Thursday, December 11, 2014

Two Studies on Differences Between Men And Women At Work


These studies are, first,  on comparing men and women who acquired a Harvard MBA on various measures of success and satisfaction, and, second,  on the ratings online professors get, based on the gender the students think they belong to.

They both share a certain flavor of not being the last word on the topic, but they are also worth dipping into for what they can tell us.

 

1.  The Harvard MBA study surveyed men and women who can be said to be ambitious, competent and at least somewhat career-oriented.  The survey wished to establish why a gender gap among senior management prevailed in that group of highly educated individuals:

Among those graduates who are employed full-time, men are more likely to have direct reports, to hold profit-and-loss responsibility, and to be in senior management positions. Setting aside those measures of success, since not everyone aspires to them, we found that women are less satisfied with their careers. Whereas about 50% to 60% of men across the three generations told us they were “extremely satisfied” or “very satisfied” with their experiences of meaningful work, professional accomplishments, opportunities for career growth, and compatibility of work and personal life, only 40% to 50% of women were similarly satisfied on the same dimensions.
But none of the theories usually proposed in this context succeeded in explaining the gender gap, not the greater family responsibilities women may be assumed to have, not any sort of opting out or longer breaks from the labor market.  The researchers tell us that they could not explain this particular gender gap by using the data they had in the survey.

I haven't seen the data itself, but asking individual workers about their personal views doesn't tell us anything about the institutional constraints or about statistical discrimination (putting mothers on mommy tracks whether they wish to be there or not, say) they may be facing.

More work is needed on that question.

The study did have interesting findings about the difference between expectations* and reality when it comes to whose career should take precedence in the family and who is expected to be responsible for hands-on childcare.  On both of these measures men's expectations at the point of graduation from Harvard were much more traditional than women's expectations: 

More than half the men in Generation X and the Baby Boom said that when they left HBS, they expected that their careers would take priority over their spouses’ or partners’....
Notably, this expectation was less prevalent among men of color than among white men. Forty-eight percent of the former—compared with 39% of white men—anticipated that their spouses’ careers would be of equal importance. Meanwhile, the vast majority of women across racial groups and generations anticipated that their careers would rank equally with those of their partners. (Only 7% of Gen X and 3% of Baby Boom women, and even fewer of their male counterparts, expected that the woman’s career would take priority over the man’s—an arrangement we call “progressive.”)
Reality turned out to be even more traditional than the men's expectations at the point of graduation.  This could account for the lesser career satisfaction women in the survey declared.

Similar results apply to expectations and reality about the sharing of childcare duties:

We had asked a parallel set of questions about child care: How had graduates who were expecting to have partners and children (91%) anticipated dividing child care responsibilities when they left HBS, and how did they actually divide them?...
At the time they graduated from HBS, more than three-quarters of men expected that their partners would do the lion’s share of child care. Black men were somewhat less likely to expect such an arrangement. Meanwhile, about half the women expected that they would take on the majority of this work. Latinas were the least likely, at 40%, to have expected to shoulder most of the child care.

Once again, reality matched the traditional expectations better:

Ultimately, more-traditional arrangements did win out. Healthy majorities of Gen X and Baby Boom women took responsibility for most of the child care in their families. Even higher percentages of Gen X and Baby Boom men reported having spouses who did so. Black men and women were the least likely to have a traditional arrangement; their numbers were lower by roughly 15 to 20 percentage points.
And here's an interesting paradox:  The researchers point out that roughly half of the women who expected their careers to be given the same importance as their partners' careers also expected to do most of the childcare.

But those two expectations conflict with each other:  If a woman does all the household chores and child-rearing, takes time off when the children are sick and so on, her career will soon suffer from either her absences or her fatigue, whereas her partner's career might even benefit from that.  It's likely, then, that he gets salary raises and promotions and she does not.  After that holding her career equally important doesn't make monetary sense.

The linked survey has nice pictures of all these findings by generational cohorts and more information on ethnic and racial differences in the sample.

What's the take-home-lesson from this?  Mine would be that the second wave of feminism focused on equal treatment of men and women in paid work but left the question how to divide unpaid work at home unanswered.  We are still struggling with the latter.  As long as women are going to do the majority of household tasks women are also going to be less well represented in various leadership roles outside the home.

2.  A pilot study on student ratings of online professors by Lillian MacNell, Adam Driscoll and Andrea N. Hunt suggests that being female costs a professor quite a bit in student ratings.  This finding must be taken with some reservations, because the study used very small student samples:

To address whether students judge female instructors differently than male instructors, the researchers evaluated a group of 43 students in an online course. The students were divided into four discussion groups of 8 to 12 students each. A female instructor led two of the groups, while a male instructor led the other two.
However, the female instructor told one of her online discussion groups that she was male, while the male instructor told one of his online groups that he was female. Because of the format of the online groups, students never saw or heard their instructor.
At the end of the course, students were asked to rate the discussion group instructors on 12 different traits, covering characteristics related to their effectiveness and interpersonal skills.
“We found that the instructor whom students thought was male received higher ratings on all 12 traits, regardless of whether the instructor was actually male or female,” MacNell says. “There was no difference between the ratings of the actual male and female instructors.”
What's nice about this study is the way it can hold everything else the same (personal characteristics, instructor competence, course material) while letting the presumed gender of the instructor vary.  The differences the study finds, then, cannot be for any other reason except that gender.

The reason why the study should be taken as a pilot (like a pilot boat leading a long row of ships) is because of the small sample sizes in the four groups.  A few extremely bigoted students could drive the findings in small samples.**

It's important to carry out more research into this topic.  Note that a professor's salary and advancement can depend on student evaluations.  If those evaluation themselves are biased against women, female professors will be running the promotions and salary races with an extra weight around their ankles:  One which has nothing to do with actual competence.

----
*"Expectations" is intended to measure what the survey participants thought would happen when they graduated from Harvard.  "Reality" is what actually happened.  As far as I can tell (and I may be wrong), the questions about expectations were asked now, not in the past when the person actually graduated.  This means that "expectations" should perhaps be called "remembered expectations."

**I have not had access to the study itself.  That means I cannot tell if the bias against a professor presumed to be female was found across all students or not.  

Neither could I find out, without access to the study, what the online course covered.  It's theoretically possible that sexist bias would only apply in some kinds of courses and in some fields and even that a reverse sexist bias is possible in other kinds of courses and/or fields.  I doubt that because authority is coded male in the US culture, but the possibility exists.