Friday, January 15, 2016

On Neanderthals and Political Debates

That headline is quite likely to make you expect something different from this post than what it actually contains!  And that's part of its lesson.  Let's start:

I spotted this headline a few days ago and stashed it away as a potential source for delicious thinking:

You might have a peanut allergy because your ancestors had sex with Neanderthals

What's weird about that headline is pretty obvious:  The Neanderthal in that headline would also be your ancestor if the described sex was productive, and if it wasn't productive that ancient nookie cannot explain your peanut allergy.

So I began this post with what looks like hair-splitting.  But it isn't, really.  It can be a valuable lesson about how we introduce new information (the idea of a Neanderthal ancestry of some humans is a novel thesis) and how we interpret information in general.

People tend to have a certain prior structures (or strategies) of thinking and those can affect what we read from certain information.  In this example the prior structure is the assumption that the Neanderthals didn't pass their genes on to any currently existing human populations, that they were out-competed by the Cro-Magnons*, that they were a genetic dead end.  Besides, they were ugly, and "we" beat them!

Thus, a useful starting point is to assume that most readers of that headline see the Neanderthals as outsiders, not really in the ancestry chain, and to inject them into the conversation from that starting position, as if our "ancestors" had a bit of illegitimate exotic sex on the side.**

The political implications of this are enormous, my erudite readers, enormous!

Just spend some time reading how the right-wingers wrote about the Cologne mass sexual harassment and then how the left-wingers and feminists wrote or didn't write about it.  Granted, those particular knots have many more threads intertwined, but one important thread consists of the thought structures people have earlier built about and around the issues.

And by those I don't mean solely the basic value structures, but also personal experiences, the selective sieves we use to filter reality, the values of the groups we belong to and how those groups can give us both pleasurable praise or make us suffer from shunning, and, of courseAll those affect the information we see and how we go about interpreting it, and any new information will be interpreted within those thought structures we already have and the older information we have accepted into them.

None of that is new thinking, of course.  I'm sure that experts have proper names (or detestable acronyms) for the phenomena I grasp with here.  Still, it can be worthwhile to enter an old room through a new door, right?  
* We are not supposed to use that term anymore.  But, honestly, the proposed alternative EEMH for European early modern humans is nowhere near as memorable.  Besides, I detest acronyms almost as much as I detest raisins.

**  Assuming, of course, that EEMHs and Neanderthals did interbreed in Europe or Asia.


Thursday, January 14, 2016

The Old Canard Returns: Women's Lower Earnings Are All Because of Choice

Dana Loesch, a conservative television host on The Blaze tells us that women earn less ONLY because they "choose" to earn less, that there is no gender wage gap attributable to anything else, and that everybody knows this to be the truth.

I love her audacity and deplore her stupidity, the latter, because economists who actually study the question don't agree with her conclusions.  The world is complicated and the economics of the gender gap in earnings are even more so.

Does it ever annoy you that it's much easier to make soundbites which are not true than to present what is known, so far?  It annoys me, because I'd like to shoot a short soundbite back at our Dana.  Instead, links to relevant material must suffice:

This post provides links to several others for those who really want to get their teeth on the topic.   This post, as well as this one, have more on the idea of "choice" in this context.

Finally, this post talks about a study which the conservatives have widely used to draw the conclusion Loesch makes.  More on the topic here.

I get very angry at the usual conservative political lies about the earnings gap between men and women and also get somewhat angry at the lefty political interpretations of the gross earnings gap* as being completely due to discrimination against women.  The truth is somewhere in the middle.  But no way can we state that it's only feminine choices** which cause the average earnings of women to be lower than the average earnings of men.

*  The gross gap hasn't been adjusted for anything.  It's the net gap, calculated after data is adjusted for non-discriminatory factors such as working hours (if the earnings data is not per hour), education and experience, that economists actually analyze when trying to measure the impact of discrimination on earnings.

In other words, not all of the observed average difference is attributable to differential treatment of women and men in the labor market.

**  As I write in one of the linked pieces, the meaning of "choice" matters here, too.  If the society expects women to be in charge of most child-rearing, shopping, cooking, cleaning and laundry then the options between which individuals choose are already affected by their gender.

Things get even more complicated when we acknowledge that the US labor markets are still largely based on the assumption that someone else than the worker takes care of those chores.

Conservatives toss the term "choice" into the conversation as if everybody was choosing from exactly the same menu for their lives.  But that's not the case, on average, and even less so for poorer workers whom the labor market treats with greater harshness and fewer perks.

Note, also, that there is trivial choice, such as choosing between drinks at dinner, and non-trivial choice.  The latter is the proper interpretation for the production of the next generation.  But most conservatives treat this topic as if having children is just an expensive hobby of no greater societal significance.  Odd, given how "family focused" many of their organizations are.

Wednesday, January 13, 2016

Teamwork in Economics Research: Who Gets Credit?

The ivory towers of academia are not a place to lounge on your laurels.  Rather, professors compete ferociously for the diminishing numbers of jobs with actual benefits and job security, i.e., tenured jobs.  And in that competition being female may work against you, at least according to a new study* which looks at how much credit researchers get for writing articles.

Publications are important, because of the publish-or-perish principle of tenure acquisition.  Heather Sarson's doctoral thesis looks at the impact of gender on how much credit someone in the economics field gets for her or his publications.  What she argues her study shows is this:  Women get less credit for co-authored articles than men, but roughly equal credit for articles they have written alone:

While women in the field publish as much as men, they are twice as likely to perish. And this higher rate for women being denied tenure persists even after accounting for differences in tenure rates across universities, the different subfields of economics that women work in, the quality of their publications and other influences that may have changed over time. 
But Ms. Sarsons discovered one group of female economists who enjoyed the same career success as men: those who work alone. Specifically, she says that “women who solo author everything have roughly the same chance of receiving tenure as a man.” So any gender differences must be because of the differential treatment of men and women who work collaboratively.
The career benefit from publishing a solo paper is about the same for women as it is for men. But unlike women, men also get just as much credit for collaborative research, and there is no statistical difference in the career prospects of authors of individually written papers and those of papers written as part of a research team.
Unfortunately for women, research done with a co-author counts far less. When women write with co-authors, the benefit to their career prospects is much less than half that accorded to men. This really matters, because most economic research is done with co-authors.

What's going on there?  To learn more, Sarsons looked at the possible impact of who you are writing those articles with, and found that what really hurt women was co-authoring only with men.  In those cases the women got essentially no credit  for the article.  Men, on the other hand, received approximately the same credit for co-authored articles independently of the gender of their co-authors.

Justin Wolfers, the author of the NYT piece on the study, suggests that whenever there is ambiguity about how to divide the credit for an article those who are to decide on that division tend to fall back on their (probably subconscious) biases.  He also notes that Sarsons carried out a parallel study in sociology and failed to find any gender effect.  That might be because in sociology the first author listed is the main author of the article, whereas in economics the authors are listed alphabetically.  The former practice gives some help in deciding how much credit to give to someone (more to the first author), the latter leaves that more open.

Finally, this quote from the NYT piece is interesting:

Many female economists have shared with me their experiences of research being taken less seriously simply because it was written by a woman. The great economic historian Deirdre McCloskey, a distinguished professor at the University of Illinois at Chicago, has a unique perspective on all of this, having spent the first half of her career as the male economist Donald McCloskey. Today, she reports that it is quite common for her colleagues not to acknowledge a point she has made until it is reinforced by another male economist. That rarely happened when she was Donald.

I was trying to think of alternative theories which would explain Sarsons' findings, such as asking whether the gender of one's co-authors gives a different signal about male and female authors.  Something along the lines that, on average, women in economics are at earlier stages in their careers than men, so perhaps women are more likely than men to co-author with more senior researchers, and senior researcher tend to get more credit.  But Sarsons tests that (p. 24) and finds it not to be the case.

Her findings have wider importance, if similar considerations affect how employers divide the credit for teamwork more generally.  Indeed, the findings may be stronger in areas where there is no choice whether to work in a team or go solo.


This seems to be the working paper on the topic.   Sarsons writes:

While women who solo-author everything have roughly the same
chance of receiving tenure as a man, women who coauthor most of their work have a significantly lower probability of receiving tenure. The penalty is not explained by coauthor selection and is robust to controlling for productivity differences, tenure institution, year of tenure, and field of study.