do women make (insignificantly) better social science researchers?

researchers at newcastle illustrated (PDF) that simply a picture of eyes can significantly increase likelihood that individuals will contribute to an honor-system box for drink payment. this is similar to a study last year (PDF) by researchers at UCLA in which eyespots significantly increased generosity in anonymous one-shot dictator games.

something else i found cool about the study, although it's technically insignificant: the newcastle study hinted at an inverse linear relationship between femininity and contribution (see chart, page 2). the most feminine eyes induced the fewest contributions and the most masculine the most, with eyes straight out of a ann rice novel in the middle (??). it's a straight linear line, with no outliers - too perfect to be coincidence! who really believes in statistical significance anyway?

so, although men may be better at administering medical care, women may be better than their male counterparts at leading behavioral experiments by influencing their actions less. this study only ran for a few months, with only two sets of female eyes. i'd like to see a larger study with higher variation in both femininity, age, and expression.

of course, insignificant gender correlations aside this has interesting implications for social science experiments involving prosocial behavior. regardless of whether the eyes activate something that harks back to group selection or simply to a reminder that another person may enter the room, researchers must be very careful to minimize felt observation.

Labels: social