Gender and language in letters of recommendation

I don’t have time to write about this properly, since I’m going up to Seattle for the Cascade Topology Seminar in a few hours, and then going to Japan for two weeks, but this seemed too important to not point out: new research suggests that people write recommendation letters for women and for men in different ways, which act to the detriment of women.

Essentially, letters for women tend to play up their “nice” side and those for men their “active” side. Interestingly, the effect didn’t seem to depend on the gender of the recommender, and men who received “nice” recommendation letters rather than “active” ones also had more trouble getting hired, controlling for other factors.

Just a reminder that we all have to be really self-aware about how we write recommendation letters.

5 thoughts on “Gender and language in letters of recommendation

  1. The study makes no attempt to block for how supportive or ambitious the recommendees are. It’s anyone’s guess whether this is a letter-writing phenomenon or a working-in-academia phenomenon.

  2. For me, the takeaway is the last paragraph, which contains a really good idea. “Hebl urged those seeking letters of recommendation to not be afraid of talking about the issue with their letter writers. “Give them a copy of the research,” she said.” Well, I wouldn’t go as far as actually printing out copies of papers on this issue— but raising the topic in conversation is a very good idea.

    More generally— letters of rec are improved by _any_ amount of direct communication between the recommender and recommendee. For example if you are applying only to research oriented jobs, make sure your recommenders actually know that. It’s so obvious that it shouldn’t really qualify as a “tip”, but it seems to me that many people who aren’t described as “assertive” or “aggressive” or “ambitious” might also be people who avoid having direct and recommendation-specific conversations with their recommenders. (At least: I am such a person, and I personally avoid such conversations as much as humanly possible.)

  3. Not sure if “active” is the right word to use here; the study used “agentive”. As I’m sure you also know from personal experience, being supportive and friendly in an academic environment requires a lot of action! It seems to me part of the problem is that those types of actions aren’t being recognized as valuable and important.

  4. The study reported here may be new, but I’ve heard anecdotal evidence of this phenomenon for years.

    Also, I’ll add to Ben’s reminder that we all have to be really self-aware about how we read recommendation letters.

Comments are closed.