpoints can actually improve the decisions that we make (Smith et al., 2006).

On the more pragmatic side of things, gender imbalance in discussions of best research practices matter because these discussions will change our field in the future. If women are not invited to this discussion explicitly (i.e., not given invitations to speak) or implicitly (i.e., the costs of speaking up and getting it wrong are harsher for members of lower status groups; Brescoll & Uhlmann, 2008), then the field is negatively impacted. Given that women represent a majority of the early career positions in SPSP, our discussion of best research practices is missing the voices of many of our newest and brightest colleagues. Bummer for us!

(3) Imbalance is a bigger problem at the top
Recall that in point #2 I mentioned that women represent the majority of early career psychologists. You might take that, as others have, to suggest that we don't have a problem with gender imbalance in our field. Hooray! Gender discrimination is over! Or maybe, wow, we should be concerned (instead) about men having a fair chance at early career positions in psychology.

If that was your reaction, I would like you to look at those numbers again, except now look at full members of SPSP. There you see the status quo: Men represented more than women. I expect that a gender imbalance of this kind is a potential problem for our field: People at the full member career stage are making many decisions in peer review, promotion and tenure cases, or about whom to invite for a symposium on best research practices. Reducing gender imbalance in positions of power is critical so that potential explicit/implicit discriminatory practices are addressed for the early career members of our field, who happen to be mostly women.

(4) Keep your anecdotes out of my science
Often, arguments about gender imbalance devolve into bludgeoning one another with anecdotes. One person might say "In my experience, you just need to take risks and everything will work out better for you." and another person might reply "I'm the biggest risk-taker in the world but I still face gender discrimination." I expect to field these kinds of arguments during Thanksgiving dinner, but not from scientists in a debate about the existence of gender imbalance in best research practices debates. One person's unique experiences do not invalidate another's (or whole studies, for that matter).

Too often, when we talk about gender imbalance in conversations of best research practices people (usually men, and I've made this mistake too) will chime in and say basically, "Don't worry. I don't know everything, but I ventured into the discussion of best research practices and you can too!" These comments ignore the possibility that people (e.g., women, people of color) face differential pressures and costs for speaking up in academic settings (e.g., Brescoll & Uhlmann, 2008).

(5) Discuss gender imbalance on its own terms
Finally, I want to acknowledge that issues of diversity in science are important for all different types of diversity. This includes race, gender, class, sexual orientation, political ideology, mental/physical disability, conceptual/direct replicators, etc... We can (and should) care about all forms of imbalance in our field and in discussions of best research practices.

However, we should discuss each diversity issue on its own terms: Discussions of gender imbalance that derail into discussions about whether we would care as much if the imbalance were, say based on political leanings or single/married status are counterproductive. This is because all diversity issues are not likely to be caused by the same underlying factor, or to be solved by the same magic bullet. If gender imbalance exists and people want to eliminate that imbalance, unique efforts can be deployed to deal with that unique imbalance. To illustrate, racial imbalance in SPSP's membership is large and systematic across all levels. It would be presumptuous of anybody to assume that solutions for improving gender imbalance would work for race, or to assume that race and gender imbalances arise from the same underlying process.


  1. Nice post. Regarding your last paragraph-- It is crucial to keep in mind intersectionality. Not all women social psychologists are white, straight, middle class, etc. So if, say, a woman of color wants to talk about how race affects her experiences, that isn't derailing.

    Of course, this is very different from your example of intentional derailing by drawing false equivalences.

    1. That's an important distinction. Thanks for sharing!