Monday, January 10, 2011

Should chemistry graduate school rankings be based on employment?

Over the weekend, there was a rather wonderful article in the New York Times about the perils of trying to become a lawyer. While the main story was the difficulties of keeping up with student debt (and the associated consumption issues), the article also covered the problems with tracking whether graduate law school students were unemployed. To wit:
Even students with open eyes, though, will have a hard time sleuthing through the U.S. News rankings. They are based entirely on unaudited surveys conducted by each law school, using questions devised by the American Bar Association and the National Association for Law Placement. Given the stakes and given that the figures are not double-checked by an impartial body, each school faces exactly the sort of potential conflict of interest lawyers are trained to howl about.
The surveys themselves have a built-in bias. As many deans acknowledge, the results are skewed because graduates with high-paying jobs are more likely to respond than people earning $9 an hour at Radio Shack. (Those who don’t respond are basically invisible, aside from reducing the overall response rate of the survey.)
Certain definitions in the surveys seem open to abuse. A person is employed after nine months, for instance, if he or she is working on Feb. 15. This is the most competitive category — it counts for about one-seventh of the U.S. News ranking — and in the upper echelons, it’s not unusual to see claims of 99 percent and, in a handful of cases, 100 percent employment rates at nine months.
A number of law schools hire their own graduates, some in hourly temp jobs that, as it turns out, coincide with the magical date. Last year, for instance, Georgetown Law sent an e-mail to alums who were “still seeking employment.” It announced three newly created jobs in admissions, paying $20 an hour. The jobs just happened to start on Feb. 1 and lasted six weeks.
A spokeswoman for the school said that none of these grads were counted as “employed” as a result of these hourly jobs. [snip]
As absurd as the rankings might sound, deans ignore them at their peril, and those who guide their schools higher up the U.S. News chart are rewarded with greater alumni donations, better students and jobs at higher-profile schools.
It's been years since I've seen the US News ranking for chemistry departments; as I recall, they lined up with the conventional wisdom quite well (H-bomb on top, etc., etc.)* I don't think they relied particularly strongly on post-graduation employment. (Law school rankings, I believe, are far more reliant on those numbers.)

I suspect that if the question was "At year 3 after receiving your doctorate, how many of your graduates are employed by private employers?", the numbers would begin to look a lot worse. Certainly postdoctoral fellowships should not count as 'employment'; it's unknown if BLS counts them as "employed", but I know that ACS does not. (It doesn't count them as being 'unemployed' either, but breaks them out into their own separate category.)

At some point in the future (especially if things don't recover with the overall economy), these questions will have to be answered by graduate schools in chemistry.

*Anon011020110850a challenges me to update the rankings. For overall chemistry graduate schools, they are: tied for 1st: Berkeley, CalTech, MIT, 4th: Harvard, Stanford, 6th: UIUC, 7th: TSRI, Northwestern, UW-Madison, 10th: Columbia and Cornell.


  1. Is there any good way to make graduate programs start doing this? I don't see it happening until there are a couple of grad schools on board of their own accord. This just seems really implausible.

  2. Yes, people don't really volunteer to be graded on their most difficult problem areas.

  3. While I'm not preaching a new sermon here, I will nevertheless restate a case made by many (including myself) before.

    Universities are *not* job placement agencies. They are not designed as such and -- to be honest -- aren't equipped for it. Nor should they be. However, as professional schools continue to push this issue (come to law school for a job, not for the law), other units on campus will be pressured to do the same thing. It's short-sighted, thoughtless, and potentially sets students up to make less-than-wise decisions.

    Far more useful and relevant information for potential graduate students (in STEM, humanities, professional, and social science fields alike) would be:

    1. Completion Rate (Men, Women, etc.)
    2. Time to Degree
    3. Departmental Atmosphere
    4. Did your advisor provide you with helpful information and mentorship regarding *YOUR* career path of choice.
    5. Were you treated like a valued member of the department or simple a tuition-paying lab grunt?

    Some of this information is available already and much of it (or something similar) is available from the Survey of Earned Doctorates.

    Anyway.... job placement is a reflection of the job market rather than the quality of an education. To confuse these issues is a disservice to students, faculty, researchers, and your field of choice. Students ought to have a realistic source of information on their job prospects. But there are so many variables to consider on that point that assuming "how many graduates from this program are employed" is a relevant stand-in data point would prove, I think, superficial and amateurish.

  4. The NRC grad school rankings did take into account the number of (former) students in academic positions. Though that really leaves out most graduates, and I'm not sure if "academic positions" included postdocs or not. But with their crazy multiple variable analysis/subjective rankings, surely they could throw in some stats on employment.

  5. PS... It's a hugely important conversation to have. Thanks for bringing it up!

  6. Through my conversations with undergrads who are contemplating graduate school and also current graduate students, a _major_ topic is: "how will this time that I spend in a graduate program prepare me for the job market?" On the other hand, chemistry graduate programs blatantly advertise themselves on the basis of job opportunities. And those opportunities are currently in very short supply.

    And so, Mr/Ms/Dr jeconnery, I believe that your criteria are very naiive. Anyone in the organic chemistry field can tell you that the doctoral pedigree (location research director) is very important in landing a position in either academia or major chemical industry.

  7. Responding to jeconnery, I would like to say that while you're right that "Universities are not job placement agencies," on the other hand Universities do absolutely perform job placement functions. Many (if not most) universities have career placement services that (ostensibly) have some mission to help graduate students.

    Personally, I believe that there is some reasonable expectation that a Ph.D student should be able to get some career placement assistance at both the University and Departmental level. I believe this in large part because my particular University, relative to its size and prestige, does an absolutely abysmal job at this for Ph.D science students. I base this on peer institutions that (a) are hosting far more on-campus interviews and (b) have professors that actively help their students find jobs in the private sector by aggressively seeking interviews for them.

    So I think that it's absolutely reasonable to expect that some combination of my University, Department and Faculty will step up to the plate and help me with my job search. In today's market, where employers don't seem to be even reading resumes anymore, there doesn't seem to be any other way to gain an advantage.

  8. Aww...CJ, what happened to your pedagogical idealism? Just as there are starving artists, shouldn't starving chemists be allowed to toil in graduate school purely for the sake of understanding one aspect of the universe?

  9. @Chemjobber: I believe a correction is in order. UC Berkeley still occupies the overall top spot in the US News Chemistry Ranking, while Hardfart continues its stranglehold on the organic category. Unfortunately, the ongoing California budget crisis has left the Golden Bear defanged, declawed, and mangy.

  10. A8:41a: I'm not really sure that I had any to begin with, even though I sincerely think that I am an idealist at heart. I dunno -- was your comment rhetorical or did you have something specific in mind?

    A8:50a: Check update.

  11. I actually agree with jeconnery that schools' job placement rates should not be published, but only for the reason that their own numbers are not to be trusted.
    I graduated with a masters (in inorganic chemistry) and, not being able to find an industrial job, went back to a different university to re-train as a research associate in organic chemistry. Googling my name a few months later, I came upon an internal review from my old department that put me in the "employed" pile. My $19k/year entry-level job counted every bit as much to them as it would have had I been making $60 k in industry.

    In response to jeconnery, job placement is *largely* a reflection of the job market, but to say that it "does not at reflect on the quality of the education" is ridiculous. I would say that if universities use improved job prospects as a selling point for their program - and find me a chemistry department which does not, in some way, do this - then they should be prepared to be judged on that criterion as well.

  12. @Sharon- the National Research Council's new grad school rankings criteria on '# of students accepting academic positions' does indeed include postdocs. It's a crude measurement at best, but my guess is that most uni's wouldn't have had any better stats on employment to give them. The rankings also include a couple of jeconnery's criteria. You can argue until the cows come home as to whether they're the right criteria for ranking programs, of course.

  13. The rankings of American chemistry graduate schools by U.S. News appears to be based on college football and basketball records. Or perhaps the rankings are based on number of faculty members who have subscriptions to the magazine. Apparently none of the staff bothers to visit the chemistry departments. The editor might also be surprised that many PhD chemists are not interested in searching for the 3 or 4 available tenure track jobs available each year. Nor are they willing to work for $20,000/yr for the rest of their lives filling in for faculty on sabbatical. Industrial research is both financially and intellectually rewarding.

  14. We're trying to collect information on the # of PhD students accepting academic positions:


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20