Monday, November 22, 2004

How to Report Gender Science Results



This is an interesting article. It refers to a 2004 study about how newspapers report on research in gender sciences, in particular, whether the newspapers stress a biological or a cultural/environmental explanation to any findings of behavioral differences:

Two Yale University researchers--Victoria Brescoll and Marianne LaFrance--analyzed articles on sex differences that appeared in 29 large-circulation U. S. newspapers published between January 1994 and February 2001.
After going through all that, they found that the political leanings of newspaper publishers and managers color reporting on sex differences. While conservative newspapers tend to use biology to explain those differences, more liberal newspapers explaining them in terms of socio-cultural effects.
...

The study, published in Psychological Science (Vol. 15, No. 8, August 2004), raises serious questions about how well science journalism serves newspaper readers.
The articles were coded for the type of explanation provided for sex differences and also for the degree to which the newspaper was conservative or liberal and the degree to which the newspaper articulated traditional sex role beliefs throughout its pages.
Brescoll and LaFrance also ran experiments to see if articles proposing biological explanations for sex differences would help foster gender stereotypes. Not surprisingly, the answer was yes. When faced with press coverage that favors biological explanations, guess what, readers' gender stereotypes are indeed reinforced.


This is one my fields of interest, and I have been following the reporting of studies, as well as the studies themselves, for quite a long time. The authors of the study are correct in their findings: most popularizations of scientific findings are terrible and appeal to our stereotypes and prejudices, whatever they happen to be. But mostly they actually appeal to those views that would maintain the status quo. There is a pretty strong bias towards explaining all results about gender differences as inborn. I would be a rich goddess if I had been given a dollar every time I read a headline stating that :"Girls are different! And it's in our genes!" Or one about why women can't read maps (it has something do with the map reading brave prehistoric warriors had to engage in while the women were watching the prehistoric equivalent of today's daytime soaps in some cosy cave).

There are the opposite popularizations, too, of course. But they are nowhere near as common. I can only remember two items in the last ten years which actually argued for a hundred percent cultural explanations to all behavioral sex differences.

But the actual studies are far too often not much better. Spend some time googling "evolutionary psychology sexual selection" and I bet that your hair will stand up. There are quite a few researchers in that field who have an axe to grind, and the point of the grinding is to prepare the axe to chop off some uppity female heads. Never mind if there is no fossil evidence or evidence about the psychology of our prehistoric ancestors: the results always conveniently support the researchers' own biases.

This is probably unavoidable, but a field of this type tends not to attract people with the opposite bias in adequate numbers. Thus we are left with an ideologically non-neutral science. Maybe more feminists will become evolutionary psychologists in the future, even though this is supposed to be a contradiction in terms right now.

The truth about most of the studies which compare men's and women's brains and so on is that the research is taking its very first infant steps, and any sort of conclusions are very premature. For example, differences in the PET scans between the sexes doesn't necessarily mean that these differences are inborn. It is known that how the brain is used affects its structure. (See, for example, the study of London taxi drivers who have unusually large areas reserved to long-term memory because of the intensive memorization of street addresses that is required for getting the London taxi license. Also, studies about the effects of depression on brain show long-term changes caused by the condition.)

The immaturity of the field doesn't stop the popular interpretation of every new research as the major breakthrough which explains why women do what they do and so on. These stories hardly ever give any statistical results, because to do so would reveal the great variability of the characteristics within each sex, as well as the large amount of overlap in the distributions of results for the two sexes. Neither do most of these popular reports even mention that whatever behaviors we observe today may also have cultural explanations, or that the nature-nurture debate is actually a lot more complicated than the simple idea of "one or the other" suggests.

I don't know if it is really possible to report gender science neutrally, but it certainly isn't happening right now. Most reports are intended to prop up a traditional view of the sexual division of labor by explaining that it is biologically decreed. This feeds right into the wingnut ideology for those wingnuts who don't have the fundamentalist interpretation for women's inferiority. Even the studies themselves have a certain type of bias in this sense: How often do you hear about studies that found no sex differences in behavior? Yet clearly men and women do an enormous number of things exactly the same.

I'm not neutral in this field, either, of course. When someone tries to sit on me and tell me that I am coy and timid and uninterested in sex and really bad at map-reading and unable to fix my car and that all this has been proven by Science, well, the correct response is to take the other side and to fight back. But I don't presume to know the exact amounts by which our behaviors are predetermined. I wish more people adopted the same quasi-humble attitude.