MIT Students Conducting Study To Predict Sexual Orientation Using Facebook Friends Information, Titled Project Gaydar


Two students at the Massachusetts Institute of Technology have partnered up to conduct a study in determining one’s sexual orientation. The basis is that those that tend to have many homosexual friends will end up homosexual themselves. The two students even designed software using specific algorithms to predict sexual orientations. The software worked only for men, but not for women.

“When they first did it, it was absolutely striking – we said, ‘Oh my God – you can actually put some computation behind that,’ ” stated MIT computer science professor Hal Abelson in an interview with Boston.com. “That pulls the rug out from a whole policy and technology perspective that the point is to give you control over your information – because you don’t have control over your information.” The two students involved are Carter Jernigan and Behram Mistree.

Based on the concept, the computer software algorithm could technically make assumptions about a person’s political preferences and emotions. As a matter of fact, Professor Murat Kantarcioglu at the University of Texas-Dallas used a similar concept to predict political preferences. Kantarcioglu and a student used 167,000 profiles to make 3 million links.

In the case of the Gaydar project, the students used a sample size of 1,544 straight men, 21 bisexual men, and 33 gay men. The 33 gay men had a great proportion of gay friends than the straight men.

This article was written by Amit Chowdhry. You can follow me at @amitchowdhry or on Google+ at

MIT Students Conducting Study To Predict Sexual Orientation Using Facebook Friends Information, Titled Project Gaydar Comments

  1. mrchen says:

    This article written by pretty good, if you want to see more interesting topic, please pay attention to my blog!

  2. mrchen says:

    This article written by pretty good, if you want to see more interesting topic, please pay attention to my blog!

  3. JB says:

    I suspect this will work on groups that have an associated subculture — i.e., if you are into the gay culture you probably also (1) are gay and (2) have many friends from that subculture (who are gay). Perhaps lesbians and bisexuals are less affiliated? (Like I would know.) I would hypothesize this will work on groups that have sub-culture: Pagans, insular Christian denominations, hippies, goths, members of activist movements, and people with certain disabilities that either have there own culture (e.g., Deaf) or affiliate a great deal through support groups, all might be example of groups it might work on. Also, high schools and universities someone has attended, and home towns to a lesser extent by extension. Probably not major political parties, mainstream musical generas, etc., and probably would not be able to tell who is professes to be Christian. Again, I suspect this will hit strong for and based on (sub-)cultural / crowd membership and geography — and probably not all traits.

Leave a Comment