minorities and academia: some further thoughts

When thinking about increasing the presence of under represented minorities in the professoriate, I think of the pipeline process model. Roughly speaking, a pipeline process suggests that something happens in multiple stages. The immediate consequence of the model is that if you want X to happen you have to make sure that all the stages that make X are working properly. In terms of faculty diversity, that means recruitment to graduate school, professional training, job placement, career development, and the tenure process.

A while ago I reviewed evidence from ASA reports showing that the pipeline is leaky. On the one hand, graduate programs seem to recruit a fair number of minority students. Then, once training is complete people seem to do well getting the jobs. Then, there is a massive drop in the pipeline as people go up for promotion.

Now that I’ve been on the job for a while, I think the following is happening: the core faculty of the PhD programs are not working with minority PhD students. They are admitting students, awarding degrees, and writing letters of recommendation, but they are not collaborating with students in ways that lead to publications and grants. In other words, most successful students work with faculty who “get them started” while their own research takes a little time to develop. My hypothesis is that if you looked at PhD minority students they are way less likely to co-author with faculty and that they are less likely to receive an offer of co-authorship. I’d also hypothesize that this gap is largest for top tier journal publications. This will be small or non-existent in areas focused on race and ethnicity. In other words, when faculty build teams to shoot for that ASR or AJS publication, the minority students come last for invitations, except in race & ethnicity areas. I didn’t think this is conscious, but this might be happening and explains the drastic leaking throughout the later stages of the pipeline.

Am I right? If you are a faculty member at a top 20 or 30 program in your field, the test is simple. Look at your list of co-authors for your big papers. Look at your list of minority students. Look at the overlap. Use the comments section.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz


Written by fabiorojas

April 30, 2014 at 12:01 am

Posted in academia, education, fabio

14 Responses

Subscribe to comments with RSS.

  1. “Am I right?” If you ignore self-selection at each “stage” of the pipeline, you might be. You know this, so you specify “except in race & ethnicity areas.”



    April 30, 2014 at 2:13 pm

  2. If the discipline & its departments are largely relying on “cooling out” to thin their ranks, can we really disentangle “true” self-selection from institutional practices?


    Cooling Out

    April 30, 2014 at 3:58 pm

  3. I don’t get it. If the attrition is mostly at the assistant/associate hurdle then why would it be caused by dynamics during grad school? I could see this if we assume that assistants are mostly just publishing work they started in grad school, but most departments (at least in sociology) try very hard in the tenure process to establish that the candidate has new work distinct from that started under the close mentorship of his/her advisor.

    Or did I misread your post?



    April 30, 2014 at 4:29 pm

  4. @Lisa: Self-selection only goes so far. Do you really think that minority students actively avoid all non-race/ethnicity topics? Do minority students turn down all requests for collaboration from non-race & ethnicity faculty?

    @Gabriel: I omitted some details in my post that might lead to confusion. Here is my model for the modal academic career in sociology and other journal oriented fields. During graduate school, senior faculty (or very active juniors) recruit students to help with data collection and articles. For the typical student, this co-authorship creates the CV for the job market and, to a lesser extent, the midterm review. Thus, grad school behavior spills out into your first job for most people.

    This is important because social science journal publishing is usually a multi-year project. So, if you have a midterm review at the beginning of year three, that means you have to submit a review packet after 24 months or so after showing up. Many articles simply won’t be accepted by that time, especially those from the dissertation. Thus, initial hiring and midterm renewal depends, usually, on having something done that was submitted while in graduate school, which is often from an adviser’s project.

    Of course, you correctly point out that hiring/promotion committees try to ascertain how much of the work is from the student. But, in practice, it is not too hard for people to come up with a semi-plausible story that explains that, well, yes, this article is co-authored but in fact the student did all the work. Sometimes it’s true, sometimes not. My point is that most of the time people tend to be generous toward younger scholars.

    Now, this story might simply be incorrect. But my observation of a lot R1 hires (including my own institution) is that they almost always have faculty co-authored articles and that these are the main publications for a couple of years. This ensures successful midterm reviews and eventually successful tenure reviews once they publish their own work.

    Getting back to Gabriel’s point. If you believe this model, then the co-author gap won’t have an immediate effect. The reason is that lots of faculty are hired with few pubs, or even none. So it only appears later when you get lukewarm midterm reviews, or your publication is slow or “wrong” (low ranked journals) because you simply never got the training from the co-authorship experience.

    There might be other forces at work. Lisa is not the only one to focus on self-selection. For example, if 100% of minority students do race &ethnicity you will get a suppressed count of minority faculty. But at the very least, co-authoring is a simple and obvious strategy that directly solve the issue.



    April 30, 2014 at 5:37 pm

  5. Gabriel: maybe review and publication times in sociology are so slow that the successful tenure candidates are submitting their post-PhD work while they are still in grad school.

    More seriously, I think most departments expect publications off the dissertation AND off post-dissertation work. If assistant professors of color are starting their TT jobs with fewer publications off their dissertation or pre-dissertation projects than other assistant professors, they will spend proportionately more of their pre-review years publishing this “old” work. Hard to catch up.

    (This doesn’t mean that everything else about the assistant professor experience is race-neutral.)



    April 30, 2014 at 6:25 pm

  6. I take this all as very compelling arguments that for the sake of diversity in the ranks of our associate professors we must expedite the journal review process. Every time you decide to leave a “you forgot to cite this” comment out of your review report, an angel gets its wings (and quite possibly a minority gets tenure).

    (If Fabio can say that his pet issue is an obvious “common ground” proposal, then surely I can say the same of mine).



    April 30, 2014 at 7:29 pm

  7. Gabriel, you get the common grounds seal of approval.



    April 30, 2014 at 7:31 pm

  8. There’s a problem with the leaky pipeline metaphor. It suggest that the people who leak are dripping into some puddle of waste. In reality, we don’t know where people who “leak” are going. Maybe they end up being unemployed. But maybe they end up become community college professors. Maybe they work for think tanks. Maybe they join the civilian arm of the government. Maybe they join the military. I wouldn’t classify any of this as “leaking.”

    In addition, the ASA relies on cross-sectional data instead of longitudinal data. One would expect the *percentage” of full time faculty to be mostly white because the composition of that group comes about from the lagged effect of educational trends in the 1960s, 1970s, 1980s, 1990s, and early 2000s. This is not true of post-docs, docs, and undergrads. If you look at full professors but exclude those who were hired before say 1995, you may find something else.

    The metaphor suggests that if we just keep on engaging in social engineering, we’ll arrive at some ideal state. Sometimes that’s true. But sometimes people just have autonomous desires that won’t submit to social engineering.

    P.S. The link to the ASAnet report in your other blog post is broken.


    Chris M

    April 30, 2014 at 9:05 pm

  9. Autonomous desires? That is, minorities with the PhDs do not want tenure at the same rate as white ones? Or departments sometimes have the autonomous desire to use junior faculty as tokens?


    Institutionalized racism?

    May 1, 2014 at 3:36 am

  10. Not that they don’t want tenure, but that they do want jobs doing other things. I was mainly referring to the leaky pipeline metaphor as a whole, not just the tenure-track section.


    Chris M

    May 1, 2014 at 4:03 am

  11. @Fabio: “Do you really think that minority students actively avoid all non-race/ethnicity topics?” Of course not. The point is that the restriction you impose – “except in race & ethnicity” – stacks the deck in favor of your argument. If we did the research – maybe it has already been done? – I’d expect to find that minority students are more likely to co-author with faculty. I would also expect the difference to be smaller outside of “race & ethnicity.” So, like I said, only by ignoring self-selection might you be “right.” Also, while the mention of self-selection has raised the predictable knee-jerk reaction in the comments, I was referring to the “over-representation” of minority students in areas like “race & ethnicity” as the outcome of self-selection, not about the assistant/associate hurdle.



    May 1, 2014 at 2:37 pm

  12. If this is right (and I’d be interested to see more information that would help to show whether it is), I think another corollary is that, if assistant professors of color are more likely to pick up the slack in mentoring students of color, that might slow down their progress as well (to the extent that it gives them a larger advising burden overall compared to other assistants).



    May 1, 2014 at 6:27 pm

  13. Did anyone actually take you up on the offer?



    May 3, 2014 at 6:52 am

  14. […] weeks ago, I suggested that some racial disparity in the professorial ranks might be due to low rates of co-authorship between the m …. The theory is that these collaborations provide a stream of publications that sustain people […]


Comments are closed.

%d bloggers like this: