What kind of letters of recommendation are written for women?

There is a new study:

In the scholars' analysis of the words that appeared in the letters of recommendation, they found clear patterns of word use for women's and men's letters. Women were more likely to be described with words such as those cited above, as well as "nurturing," "kind," "agreeable" and "warm." Men, in contrast, were much more likely to be described in words classified as "agentive" — words such as "assertive," "confident," "aggressive," "ambitious," "independent" and "daring."

What the analysis showed is that letter writers didn't need to use words like "feminine" to create female stereotypes — and that they did so, time and again, with women who had the same intellectual achievements as their male counterparts.

…The research is based on a content analysis of 624 letters of recommendation submitted on behalf of 194 applicants for eight junior faculty positions at an unidentified research university.


assuming the answer. Endogeneity.

Oh, the horror! Despite best efforts, men and women are still, STILL! perceived differently. Let's all wear full-length veils and legislate a "don't ask, don't tell" policy on sex in public life. And talk in sign language, after all what right do we have to talk with voice when there are deaf people?

I've written dozens of letters of recommendation and never used any of those words.

In fact, since my letters have always been successful, I usually don't change the wording much at all. Why change what works? Good students generally have the same desirable characteristics which don't vary by gender.

To the extent I customize a letter, I do so based on the exact wording of the job announcement or admissions criteria. So, if there's any gender differences in my LoRs, they come entirely from the positions for which they apply, not my assessment.

Perhaps differences in LoRs depend on the position applied for, not personal qualities. Women and men still do segregate themselves into certain types of jobs and certain fields in academia.

I might have missed it, but did they control for the gender of the writer? Women might be more inclined to mention or praise one's emotive nature.

A post containing a link and a quote doesn't provoke thoughtful comments, Math.
> Well, it's unanimous: 100% of Marginal Revolution agree that studies investigating gender bias should be dismissed out of hand.
Not subtle enough.

Just because these comments are dismissive, it doesn't mean they aren't thoughtful. Few outside the humanities would do this type of research and then upon noting the correlation conclude that stereotypes are created. As any good economics reader knows as gospel: correlation does not equal causation. Evolutionary biology and sexual selection suggests that over time female and male populations become more different over time. Until we start reproducing asexually, why should we expect men and women to become more alike?

Which is more biased:

(1) identifying more women than men as caring; or

(2) suggesting that caring is a negative character trait?

"Tyler Cowen is one of the most thoughtful bloggers I know of his. His commenters appear otherwise. Why?"

1. Your exposure to blogs may be fairly limited if you believe Tyler is one of the "most thoughtful" bloggers out there. (No disrespect to Tyler.)

2. Your exposure to blogs may be fairly limited if you believe MR's commenters are not. As crazy as some of the regulars make me, this place is head and shoulders above what's found around most of the web.

Sounds as though you've got a newly gored ox out back.

"Well, it's unanimous: 100% of Marginal Revolution agree that studies investigating gender bias should be dismissed out of hand."

Huh? Go review the response to any study on any subject. This one probably gets off easy. Gender bias studies probably deserve the same treatment as most soft science studies.

Lastly, as a researcher and aspiring hiring manager of researchers, I'm planning to favor people who are 'caring.' I've been screwed over by plenty of these 'daring' go-getters. I'm not interested in zero-sum winners. However, if I do ever get the chance, perhaps the incentives will demand that I prefer the go-getters. I think this is partly a problem of academia where one has to "move the needle" in a very short period of time. And in a system I consider broken, I'm not sure gender bias is the big problem. But, I'm a dude.

It looks like someone HAS hijacked the thread, eliciting at least two (er, now three) comments.

In this particular post, Tyler expressed no opinion, thoughtful or otherwise. Math is Hard seems to assume that, by merely posting the article, he has thoughtfully endorsed it.

Whether he agrees with it or not, Tyler has always invited thought, debate, and expression. There are tens of thousands of academics who would accept the conclusions of this paper without the slightest questioning of its methodology and without providing opportunity for dissent.

Hop on over to Calculated Risk, which is one of the finest blogs for Economics, Finance, and Real estate, and its comment board is filled with blithering idiots, spouting conspiratorial nonsense and engaging in snarky chit chat. Most of the posts on Krugman's blog come from sycophants.

Apparently more is hard for Math is Hard than math. Is that you, Vivian Schiller?

While the tendency towards "girly" words in recommendations for women is interesting, the more important point to take from this study that the lady-like letters of recommendation also mean the candidate is less likely to be hired.

It's not the words that are biased, for both genders they were all positive in the analysis. It's the reader's perceptions of the gender associated with words, and the apparent preference to hire the masculine sounding positive traits. It's an interesting thing that occurs when markets and society push women to adopt more masculine traits to be more competitive (say aggressive leadership), and sometimes men too are nudged to play the more feminine cards. It appears in this study, women would be more pushed towards the masculine spectrum, which is neither good nor bad (all the letters were positive), just interesting.

It would be interesting to see how negative work performance reviews vary by gender (and other trait subpopulations).

Comments for this post are closed