Friday, October 17, 2014

Slow years and job predictions: a graduate seminar in advanced straw-clutching

Today is the sixth week of the 2014-15 job market in German Studies. Some new jobs have appeared. Based on the new data, should I revise the number of jobs to be expected? Maybe.

To review: My previous estimate of the number of TT jobs to appear this year was 16, based on 9 TT jobs on opening day * 1.75. The basis for this was in history. The assumption is that a similar number of departments have their acts together enough to get a job ad into the opening JIL, and that's generally true. Since 2003, an average of 57% of all TT jobs have appeared on opening day, so 9 / .57 yields 15.8 jobs. Grim.

But wait! Now we have more data to go on. As of today, we're at 17 TT jobs, already ahead of the projection. Are more jobs coming? Since 2005, an average of 81% of all TT jobs have been advertised by week 6, so 17 / .81 yields 21 jobs. That's four more jobs, but still grim.

But wait! Maybe it's a slow year and lots more jobs are on the way!

No. No charge of the MLA job list stampeding in over the horizon is going to save you.

It's possible, however, that there has been an increase in the number of universities and departments whose  budget situation is so screwed up that they can't get a job posted by Halloween. And in fact there's some evidence for this. Between 2005 and 2008 - "the good old days, except we thought 2008 was just a down year and not the new dismal" - an average of 58% of TT jobs were posted on the first day, and 87% by week 6. Since then, it's been 51% on the first day, and 77% by week 6. If we use the more more recent averages from the dismal years since 2009 - the years most like 2014, in other words - we would project 18-22 jobs (using the first day/week 6 averages, respectively).

Which one is more accurate? Since 2009, the first-day projection as proved to be more accurate twice (in 2009 and 2011), while the week 6 projection has proved to be more accurate twice (in 2010 and 2013); they were tied (and dead-on accurate) in 2012. Taking the more recent years as a guide would suggest that the week 6 projection might be more accurate for this year: 22 jobs. Another way to look at it is that the lower projection was more accurate in 2005-2008, while the higher project proved more accurate in 2009-2013, again suggesting that the higher projection will be more accurate this year.

So I'm revising the projection upward by 6 jobs from 16 to 22. There might just be 5 more TT jobs out there waiting to be advertised. There's historical precedent for this: Since 2009, an average of 6.2 TT jobs have appeared on the JIL after week 6. The rule of thumb for next year might even become the number of TT jobs on opening day times two.

In other news, the number of non-TT jobs now advertised (11) is tied with 2005 and 2007 for the highest at this point in the last decade. Yay?