53

As a researcher in computing science in Western Europe, I get depressed seeing the number of female students in our department. (Hint: near-zero.)

I recently heard that certain words in a PhD candidate ad seem to "scare" female applicants off, for example, "excellent", "commitment", etc. (I got my information from our anti-discrimination representative and a female researcher. I did not check the information myself. The latter claims having spoken to students herself. I agree that the ad is not the only problem with skewed gender distribution, but if this contributes to the problem then I definitely want to eliminate it.)

Is there a list of DOs and DONTs when writing ads to ensure that female students are encouraged to apply?

1
  • 2
    Comments are not for extended discussion; this conversation has been moved to chat.
    – ff524
    Commented Dec 20, 2017 at 19:03

6 Answers 6

66

It is believed that the use of "gendered" language in job ads can make ads appeal more to men or women [1]. See no reason that this should not also apply to ads for PhD students.

There are several websites that claim to test your job Ads for gender bias based on the word lists in [1].

E.g.

http://gender-decoder.katmatfield.com/

or

https://www.totaljobs.com/insidejob/gender-bias-decoder/

The other common advice is to avoid specifying requirements that are not really requirements, but actually nice-to-haves. Widely cited research [2] suggests that men are far more likely than women to apply for a job even if they don't meet all of the requirements. So for example, if you say that you require students to have taken a class in X, but when it actually comes to it you get no good applicants that have, but decide to take a student who is really good in all other ways, but does not have class X, then class X shouldn't be specified as a requirement, but a nice-to-have. Saying in your ad that applicants must have X is then biasing against women.

Mind you, I always follow all this advice (the gender checkers even rate my ads as slightly feminine bias), but I still get next to no female applicants.

[1] Gaucher, Friesen, and Kay, Journal of Personality and Social Psychology, July 2011, Vol 101(1)

[2] https://hbr.org/2014/08/why-women-dont-apply-for-jobs-unless-theyre-100-qualified

1
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – ff524
    Commented Dec 26, 2017 at 3:19
55

This is an indirect answer from my experience hiring software developers.

TL;DR Stop focusing on the ad itself and instead focus on where people see the ad, in particular get small groups of people to learn about the opportunity in an intimate setting.

As opposed to focussing on the language of the ad, I find it's more effective to focus on where the ad gets placed.

For example, I have posted several jobs on LinkedIn using their "easy apply" feature. Sure, it's easy to apply but applicants also generally have a picture and some personal information on their profile which can create a stigma for potential minority applicants owing to the bias of recruiters, which is unfortunately very real. The net result is that you get fewer quality minority applicants.

You may not be searching for applicants on LinkedIn but I suspect that you are in a male dominated field (like software) which itself can create the same stigma.

My suggestion is to change your recruiting tactics. During my last search for a developer I posted the ad on LinkedIn but I also reached out to local developer meet-ups and asked for referrals from current employees. The more informal and more personal recruiting effort resulted in qualified women applying to job at a rate of 2:1 compared to men despite no substantial change in the local candidate pool.

Did recruiting in this manner take more effort? Yes. Was it worth it to create a great, well balanced team? Definitely.

1
  • 3
    European academic recruitment happens in the weirdest of places, but thankfully at least currently, LinkedIn is not one of them. Commented Dec 24, 2017 at 15:03
28

It may seem against some ideologies out there, but as a husband, a father, and a colleague of numerous women in academia I can approximately tell you what women considering career in academia want to see in your ad.

How family friendly is your work environment? Any benefits for parents? Not just women, their husbands too. Vacations, breaks, holidays? Available housing close to campus? Hospitals / kindergartens / schools nearby? Public commute options? Low stress, friendly work environment? Stability of employment?

If you want a more accurate opinion, ask some women what they want from an academic institution to consider a career there.

Then write your ad describing what your institution has to offer women. Specific, succinct, and solid offers. No 'maybe', 'if then maybe', etc. Then you will attract top talent. Sure, many under-qualified will apply, but you can reject them based on their CV and pick the best.

Don't present requirements as 'requirements'. The people you really want will have requirements for you. Instead, ask applicants to please include in their resume 1), 2), 3), etc. without using the word 'requirement'.

One thing that won't work for sure is making a flashy ad like for selling a vacuum cleaner. With career decision of such critical importance, there is no chance fooling anyone into it. Especially smart, young women. Age 24-30 is a good time to make a family, give birth to a child, and put a down-payment on a house. And smart women do want it good.

In fact, many people disagree with such a pragmatic view, but I honestly don't understand why. It works. The question is not about letting women to participate or not. I am assuming there are equal rights. The question is about attracting talented women.

9
  • 15
    How are any of those desires specific to women? Commented Dec 21, 2017 at 0:55
  • 19
    @astronat Time is factor. A man has more risk tolerance because he can make a family after 30 and will not feel at a disadvantage. A woman will. In my experience, there are very very few women who put career above family. In fact, it is a theoretical few I read about and saw in movies. I am yet to meet one in person. Commented Dec 21, 2017 at 1:28
  • 15
    @ArthurTarasov I've met several women who put career first. Especially at the Phd application level, it doesn't play a role for many, yet, in practical terms. Still, statistically - from a personal experience pool - you are right that these are things that are more likely be of interest to woman because, a) there is a higher percentage of single female parents (who have full/main custody) and b) apparently more of them think about these things before actually getting a child. Once a man is a father, I think this also applies equally to them as to mothers. Commented Dec 21, 2017 at 1:54
  • 21
    You are talking about stability of employment and low-stress work environment when the ad is for a PhD position ... ?
    – Szabolcs
    Commented Dec 21, 2017 at 9:43
  • 8
    @Szabolcs there is a difference between guaranteed employment for four years and "we'll decide if we keep hiring you every year", for instance.
    – Davidmh
    Commented Dec 21, 2017 at 16:55
14

tl;dr- Apparently Carnegie Mellon University recently got an incoming first-year undergraduate class that was 48% female for Computer Science. They seem light on the details, but their press release stressed a focus on social activities in the curriculum.

Maybe try whatever Carnegie Mellon did: increased social focus?

In this question's chatroom, @ElizabethHenning had pointed out that Carnegie Mellon University has recently reported near-gender-parity in an incoming class of first-year undergraduate Computer Science majors:

Women make up more than 48 percent of incoming first-year undergraduates this fall in Carnegie Mellon University's top-ranked School of Computer Science (SCS), setting a new school benchmark for diversity.

[...]

Women and men are judged by the same standards for admission, and retention rates historically have been the same for both, he noted.

-"Women Are Almost Half of Carnegie Mellon's Incoming Computer Science Undergraduates" (2016-09-11)

This was a goal that they'd sought for a few decades now. How'd they do it?

Blum was instrumental in establishing Women @ SCS, a faculty/student organization that helps women make connections across the school, and in recognizing that it's the computer science culture — not the curriculum — that needs to change to accommodate women.

[...]

"Computer science at Carnegie Mellon has a very social atmosphere," Blelloch said. "Most of the classes encourage students to work in groups, developing skills they will need to function in teams in the workplace. It's one of the reasons employers like our graduates so much."

-"Women Are Almost Half of Carnegie Mellon's Incoming Computer Science Undergraduates" (2016-09-11)

Still, if Carnegie Mellon only provided a more social environment, that'd seem to help explain higher retention rates, but not higher enrollment. So, presumably they somehow advertised this more-social-culture. I'm not immediately seeing the details about how they accomplished that.

There're other issues that aren't so obvious. For example, apparently the same people claiming credit for this have been trying to increase female enrollment for decades now. So, why success only recently? Did they finally figure out some secret or fine-tune their process? Did their strategy take time to start working? Was this a fluke?

Whatever the case, a lot of the information and claims on this topic seem like bunk. And since Carnegie Mellon appears to have empirically demonstrated a method, it seems like they'd be a potentially interesting case study.

For further reading, Carnegie Mellon has a women's group called women@scs. Their page includes papers and a FAQ.

7
  • 8
    This is exactly the point: "it's the computer science culture — not the curriculum — that needs to change to accommodate women." And you have to communicate your culture. My university has a computer science program for women that advertises the group work, the collaboration, the insight into real business cases. Ther is no sudden magic words you can put into an ad to attract a diverse population, it has to be a culture change over time that gets communicated. It takes a lot of work. Commented Dec 21, 2017 at 16:11
  • 5
    Something else they did was to stop making previous programming experience a requirement for entering the major, because it was found that this screening requirement had no effect on the outcome of the quality of the graduate. But it had a very strong effect on the number of women who were qualified to be admitted to the major. Commented Dec 21, 2017 at 21:04
  • 5
    On enrollment vs retention: note that potential applicants to top-tier universities generally do a pretty large amount of research, and will discover information that's out there even if it's not literally an ad, whether it's through current/former students or official literature about the program. On the latter, if you look at the department's "Mission & History" page, one of the top-level values is "A supportive culture that brings out the best in people." So maybe official department information is a place for the OP to look too.
    – Cascabel
    Commented Dec 21, 2017 at 22:31
  • 5
    @Jefromi "Supportive environment" might sound like HR-BS, but it's actually very important. (Of course, it's even more important to have clear ideas about how to make the environment supportive and actually deliver on them.) People from non-underrepresented groups often don't understand how much effort URGs routinely expend while trying to either fit in or educate the people around them. This is extra friction which wastes energy that could be put into actual work. It's up to the program to step up and minimize this friction so that URGs have a more equal chance to succeed. Commented Dec 21, 2017 at 22:52
  • 4
    Carnegie Mellon simply employed technically good women in CS throughout the ranking. Senior women started attracting other women - that's how it worked for a few I knew.
    – Leon Meier
    Commented Dec 22, 2017 at 3:02
3

Most of the job ads I see suffer from different, more important problems:

  1. The ad doesn't clearly say who is expected (say, "researchers from all areas can apply"), but private talks point out that some areas are already full.

  2. There is a mismatch between the required documents according to the job and the documents that the web form requires.

  3. The job ad in your country's main language and the job ad in English don't match: they are not a translation of each other.

  4. It is unclear to whom the application has to be addressed.

  5. Precious space is spent on advertising the environment (city, university, "vibrant community", "major impact", public transportation, and the like). If the applicant is going to do the research job properly, she/he won't see the environment too much anyway, and they know that. They either know the surroundings (and are aware of the "major impact") or don't care. Usually, such a para smells fishy. E.g., Saarland is often claimed to be in the middle of Western Europe; this can be made true in some sense but doesn't help them in any sense...

First, make sure that your job ad has no issues in the above sense. Only then tackle the gender issues.

An overwhelming majority of the job ads I've seen so far have a para about welcoming female applicants; the strength of encouragement varies. However, this para drowns in other boilerplate-language paragraphs. Therefore:

  • Add a para about family/kids friendliness. After all, it's mostly women who give birth to kids and raise them up, and it's unlikely to change. You will get more women who are already pregnant or thinking about getting pregnant soon. It is a matter of discussion whether you really want it.

  • Mark your paragraph about prioritising applications from women bold and increase the font size! HTML allows you to do that!

  • If allowed by the legislation, don't use "they", don't use "the applicant", don't use "he/she", use "she/he" or "she", adding a disclaimer that men are also meant.

8
  • 8
    "If the applicant is going to do the research job properly, she/he won't see the environment too much anyway, and they know that" I completely disagree with that. Doing your job properly doesnt mean not having time to go to a city for a drink or sightseeing. Being located in a nice place is always a nice bonus.
    – Maciej
    Commented Dec 21, 2017 at 10:14
  • 6
    Expecting your new applicant to spend all their waking life on your job is EXACTLY how you keep women away.
    – CCTO
    Commented Dec 21, 2017 at 18:17
  • 1
    @CCTO That's EXACTLY what you have to tell to the person asking the question. Then he/she will have to take a decision about what is more important: progress in job or hiring women. This is exaggerated, of course: I knew women who were obsessed with their work, too.
    – Leon Meier
    Commented Dec 21, 2017 at 23:30
  • 2
    @LeonMeier Giving up to this US-borne paranoia is one of the huge factors destroying scientific community and I, for one, refuse to hop on this band-wagon. No great discovery, not even a relatively-important discovery was made by pulling 14h/day 7days/week insanity.
    – Maciej
    Commented Dec 22, 2017 at 14:24
  • 2
    @JiK You have to believe me on that: it my area, it is hundreds of good cadidates with supportive supervisors per position. Source: private communication.
    – Leon Meier
    Commented Dec 24, 2017 at 14:30
2

Problem is that grad students are treated like warm bodies to be drafted rather than early career professionals worthy of recruitment and competition.

In real situations, such as hiring for professors, any desirable attribute in prospective candidates is rewarded with improved offers. So to attract more female PhD candidates your department should identify capable applicants and offer them better stipend offers than they would get elsewhere. Making better worded recruitment materials should at least give you a better pool to recruit from but doesn't fix your stated demographics problem of bringing students in and supporting them through the PhD process.

It should also go without saying that your department should have female representation at every faculty seniority level. However many do not so I will say it. Having some token female assistant professors that you cycle through every 5 years in perpetuity doesn't count.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .