“According to my clock it’s 1100EST, let’s get this #DiversityJC rolling!”
And with that, we were off.
You know, it’s funny how things get started these days (especially for a twitter newbie like myself). One little tweet can ring so true, and next thing you know, you’re talking with complete strangers over the interwebs. But they don’t really feel like strangers, do they?
The Diversity Journal Club (#DiveristyJC) was born of conversations between Jonathan Goaya (@jkgoya), Doctor PMS (@Doctor_PMS) and myself (@DrEmilySKlein). For our first paper, we discussed the one that got it all started (learn more here, and read the article here. The premise of the paper was that women get more positive feedback on their grant proposals than men do.
And our thoughts?
First and foremost, many had concerns with the methods used. Of course, as scientists assessing study in an unfamiliar field, we tempered that with a healthy dose of “well, the methodology was new for me so…”. Regardless, we generally agreed that the methods had to be somewhat addressed to assess the limitations of the study and critique results.
Concerns with methods included weighting very different words with the same value (i.e. being a genius is apparently the same as having a knack for something), distribution of words used, and how words were categorized and what those categories meant. Re-doing the study as a double-blind would also help, as well as a more narrow choice of words in general (such words like “queen*” and “king*” seemed… excessive, and were – let’s hope – doubtful found in a single response.) Authors should have limited the words addressed by their analysis to those that may have actually been used by reviewers instead of what appears to be a standard list of YAY and NAY words. Also, pretty small sample size. Of course, as Beth Hellen (@PhdGeek) pointed out, it could be a pilot study. In any event, we agreed that the methods needed some work, but we’d all like someone versed in them to provide a better critique.
However, a note: My understanding was the words used did come from other, published studies and methodologies. Either way, there did appear to be a underlying bias in the words used – perhaps indicating a larger, systemic problem with the words we use to define people along the gender binary, and how we code and value those words. As Cheng H. Lee (@chenghlee) noted, “In perhaps not-so-subtle way, this could have incorporated broader biases about masculine vs feminine words into the analysis.”
Moving on to those “larger issues.” First and foremost, why would women be praised more than men? Especially given the increasing evidence that women are seen as less competent and are overlooked for jobs, tenure, even mentoring, and described using less capable words even by people trying to get them a job/money (e.g. in letters of rec)?
Given the findings argued by this study… What’s going on with grants?
Well. Perhaps we should be looking at it a bit differently. As Jonathan Goya (@jkgoya), the premise could be seen another way: “Do reviewers use different language to review women and men?” Ahhhh… now we’re getting somewhere…
For the first potential explanation, an old favorite of mine: Chivalry and the gender binary. Ladies be all soft and sensitive, duh, and, moreover, a true gentleman is not rude to ladies. Yes that’s an exaggeration, but you get my drift: Men used “nicer” language when speaking with women than with men.
Although… the women were also better funded, according to the study, so there’s less evidence that men were just being nice and letting the ladies down easy. That said, Jonathan Goya (@jkgoya) still noted “from personal [experience], I’m pretty sure men speak to each other in much more directly critical language than between men and women.” Also personally believing this to be true, I’m unwilling to throw out this explanation as a possibility. In addition, R. Deborah Overath (@scienceknitster – yes on that name) pointed out that, given how awesome the words were for women, they should actually have scored better.
Alternatively, as Ruth Hufbauer (@hufbauer) suggested, maybe men are becoming more aware of their biases and are overcompensating or being too careful… ?
Another explanation: The bar is lower for women. Many people are just surprised (surprised!) when women write a stellar grant proposal. They don’t expect it, so are more glowing in response. Moreover, Jonathan Goya noted “…to get the same scores, women have to really wow the reviewers” and Lauren Sakowski (@LaSaks87) reiterated that the women may have put in “extra effort for the same funding/recognition”.
Basically, we either don’t expect women to do well, or they have to put in the extra effort for the same recognition. Or both.
As some additional evident, Beth Hellen (@PhdGeek) noted that there is a larger difference between the positive and negative words used for funded women, but little difference for men. Perhaps the actual content of proposals from men somehow counted more. In addition, again to R. Deborah Overath’s (@scienceknitster) point that women possibly should have gotten better scores given the words used in their reviews. Perhaps the bias is not in the language used, but in the actual score that, you know, actually matters.
Of course, perhaps women just write better grants.
Or… they have more experience with them and exactly what reviewers are looking for. We all know grants can be formulaic. We also know women are less likely to be tenured or in leadership positions. Maybe these women have spent more time in soft-money jobs that simply require more grant writing to stay afloat in the world. Consequently, they’re just better at it.
Yet, again, as scientists assessing a study from a generally unfamiliar field, we craved more information. We speculated on additional variables that may help us piece apart the methods and the results, and really assess why this paper, on its surface, seems to contradict what more and more studies are telling us, and what many if not most of us know from experience: women are biased against in the sciences. We are still going up the stairs, when men have had an elevator.
Finally, Ruth Hufbauer reminds us that, yes, we’re scientist, but we’re also human. Just like the reviewers and the authors. It’s difficult to do things, like be on review panels, without either being biased in some way, or at least being worried about it.
This would be the final point I’d make, and I hope it’s one we come back to repeatedly in the Diversity Journal Club: As Dr. Wrasse (@labroides) asked: does a fish know it’s in water?
How do we recognize our own biases?
And what do we do about them when we start to figure them out? How do we deal with them in others?
That was where the Diversity Journal Club left off: contemplating how we become more self-aware, how we educate ourselves and others, and how we raise awareness.
We hope we’re doing our little part of this, in our own little corner of the twitterverse. Until next time, kids…
Lastly, a quick but very important note from Jonathan Goya and Beth Hellen that I was absolutely guilty of: avoid using karyotype (e.g. XX or XY) when discussing men or women. This automatically assumes sex, and negates a person’s right to self-identify their gender. Therefore, please instead use M, F or alternative. And please keep these critical hints coming!
The next Journal Club will be next week! We will post a paper on Monday 9/29 to review on Friday 10/3 at 11am EST. Have one we should read? Let us know! Since we just did one on gender, let’s have a new diversity topic for the next one – any and all welcome.
Thank you to all participants (and give them a follow, they’re awesome!)*
Doctor PMS (@Doctor_PMS) – the one who got us started with one little tweet!
Jonathan Goya (@jkgoya)
Dr. Wrasse (@labroides)
Beth Hellen (@PhdGeek)
biochem belle (@biochembelle)
Ruth Hufbauer (@hufbauer)
Cheng H. Lee (@chenghlee)
Lauren Sakowski (@LaSaks87)
Deborah Overath (@scienceknitster)
Ian Street (@IHStreet)
Storify of the #DiversityJC
… and anyone else who checked in and followed the discussion. Again, we invite any and all participants, as long as you read the article and no trolling, please (although that just means we’ll ignore you. Which is a bummer. For you.)
‘Til next time!
*Let me know if I got any names wrong or you have trouble finding someone!