Friday, March 14, 2014

How the job market in German really works. Part four: Who gets hired?

(This is lightly updated from what I posted at pan kisses kafka.)

Let’s pretend that you are an American college student and, despite everything that people are telling you, you are determined to get a Ph.D. in German because you want to be a college professor. What grad programs should you apply to? Obviously—you think—you should apply to programs with a good track record of placing their graduates into tenure-track jobs.

Rule #1: Treat everything that departments say about themselves as delusional or deceptive until proven otherwise. Placement rates are ratios where both the numerator and denominator get twisted to the department’s benefit because it is a matter of survival for them.

First question: Do you have a trust fund or savings in the seven figures? Do you already have a guaranteed job waiting for you when you graduate? If so, it hardly matters what the placement rate is. But if you don’t, then you will have the academic job market waiting for you at the end of your grad program. You will have to watch the job lists, identify jobs you qualify for, and apply to them as one of dozens or hundreds of applicants.

Rule #2: What you want to know is departmental placement rates for nationally-advertised tenure-track jobs at the assistant professor level. It’s not enough that a department can point to “academic placements” (*cough* adjunct *cough*) or to graduates who are doing interesting things, or even to graduates who are in tenure-track positions.

Here’s the problem. People who earn Ph.D.s in the humanities are typically not found dead of starvation in the streets of major urban areas. We’re smart, we’re driven to finish what we start, we know how to research and how to impose order on vast quantities of information. We end up doing interesting things with our lives, often in the vicinity of universities, since that’s where we’re spending nearly the first decade of our professional lives. If a university has a qualified and competent person with a Ph.D. hanging around, it’s not too surprising that this person eventually ends up teaching classes in some capacity. That fact tells us little or nothing about the quality of the student’s grad program.

Lately, some people have suggested that doctoral programs should take some modest steps in order to keep track of what happens to their Ph.D.s after graduation. It’s a good idea, and these suggestions are made with the best of intentions, even if they’re coming about 50 years too late. They are, unfortunately, looking in the wrong place as far as you are concerned. You can’t just count up how many of a program’s graduates end up as professors—otherwise, the best qualification you could get in grad school is marrying a professor of engineering or accountancy who can swing a spousal hire for you. Instead, there is just one thing you should be looking at:

What percentage of a program’s graduates are hired for tenure-track jobs through competitive searches?

For the denominator of our placement ratio, we need to know the number of Ph.D. graduates of each program. The annual “Personalia” feature in Monatshefte is a decent starting point. They provide the number of Ph.D.s granted each academic year, and a somewhat less complete list of dissertations completed in each department. It’s the best source we currently have, unfortunately, now that David Benseler’s annual “Doctoral Degrees Granted in Foreign Languages in the United States” no longer appears in the Modern Language Journal. For the figures here, I’ve supplemented the number of Ph.D.s awarded per department with reports that I find in other places, particularly in the Proquest database of dissertations, and I’ve carefully checked any discrepancies.

For the numerator of our placement ratio, we should start, in effect, with the jobs wiki. For the tenure-track jobs that are nationally advertised each year (in the MLA/ADFL job list, in the CHE, on, etc.), who gets hired? After several years of tracking Ph.D.s granted and TT hires, we can then count up how many have come from each program. I’ve done that based on the best information I could find about new hires from the university websites of departments that have advertised positions in German. (We can’t rely on “Personalia” for hiring data because the coverage is too incomplete, and new appointments don’t show up with the regularity that you would expect in departmental listings.)

No program has 100% placement by this measure, and that’s not a bad thing. It’s great if Anna, who’s been the German instructor at a community college for years, finishes a Ph.D. and continues teaching at the same school. It’s great if Bettina signs on as the Assistentin of Prof. Dr. Dr. Großperücke back in Paderborn. It’s great if Max is hired into a tenure-track position because his wife made that a precondition of her accepting the job as dean of the business school. But if you’re looking at grad programs and don’t already have a job, or a spouse with $5 million in NSF funding, what you really want to know is which programs do the best at placing their graduates in the TT jobs that are advertised. Rather than tracking Ph.D. outcomes, we need to be matching Ph.D.s to hires in advertised job searches.

We have hiring data from 2006 onwards, but before we answer the $64,000 (in student debt) question, we need to take a quick look at when people get hired into TT jobs. For 173 hires from 2006-7 to 2012-13 where we’re confident that we’re looking at a first TT hire, the largest group (38%) appear to be hired ABD shortly before defending. That means, for example, that a job was advertised in the fall of 2006 (or perhaps in early 2007), while the dissertation was deposited some time in 2007. Smaller cohorts appear to be hired in the same year they defend (30%), one year later (17%), or two years later (8%). Before or after that, the numbers drop precipitously to a few percent or less. (Because of the different ways that degree dates are recorded, however, we have to treat the precise figures with some caution.) What’s important is that ABDs don’t actually seem to be at a disadvantage in applying for TT jobs, and most crucially for us here: we have to start looking at hiring one year earlier than the degree date, or else we’ll miss the biggest segment of who got hired. We have hiring data starting with the 2006-7 academic year, so we’ll start with the calendar year 2007 Ph.D.s.

 Figure 1: Average time from degree to first TT hire (difference between job cycle year and Ph.D. completion year)

Below, in one table, I’ve ranked graduate programs in German by placement rates based on the number of Ph.D.s produced from 2007 to the end of the 2012-13 academic year, and the number of placements into TT jobs advertised between 2006-7 and 2012-13. (I’m only counting jobs in the discipline of German, rather than multi-language positions or others not directly within German Studies. You can get the complete table as comma-delimited text here, along with the Ph.D.s and TT placements in my data set.)

But wait, you ask. Didn’t everything change in 2008? Maybe there was a “flight to quality” so that only Ivy League grads are getting jobs now. I’ve added additional columns that are restricted to Ph.D.s granted from 2009-13.

But wait again—what about the really cool jobs in R1, Ph.D.-granting departments that grad students who dream of revolutionizing their field lust after? The kind where you teach specialized grad seminars and go on sabbaticals and get invited to deliver prestigious lectures and everything else that your advisor does? I’ve added one more column for TT placements into doctoral departments.

I’ve eliminated Canadian programs (Alberta, British Columbia, McGill, Montreal, Queen’s, Toronto, Waterloo; 32 total Ph.D.s), as I find very few Canadian Ph.D.s hired for nationally-advertised jobs, and I’m concerned that I may not understand Canadian hiring enough to say anything useful about it. I’ve also moved sixteen small (1-6 Ph.D.s) or deactivated programs to a separate list, leaving us with 32 Ph.D. programs in German.

First, let’s check the Big Non-embarrassing Four. Princeton seems to be doing reasonably well (although they get an asterisk since one of those TT hires was their hiring one of their own grads), with three large public universities bunched together after that.

Next, in places 5-11, we have the one-in-three chance club.

Rounding out the top half, we have seven programs in the one-in-four club.

Coming in at places 18-27, there’s the one-in-five club:

And, finally, everyone else.

Here’s the list of small programs with less than seven Ph.D.s. Note that Florida and Duke have some decent placement rates, but not enough Ph.D.s for a comparison with the other programs. Some of these programs are being phased out.

The overall average TT placement rate is 20%, dropping to 15% if we only look at the years since 2009. (The most recent Ph.D.s will get another year or two on the market, so that number may creep up a bit.)
Note that highly prestigious universities can be found in every tier, including those with mediocre and truly abysmal placement rates. Graduates of some brand-name grad programs in popular coastal cities seem to have no interest in moving to boring flyover towns just for the sake of teaching 4-4. And the flight to Ivies after 2008? It didn’t happen. The only programs with decent and rising placement rates were Texas, Penn State, and the University of Washington.

Again, this doesn’t mean that the Ph.D.s who didn’t land TT jobs aren’t doing useful and interesting things with their Ph.D.s. Many of them are. They just didn’t compete successfully for the small pool of TT jobs that they spent 5-10 years training for.

But to really see how screwed up our profession is, you have to look at the doctoral placements. From 2006-12, there have been 47 TT hires by doctoral-granting programs (including five from programs not considered here), of which over a third of the total, 17, came from just three programs: Princeton (9), UC Berkeley (4), and Cornell (4). (Do you wish you had known that when you were applying to grad school? I certainly do.) Of those schools, only one, Princeton, has a non-embarrassing overall placement rate (and even Princeton’s has looked mortal since 2009). And note how the doctoral placements make up nearly all the TT placements for those schools: Maximizing your chances of following your R1 dreams by picking one of those programs comes at the cost of having terrible chances of landing any other kind of academic job.

* * *

Dear DGS’s and Department Heads: If you think my figures are wrong, please correct them. Do you think I’ve miscounted your graduates? Please provide an updated list of names and dates of Ph.D.s granted since January 2007. Have I missed your placements into advertised tenure-track jobs? Please provide the names and the date and place that the TT jobs were advertised. Above all, please consider the absurdity of some guy on the Internet collecting this information, instead of the GSA, the AATG, or some MLA committee. You’re in a position to change how little the discipline of German Studies understands its own job market, so please do something about it. The undergrads who are applying to your programs, and the grad students who are in your programs now, deserve better.


  1. Intriguing! I'd be interested in learning what proportion of advertised TT jobs go to German Studies PhDs as opposed to PhDs from other departments, if that's something you have on hand.

  2. It's not a huge percentage. I'll take a look at it and put together a post.