One of the most important ideas in gender today is the Gender Similarities Hypothesis. This hypothesis was proposed by Janet Shibley Hyde, a professor at UW-Madison. It was originally published in American Psychologist, as a direct contradiction to the more commonly held Gender Differences model. The Differences model argues that male and females are vastly different from one another biologically and psychologically.

Hyde tested the opposite hypothesis, that men and women are actually more similar to each other than different, by analyzing meta-analyses of gender differences. A meta-analysis is a procedure by which researchers will review hundreds, and sometimes thousands, of studies in order to determine whether a hypothesis is true or not. Meta-analyses are important since one study alone is unlikely to prove an effect. Furthermore, meta-analyses often include studies in their review that were not published in an academic journal. This gets around the "file-drawer" problem: presumably, many studies which prove that something is untrue are lying around in the file drawers of scientists. This happens because it is nearly impossible to publish a study that proves something is not true, only something that suggests that it is true.

Hyde examined the major meta-analyses that had already been conducted on gender differences. In a sense, she did a meta meta-analysis in order to see what gender differences are true or not. She started out with the hypothesis that most differences between the two sexes are negligent to non-existent. In fact, that is exactly what she found-with a few exceptions. The largest gender differences are in the domain of motor performance (such as throwing velocity and distance). A second area is sexuality, particularly in reported masturbation activity, and attitudes about casual and uncommitted relationships. Although much publicity has been given to gender differences in aggression, the differences are only moderate. Furthermore "relational" aggression which has been publicized as more common among girls shows no consistent gender difference.

For those who are at all curious, I highly recommend downloading the full article here. Many of the results are surprising - gender differences in math and verbal ability overall are quite small. Moreover, gender differences in traits like assertiveness, self-esteem, and even height are also quite small.

The question is then, why do most of us believe so strongly in gender differences despite the evidence that shows they are minimal for most things? For one, overinflated claims of gender differences appeal more to our intuitions. They sell more magazines and newspapers. They make for interesting non-fiction book titles, and they allow researchers to publish papers that gain them scientific recognition. Perhaps we start out believing in gender differences and therefore see them wherever we look.

Unfortunately, as Hyde points out, there are big costs to our beliefs in gender differences. Social psychologists have shown that beliefs often lead to self-fulfilling prophecies. In other words, the more we believe something is true, the more we are likely to act in a way that makes it come true. Men are taught to believe that they aren't good at communicating, that they lack tact, and are not good at interpreting emotions. Women are taught to believe that they aren't cut out for leadership, they're bad at math, and they should stick to certain careers that bring out their "natural" abilities. The costs for our beliefs are huge for both genders; and, given the lack of scientific data to support any of them, ought to be seriously re-evaluated. At the very least, I hope that anyone who has read this blog will have an extra grain of skepticism the next time they encounter a so-called gender difference in print or in person.