'Gaydar' More Effective On Women, Says Study
Psychologists fascinated by gaydar recently offered over 120 college students pictures of gay men and women to determine whether the gaydar instinct differs between gender. Turns out it does.
In the study, 129 college students viewed 96 photos each of young adult men and women who identified themselves as gay or straight.
Concerned that facial hair, glasses, makeup and piercings might provide easy clues, the researchers only used photos of people who did not have such embellishments. They cropped the grayscale photos so that only faces, not hairstyles, were visible.
For women's faces, participants were 65 percent accurate in telling the difference between gay and straight faces when the photos flashed on a computer screen. Even when the faces were flipped upside down, participants were 61 percent accurate in telling the two apart.
At 57 percent accuracy, they had a harder time differentiating gay men from straight men. The participants' accuracy slipped to 53 percent – still statistically above chance – when the men's faces appeared upside down.
According to the study's lead author, University of Washington psychology graduate student Joshua Tabak, the reality of gaydar undercuts conservative arguments that keeping sexuality secret will eliminate homophobia.
"Gaydar may be similar to how we don't have to think about whether someone is a man or a woman or black or white," he said. "This information confronts us in everyday life."




Facial recognition goes way down when you look at upside-down pictures. People won't recognize friends, relatives when the pictures are upside down. So it's interesting that even upside-down, people may be picking up on some sort of cue.
As for no facial hair, make-up, hair, etc., this makes absolute sense. It has nothing to do with pre-assumed stereotypes. It is about focusing on the face with no distractions which might cause a person to make assumptions based on whatever specific experiences the viewer might have. It is about eliminating a confound.
Posted by: TJ | May 16, 2012 6:36:27 PM
I don't understand why each time they do these studies they seem to spin it as "proof that gaysar exists" when the participants guessed accurately at about the same rate as a blind coin toss, give or take a margin of error.
The summary doesn't clarify very well the numbers, but it sounds like it may have been 96 gay women, 96 straight women, 96 gay men and 96 straight men.
If that's the case, a coin toss by a blindfolded monkey would have been close to the numbers and might easily have even beat them.
Posted by: Gregv | May 16, 2012 7:41:56 PM
@GregV
Uh... What size margin of error are you imagining? Can you please show the math because what you are making is an extraordinary claim. You might want to provide some evidence for that belief. Have you ever taken a science course that required experiments in the lab? I don't mean this as an attack, it is just that at a glance, the numbers were pretty convincing to me and I didn't believe in gaydar before reading this. If you are right, I have to reread this and figure out how I went so wrong.
A blindfolded monkey tossing a coin should have no less a chance of being right than you would tossing a coin. That is what makes a coin toss "fair." No one has an edge. Not even a man against a blindfolded monkey. Why the extra bit about blindfolded monkey?
Posted by: NullNaught | May 16, 2012 8:16:52 PM
@Nullnaught: Okay, I'll clarify for you.
If a trained, blindfolded monkey were presented with the upside-down photos (which he wouldn't even see, because he's blindfolded) and tossed a coin (heads for gay, tails for straight) he would be right on average 50% of the time. (That's assuming that the photos were of 96 gay women, 96 straight women, 96 gay men and 96 straight men.)
But the human beings looking at those upside-doen photos were only right 53% of the time (which is barely different from what that monkey would come up with.
Studies usually have a number larger than that as their margin of error. And a blindfolded monkey tossing coins would lose to these people slightly more often than not, but the numbers are so close that he would beat them ALMOST as often as he would lose (if given, say, 100 coin tosses each time). As I said, the numbers are so close that it wouldn't be so surprising for the monkey to beat the people in the study several times.
And yes, I have taken statistics and done experiments in courses.
Posted by: Gregv | May 16, 2012 9:46:21 PM
This is just a whole lot of wasted money on a nonsense study. But, I guess it kept food on somedody's table. Mabye somebody did an MA or PhD thesis on this. LOL!
Posted by: andrew | May 16, 2012 10:22:00 PM
@Gregv
You don't have the reading comprehension skill necessary to duplicate my argument. Trying to explain anything to you in writting would be a waste of time, as you would likely not understand. When I first read your original post, I thought you had observed something I didn't. We read the numbers differently.
Go ahead an ignore the findings of scientists. It makes no diference in your personal life.
Posted by: NullNaught | May 17, 2012 12:03:25 PM
@Nullnaught: So your argument comes down to nothing but ad himinem, I see; You won't bother finding any error in my calculations because I "probably wouldn't understand anyway."
I did not find a very detailed transcript of the study and am open to anyone whites it to explain to me that it was anything other than four categories of people (gay females and males, straight females and males) with 96 participants each. That is what it sounds like it was from the brief descriptions in the various summaries I read.
But ad himinem is a fallacious argument style, so if that's all you're offering here, I find it doubtful that you found dug up more information and have any basis to dismiss anything ai said.
In an experiment with a higher-than-3% margin of error (highly likely), numbers that are as low as 3% above blind guesses are not impressive evidence that a phenomenon exists.
Posted by: Gregv | May 17, 2012 1:15:17 PM
Edit: " anyone whites it explain to me" should read "anyone who has read it and can explain to me."
Posted by: Gregv | May 17, 2012 1:18:31 PM
Oh and to Nullnaught's question about why I referred to a blindfolded monkey tossing a coin: Of course you are right that ANYONE tossing a coin has the same chance of heads or tails (as long as he's not cheating) but I am just emphasizing that anyone (whether human or animal or machine) who/which does not even understanding the question itself could have gotten the answers right, on average, 50% of the time. I cod just as easily said that I could have put a cat on a black and white checkerboard floor (black for gay and white for straight) and asked her to answer "is this upside-down man gay or straight?" by sitting down somewhere. She would be correct 50% of the time (almost as often as the people in the experiment).
Posted by: Gregv | May 17, 2012 1:29:09 PM
Edit: "cod" = "could;" "understanding" = "understand.". (I wish I could turn off auto-correct."
Posted by: Gregv | May 17, 2012 1:33:24 PM