The U.S. Supreme Court's recent ruling legalizing marriage between same-sex couples in all 50 states follows on the heels of national polls showing rapid cultural changes in attitudes toward lesbian and gay people. A new University of Virginia study confirms that, by using a test designed to find that people have unconscious biases.
Building on previous surveys of self-reported attitudes, a team of scholars analyzed the results collected from more than half a million people between 2006 through 2013 by Project Implicit. It shows that even in 2006, implicit bias was not that bad. California, for example, is supportive of gay marriage but struck down a referendum legalizing gay marriage and then got a constitutional amendment against it, but only because the proponents wanted to use the law to sue churches and businesses that had an objection. Yet that is not bias, unconscious or otherwise, and so does not show up on a test designed to show people are 'secretly' prejudiced even if they are not.
The team found that implicit or "unconscious" bias against lesbian and gay people was 13 percent lower by 2013 and that explicit, or self-reported, bias decreased twice as much (26 percent) as implicit bias over the same seven-year period. This suggests that while many peoples' attitudes are changing at the deeper, unconscious level, some people may be less willing or able to acknowledge anti-gay bias than they were in years past. If this talk of unconscious bias confuses you, and how psychologists can determine that, it helps to know that even with a sufficiently powered test with real controls, the p=.05 value used in this test will be wrong 30 percent of the time, and in a survey could be wrong almost 90 percent of the time, so even if this test, popular among sociologists and psychologists, says you are unconsciously biased, you probably are not.
Implicit biases can occur outside of conscious awareness or conscious control, psychologists say. People may know that they have them and not be able to control them.
The authors also found that some people's attitudes were changing more quickly than others. Age, race and political orientation were the biggest predictors of attitude change. Unconscious bias decreased the most among women, as well as among white, Hispanic, liberal and younger people. Men - as well as black, Asian, conservatives and older people - showed the smallest changes in bias. Nearly all demographic groups showed decreases in implicit and self-reported bias over the seven-year period, suggesting that across the board, people seem to be developing more positive attitudes toward lesbian and gay people in general.
The findings were based on results of an online test hosted by Project Implicit, the controversial "Implicit Association Test" pioneered by Harvard, which asked participants to answer direct questions about their attitudes toward lesbian and gay people as well as perform tasks the creators believe measured unconscious attitudes. For the test related to this study, participants were shown pictures or words associated with gay people or straight people in the middle of their computer screen, such as two restroom-style female symbols side-by-side. On the right side of their screen appear positive words like "good" and "pleasant" and on the left side of their screen appear negative words like "hate" and "terrible." There are two parts to the task, the order of which is randomized among participants. In the first part, the participant is asked to pair positive words with straight people and negative words with gay people. Then the task switches, so that they're asked to pair positive words with gay people and negative words with straight people.
That's right, if you are faster at pairing negative words with gay people and slower at pairing positive words with gay people, you are prejudiced.
Upcoming in Collabra.