Collaboration, Bias, and Tenure

I was pretty bummed to read this New York Times article about the prospects for women seeking tenure in Economics.

In case you don’t have time to read it all right now: in recent study, Economist Heather Sarsons found that in Economics, women received almost no credit (measured as increased likelihood of receiving tenure) for papers coauthored with a man. Men received nearly the same credit for papers coauthored with women or with men as for solo work.

Is this bias present in Math? I honestly don’t know. This study doesn’t say anything about a similar bias in Mathematics. It does tell me that such a bias could potentially exist. I feel naïve in saying that I was totally blind to the possibility–I hadn’t considered that this might be an issue. You might think I should have at least thought about it before, because I collaborate a lot in my work. I really like doing math with other people. Collaboration keeps me motivated, and, for me, discussion is part of the process of understanding.  All of my papers have multiple authors. Some are written with other women, some with mixed groups.

My most recent finished work was done with a male coauthor. We both worked very hard on the project, and it really does belong equally to both of us. My name is listed first, just because I’m earlier in the alphabet than he is. But because that is the convention in mathematics, nobody would assume I deserve more credit than he does. This study, though, made me begin to wonder if people will actually assume that he deserves all the credit.

Don’t worry, you say, that was Economics. What does it have to do with Math? Yes, I want to be soothed. However, I do consider the fact that Math and Economics are both heavily majority male fields with overlapping skill bases. Also, the author of the article (not the study) hypothesizes that one contributing cause could be that Economics paper authors are listed alphabetically. Hmm, maybe this is relevant to Math. Also, it brings up new questions for situations like those discussed in Adriana’s 2014 post about gender bias—did those women write papers with men, for which they did not receive “full credit”?

This study opens up some frightening possibilities, especially for frequent co-authors like me. I’m not up for tenure now, but when I am, how will my papers count? Will the papers I wrote with men mean anything? As a female researcher early in my career, should I avoid working with men?

To try and put some positive spin on this whole study, perhaps this is good news for research collaborations among women. Sarsons’ study found that women did receive equal credit for papers coauthored with only other women. In Economics, papers written with only women authors are taken seriously and count towards tenure. According to this study, women do not need male coauthors to bring weight to their research.

Outside of any consideration of co-author gender tenure bias, there are many reasons that women might want to collaborate with other women. With many fewer women than men in most areas of mathematics, in can be difficult for women within an area to even meet each other, let alone begin research projects together. This is one reason that groups like the Association for Women in Mathematics (which I just wrote about last time) and focused research workshops like Women In Numbers (WIN) exist.

Adriana Salerno has written about WIN on this blog before, so you may have heard about it. In fact, WIN is how I met Adriana in the first place, along with a lot of the other great women I know in number theory. For those of you who haven’t heard, Women in Numbers (WIN) was the name of a 2008 workshop organized by Kristin Lauter, Rachel Pries, and Renate Scheidler to help build a network of women Number Theorists. Since then there have been two more WIN workshops (all held at the Banff International Research Station), each bringing together women number theorists from all over the world to work very intensively on group projects led by experts in the field. These collaborations often continue after the workshop and lead to publications, in the peer-reviewed WIN proceedings volumes as well as in journals and other conference proceedings. The workshops are very challenging and ask a lot of the group leaders and participants, but the results are tangible and exciting. I attended the first WIN in 2008 and WIN3 in 2014, and three papers on my CV came out of WIN collaborations.

I bring WIN up here because I have heard many people question the value of a “separate” network for women in number theory.  Even though these publications are peer-reviewed, won’t papers from workshops like these be taken less seriously because only women are involved?  I think that the hopeful message from this study is, no, maybe not.  At least in Economics, working with other women is a good thing.

The WIN model has expanded to connect number theorists in Europe (WINE) as well as to women in other fields, including shape theory (WiSh), algebraic combinatorics (ACxx), topology (WIT), applied math (WhAM!), non-commutative algebra and representation theory (WINART). The AWM was recently awarded at $750,000 grant to further these focused research networks. As I learned at the JMM panel, the AWM is  trying to get the word out and encourage more women to get involved–by joining an existing group or by starting a focused research network in their own field. Here is a great website about all this.  I don’t think you even need a clever acronym—just a desire to connect with women in in the field.

This has been a thoughtful month for me, considering Sarsons’ work on the research front and the recent work on gender bias in student teaching evaluations (see Sara’s blog post from a few weeks ago) on the teaching front. I’m trying to keep it all in perspective and think about the great things going on for women in mathematics. However, I’m really troubled. Partially because it seems that so much gender bias is unconscious. I believe that, for the most part, people in mathematics want to treat each other fairly. While trying to be fair, we can still make a whole series of tiny choices and judgments that add up to discrimination. We can’t see our own patterns. We (yes, we) don’t even know we are discriminating. Of course women can be biased, too. For example, one of the student teaching evaluations studies discussed in the Inside Higher Ed article (also linked in Sara’s blog) found that female students in the study showed significant bias against female instructors, while male students in that study showed no significant preference. In a 2012 study, both male and female STEM faculty members rated a job applicant named John higher than an identically qualified applicant named Jennifer.

So I’m swimming in questions. Do women in math get less credit for collaborative work with men? Are there bummer biases of all kinds lurking in academic and mathematical life? Do I unthinkingly underestimate my female colleagues’ contributions because of their gender? Do I unknowingly assess others unfairly? My answer on all counts is wow, I hope not. Ugh. But probably the truth is, at least sometimes, yes.

I want to talk more about these issues in the profession. I am dying for someone to do a study similar to Sarsons’, only in Math, and to design other studies to uncover potential biases that we might otherwise be blind to. I also think that those of us who teach should to talk to our students about bias, especially as it relates to student teaching evaluations. If, indeed, most people want to treat others fairly, we need to have these conversations so that is even possible.

This entry was posted in bias, tenure, Uncategorized, women in math, workshops. Bookmark the permalink.

10 Responses to Collaboration, Bias, and Tenure

  1. Matt says:

    I’d be more skeptical of studies like these. The NYTimes has a narrative this fits nicely into. That researcher probably went into the study intent on finding bias. It could be the result of the Texas Sharpshooter Fallacy. It could be confounding variables. It could be that the correlation is there and real, but the causal agent isn’t the apparent one. The study doesn’t really pass the “smell test.”

    I’m not saying it doesn’t happen, but it is worthwhile to be skeptical of a single study like this. For example, you link to a study that finds STEM faculty prefer males all else being equal. But the exact same journal found the opposite when the same experiment was run in 2015:

    • bmalmskog says:

      Hi, Matt. I read that other study! I found it interesting, too, but it is about a different experiment which addresses a slightly different issue. In the 2012 study, faculty were hiring lab managers. In the 2015 study, faculty were hiring tenure track faculty. I think that the 2015 study reflects the fact that many STEM departments are consciously trying to hire women because their current faculty is imbalanced. The pool of applicants is also majority male (less than 1/3 of math PhDs went to females in 2013). The desire to balance the department and relative scarcity of women on the market probably helps women to rank higher in an experiment like that. I do not think this is the same issue as the 2012 study or Sarsons’ study.

      Is there something about Sarsons’ study that you find suspect? I’m interested to hear. I did a little looking around for criticism before I posted this, maybe the most interesting thing I found was this: It discusses other ways this bias could show up, besides people intentionally ignoring papers that women coauthored with men. Of course, there could be other mechanisms here! But I am still shaken by the effect observed in the study. It paints a biased field, one way or another.

      One thing that I would like to know is how the data has changed over the last 40 years. What is the trend? I would really like to think that the field has become more fair over time.

      I agree that any single study can only mean so much. A finding like this means a question needs further study. Sure. But my post came in response to a lot of different studies that I have found upsetting. I have had such great experiences and opportunities in my mathematical life. However, I think that data sometimes reveals bias that we as individuals, living only our own experience, would never be able to notice. We need to notice these things if we are going to change them.

      • Matt says:

        That 2015 study is one of the reasons this tenure study “feels” weird. The one essentially says we find reasons to hire women in STEM to create more balanced departments (I’m inclined to believe this based on watching the hiring process at one university right now). The other says people are essentially looking for excuses to not give women tenure. It’s hard to reconcile these results.

        The other thing that feels weird to me is to put so much emphasis on a single factor. Tenure review is a huge process that takes into account massive amounts of information. My question is: how many variables did they look at before finding this correlation?

        Just to play devil’s advocate for a second, the whole question could be turned around. It seems plausible that people who are generally weaker at research would seek out joint work to get better/more results and papers. That would not only be a plausible explanation of the women data, but it means that upward sloping line on the fraction solo-authored vs tenure graph is what we’d expect, not an indication of bias.

        The question would then be: why don’t we see this trend with men? Maybe men have an inflated sense of how good they are and so they don’t try to compensate in this way. Who knows. There’s millions of possible explanations. The point is that there seems to be an assumption that the thing needing explanation is with the data from women rather than the data from men. I haven’t read the paper, so maybe this gets explained.

        Anyway. I agree that data can reveal biases we don’t know we have. But I also think it’s easy to impose biases on data by jumping on the first explanation that fits a narrative when something deeper is going on or maybe nothing is going on.

        I’d like to see an experiment for further confirmation, and also for the paper to be published (as far as I can tell this was someone’s dissertation that hasn’t been peer reviewed yet).

  2. Sara Malec says:

    This is a great, thoughtful post. To add yet another layer, this article also crossed my path this week: “Even with hard evidence of gender bias in STEM fields, men still don’t believe it’s real.”

  3. dk says:

    Although I fully agree that there’s a big gender bias in many aspects of the math career, it seems to me that this is one of the few aspects where there’s no gender bias. And if there were any, it is just a minor problem compared for instance, to the widespread biases related to ethnicity or institutional affiliation of the authors. I often hear bashing towards people who consistently publish with multiple coauthors, but this is not related to gender but rather to a cultural aspect of the way in which mathematical research has been historically developed. Anyway, writing from time to time a paper on your own does not hurt your cv, it is a good proof of independent mathematical thinking and skills and, above all, shuts some cynical mouths.

    • bmalmskog says:

      Thanks for your comment and perspective on this. Yes, I think that the biases you point out are real issues. It seems to me that collaboration is unfairly viewed as suspect and undervalued in the profession. Sure, working alone is great, and that’s the way that some people work best. But why should it be considered better than working with others? What makes ideas better if they are developed by one person? Regardless of my perspective on this, I do realize that many people see solo-authored papers as the gold standard, and I am considering how I can find the extra time and the right project to work alone.

  4. cdsinclair says:

    I have witnessed some subtle and some not-so-subtle instances of bias against women in tenure and promotion cases at my university. One situation, relevant here, was the essential disqualification of a WIN paper/conference proceeding from consideration as a serious item on one of my colleague’s vita during a pre-tenure review. Somehow, some of my colleagues were/are under the impression that this could not have been a serious contribution to the literature, even though none of them are number theorists (and almost surely none of them tried to actually look at the content of the paper).

    Sadly, my department is almost all men (2 out of 35 tenure lines are women, and both of these are assistant professors), and any suggestions of impropriety/bias against women are met with an immediate circling of the wagons and horrified gasps of “how dare you …”

    My feeling is that there is an inherent tension between competition and cooperation, and a statistical separation between men and women as to the value of each of these in advancing the field. (Of course, I have no data to support this assertion). Assuming for the moment, this does play a role in the relative lack of women in mathematics, we may be able to move towards parity by ensuring that metrics of cooperation (hosting conferences, seminars, committee work) are evaluated equally and along side metrics of competition (paper counts, publication in pet journals, grants) during promotion and tenure decisions.

    Thanks for writing about this. I think it is important that we, as a field, keep talking about it.

    • bmalmskog says:

      Thanks for your comment–even though I am really disappointed to hear that the WIN paper was not counted by your colleagues. Depressing. Did they not realize that the proceedings volumes are peer-reviewed? Out of curiosity, have you had the chance to see what the same people thought about a similar mixed-gender or male-authored collaborative paper?

  5. Jen says:

    FWIW, I got a “it’s not clear that she can do independent research” comment from my tenure committee because much of my work is collaborative. I’m suspicious that this was exacerbated by the fact that my primary collaborator is male and I am female.

Comments are closed.