Saturday, July 16, 2016

If Women Discriminate Against Women, Is It Still Discrimination? Evidence From Studies.

Here's something for you to chew on:

Several studies  I've written about here over the years show that it's quite possible for not only men but also women to discriminate against women in the sense of pure sex discrimination.

Examples of this can be found in studies about customer ratings of sale force, about student mentoring and selection, about the rating of plays by critics where the plays are the same but the presumed name of the author has been changed, and about how much a fictitious worker should be rewarded for some project when that worker's accomplishment remains the same but his or her sex is changed*.

If women do it, too, at least in some studies**, does this mean that what we are seeing shouldn't be called sex discrimination?  Even if it affects women negatively as a class?

Would your answer to that be different if I told you that I have found studies which show something comparable when it comes to ethnic or racial discrimination, where members of racial and/or ethnic minorities are as likely to discriminate against their own group as are whites***?

It's important to stress that I have picked these studies out of much larger fields, that other studies don't necessarily show being a discriminator as an equal opportunity role, and that I'm saying nothing about the relative prevalence of the above results, just that they exist.

How would we explain the phenomenon of women and/or members of racial or ethnic minorities treating "their own" with bias?

My preferred explanation is twofold:

First, we are all like the fish who swim in the water and cannot tell if it has a taste because they have never experienced the alternative.  This means that both women and men, say, have been largely exposed to the same gender stereotypes, that both women and men have seen more men in positions of leadership and, in general, have been told roughly the same stories about the proper roles of women and men in the culture.

Second, all that is hiding inside most of us, and may crop up in situations of the type the linked studies create.  I believe that the responses with bias are largely subconscious ones, not explicit choices where some kind of inner conversation goes on at the same time as the choice is made.

As an example, study subjects pretending to be the representatives of a firm might award a female employee a smaller bonus than an equally productive male employee  not because the subjects believe that women deserve smaller rewards for the work, but because women, on average, get smaller rewards and some women, at least, seem to accept them, and if you are playing the role of the employer, paying smaller bonuses whenever possible is a good thing.

And to answer my own headline:  Yes, it is still discrimination.  Who does the discriminating is irrelevant from that angle.  As long as the outcome is to treat equally productive/talented/etc. women and men differently, we are talking about sex discrimination.

Now to the question of how prevalent these attitudes might be.  I cannot answer that, because the studies that look at the sex, race or ethnic affiliations of those who do the discriminating are uncommon.  But I know that not all studies demonstrate such equal-opportunity discrimination.****


*  I can't find the link to this study, even though I'm 90% certain that I covered it on this blog and 100% certain that it exists.  The study is an audit study where the study subjects are asked to tell how big a bonus a fictitious employee should get.  In some scenarios this employee is female, in other scenarios male, but in all cases the job achievement to be rewarded is the same.  Both female and male study subjects would have awarded the fictitious male worker higher bonuses.

It's also worth pointing out that the play review study doesn't show that female critics rated plays with a female author name as inferior in the sense that they themselves found them inferior, but only in the sense that they predicted those plays would do worse in the marketplace.  That's somewhat different from the knee-jerk type reactions the other studies seem to suggest.

**  A recent example where female study subjects were not biased against women whereas male study subjects were can be found here. Other studies also demonstrate different levels of bias against women by men and women.  A meta-study which found that men were rewarded more for equal performance evaluations found the only exception in the industries with a high number of female executives.

***  A quote from the Airbnb study on racial discrimination:
On the whole, we find that results are remarkably persistent. Both African-American and White hosts discriminate against African-American guests; both male and female hosts discriminate; both male and female African-American guests are discriminated against. 

A quote from the study on mentoring by faculty members prior to the student's enrollment:

We have reported two counterintuitive findings: 1) representation does not reduce bias and 2) there are no benefits to women of contacting female faculty nor to Black or Hispanic students of contacting same-race faculty. These results are consistent with past research showing that, counter to perceptions (Avery, McKay, & Wilson, 2008), stereotypes are potentially held even by members of the groups to which the stereotypes apply (Nosek, Banaji, & Greenwald, 2002) and that female scientists are just as biased against female job applicants as male scientists (Moss-Racusin et al., 2012).
****  I also think that experimental studies of various types might make the study subjects more careful about revealing the kind of sex or race discrimination that is based on much more overt dislike or antagonism, especially if the study gives hints about its purpose.  It's possible that such studies don't capture all aspects of real-world discrimination of the most overt kind.