Whose problem is it?

Small Pond Science reminded us all some years ago that “disadvantaged students come from disadvantaged universities” and recently contended that NSF graduate fellowships are part of the problem of underrepresentation of minorities in sciences. The post (which is worth reading, along with the comments) continued on to excoriate NSF: “If the Graduate Research Fellowships aren’t a genuine part of the solution, than where the heck can we expect a solution to come from within the NSF?”

OK, so this is where GG got grumpy.

Now don’t misinterpret: there is a huge problem in getting minorities into earth science (GG cannot speak for other sciences) and it makes sense for there to be efforts to get minorities and disadvantaged students (these are not always the same people) into the field.  Where things start to go off the rails is in assigning this task to NSF, the National Science Foundation.  Why? Because the main goal of NSF, the main goal of the scientists who help by participating in the review process, is to get the very best science funded.  And trying to expand the pool of potential scientists probably involves investing in a lot of not-so-great science and not-so-strong universities, investments people working at the highest levels of the field may be loathe to make.  Demanding, as the Small Pond Science post does, that NSF take the lead on this asks a community to try to balance a duality of interests, one side of which is relatively foreign to many in that community.  Maybe tasks like this should be shifted to the Department of Education at the federal level, an agency with 10 times the budget of NSF. Kicking NSF in the shins because your undergrads didn’t get grad fellowships may not be the most useful response.

It isn’t even clear to GG that satisfying Small Pond Science’s desire for NSF to pump more money towards 4-year college students would succeed.  Like many walks of life, children of scientists are more likely to consider science as a career (next time you watch a professional sports match, see how many times a player’s parents come up in the commentary; sports stars, too, tend to follow in a parent’s footsteps), perpetuating economic and racial inequities in the field.  Science is not, by and large, a particularly remunerative calling; many disadvantaged but capable students will find fields with higher salaries and a shorter path to financial security far more attractive, while students from financially stable backgrounds are more apt to risk years of relatively low income (grad school and postdoc purgatory) to pursue a scientific career. After all, spending six years getting a PhD doesn’t guarantee anything other than six years not earning a living wage. And there is the issue of “do I see myself in that community”? Earth science in the US remains depressingly lacking in many minorities; jump starting minority participation will probably require a set of students who don’t care as much about what the community looks like at present. So it seems unlikely that the percentage of students getting grad fellowships will be the same for graduates of Harvard and CSU Dominguez Hills anytime soon.

So how might we solve this problem?  As Terry McGlynn (the author of the Small Pond posts and a CU graduate) notes, “If we want to diversify STEM, we’ve got to cultivate a broader foundation. If we want to diversity STEM, we can’t expect that the next generation of graduate students will come from the same elite undergraduate institutions.” Here’s the problem: you won’t solve this by trying to make every undergrad institution into a top-flight research school.  There simply isn’t the cash. That’s the path many of the CalState schools have tried to blaze, but between the limited dollars at NSF or NASA for research and the different demands on the faculty at such schools, it is hard to get much more than a watered down mess across the board.  (That said, some individuals in earth science who have simply required very little money have both managed to conduct capable research careers and foster students who have gone on to success at the bigger name places; these folks should be celebrated. But these are unlikely to be the main solution). Terry works far closer to the real issues than GG, but it sounds like a lot of the problems he sees are capable undergraduates not being steered into graduate programs.  If this is truly the problem, then some of the weight might be taken by partnerships of research universities pairing themselves with 4 year schools to provide mentoring and guidance.  Students interesting in earth science, for instance, can come out of a 4 year school with no idea of the breadth of the field or the different specializations out there simply because there is nobody at their school who knows any of that.  They might apply to schools that simply are terrible matches because, well, they simply don’t know any better. Such students–who might be very successful in the right places–might back away from a graduate education simply because they were knocking on the wrong doors. Similarly, 4 year schools might be unaware of the qualities graduate programs look for in potential students, and some kind of more effective pairing might help to inform them so they can better advise their students.

A problem GG has seen from the other side are disadvantaged students pushed towards graduate school by some well-meaning programs when those students are simply unprepared to succeed at the next level.  This is poison in multiple forms: we certainly don’t want to admit students only to have them fail, and failure at this level can be more devastating than not trying at all. Furthermore, such failures discourage faculty at research schools from risking the admission of similar students. There are two possible paths forward.  One is to have at the graduate level the resources necessary to bring such students up to the academic level necessary to complete a degree; the other is to better insure that students depart their undergraduate school with the skills and preparation necessary to succeed.  The first really requires the acquisition of skills and capability at a level where the expectation would be that you’d see some such students frequently enough to justify the cost and specialization.  There are such programs (GG’s is not one); it would probably benefit many 4 year schools serving underprivileged students to identify such programs as likely targets for their graduates. The second puts a greater burden on poorly financed undergraduate schools which might already have struggled to overcome educational deficiencies for students when they entered college.

Would changing NSF graduate fellowships solve all this?  No.  If anything, the statistics in the Small Pond post argue that the grad fellowships are a symptom more than a cause.