Sunday, June 21, 2015

How Long are Postdoctoral Fellowships? - Part 2

Following some interesting interactions with readers on Twitter and in the comments, it seems my initial post could have improved with 1) more data, and 2) break-down by gender. So, I spent some time this weekend digging through my last two Bumper Cars posts (2014-2015, 2015-2016) in order to provide a clearer picture.

How did we get here? Read the original post.

2015 - Random Trillium from a recent weekend hike

For starters, raw numbers - as of this writing, there are 194 confirmed new hires between the two lists. Of these, I was able to track down information about postdoctoral appointments for 132 (68%). Unfortunately, there's no consistent format for how candidates track their experience; I found myself cobbling it together from LinkedIn, university websites, ACS member profiles, and digital thesis repositories.

To keep myself honest, I'm pasting my assumptions below* this post, as before.

Now, to the numbers: First, the aggregate statistics - of the 132, here's the new mean/median/mode:

MEAN:3.49 years
MODE:3 years
MEDIAN3 years
MIN:             0 year (no postdoc!)
MAX:            9 years
(n = 132)

So, roughly in line with what I had before. But what about the gender gap? Do men spend significantly less time as postdoctoral scholars?

First, it helps to clarify what the real split looks like: Of 132 candidates, 39 (29.5%) are female. This may be an admittedly small data set, but I see only a slight difference** in overall time: 
3.47 years for men (n = 93), 3.54 years for women (n = 39).

Edit (6/21): A good spot to insert a quote from a 2013 Beth Halford piece in C&EN:
"For one, although there are no hard numbers to point to, some say people are spending more time in postdoctoral positions. In chemistry, one to two years used to be the norm, but that time frame may be creeping up. Some chemists tell C&EN that they are spending five or more years doing postdoctoral studies."
Seems to be the case, at least to me.

Of course, the best way to make this data set relevant is to send in even more new names! Once I can figure out a good mechanism to capture pharma / gov't hires, I'll try to expand the analysis. Who knows? Maybe we'll get a real live database set up...

*These postdocs reflect faculty appointments; I'm clearly not counting those who went into government, pharmaceutical, industry, or left chemistry entirely. If someone has a good idea for how to capture that data, I'm all ears.

Counting time: If someone gave a graduation year - "Ph.D. 2009"-  I assumed a postdoctoral stint until their faculty start date. For example, 2015 start = 6 years a postdoc. If, however, they provided a range - "postdoc 2012-2014" - I assumed that they postdoc'd the difference of that time, or 2 years, despite the fact that, depending on start and end dates, that could reasonably be interpreted as any length of time between 13 months (Dec 2012-Jan 2014) and 36 months (Jan 2012-Dec 2014).

Of the 77 new faculty starting in 2015 or 2016 (as of June 2015), I was only able to find bio-sketch information for half. The following people from my list are represented in the above statistics: Li, Engle, Hyster, Matson, Menard, Personick, Thoi, Tsui, Wasa, Blakemore, Browne, Devery, Gahlmann, Kempa, Limmer, Nelson, Sing, Thompson, Bantz, Hubbard, Garcia-Bosch, Huo, Wei Li, Mirica, Rossini, Seiple, Wu, Anand, Boudreau, Genereux, Jiang, Sletten, Theberge, Fu, Ke, Conley, Raston.

**One additional complicating factor? It's tough to tell who's a postdoc anymore. This study only includes candidates who list their experience as "postdoc", "postdoctoral fellow", "research associate", or something of that ilk. Senior researcher? NSF Fellow? Visiting researcher? Lab assistant? Are these postdoctoral positions, or not? Tough to tell, so I excluded them. Thus, I may be artificially shortening certain candidates' timelines.

Additionally, certain candidates had "gaps" of 1-2 years in their experience, and I could find no information for what they did. Took time off? Worked somewhere that didn't pan out? Had a child? I simply don't know.


  1. The data on what people did in their "gaps" would fascinate me to no end. But we'd probably have to do a sacred dance to petition the deities of chemistry to give up that info.

  2. all of those things you've excluded are postdoctoral positions or equivalent

    1. Thanks for commenting; I think it's more nuanced than that. Certain roles allow junior academics to manage a lab group inside of a larger group, as a group leader or research asst prof. Are those postdocs? I came across one person whose description read "Senior Scientist", although he worked inside a large academic group as computational support.

      Maybe I'll do another set of stats that includes just *time difference* between Ph.D. and first academic appointment. This would capture gap years as well as alternative positions.

  3. Some may be senior postdoc positions, or very well paid / fellowship supported postdoc positions, or be some sort of senior collaboration postdoc, but effectively, if your not independent, and you don't have a permanent position, your splitting hairs

  4. I'm guessing that if you factored in the post-docs who left to go into industry, the mean time would be lowered.


  5. I would like to thank you for the efforts you have put in writing this site. I am hoping the same high-grade website post from you in the upcoming as well. In fact your creative writing skills has inspired me to get my own site now. Really the blogging is spreading its wings quickly. Your write up is a good example of it.
    5 Wire Resistive Touch Screen