The white male recipient column also suggests an interesting supply-demand effect.? crude remarks he made in an interview, suggesting that he was not sexually attracted to African-American women.? oktrends.com, an associated blog where the dating service is refreshingly open about reporting the results of all kinds of interesting number crunching.?Tiger Woods‘s alleged actions suggest a similar preference (Bill Maher quipped “He doesn’t need sex rehab; he needs diversity training”). analysis of messaging at Ok suggests that discriminatory male preferences are a wider phenomenon. (I predict that the website’s database will become the basis for more than one Ph. dissertation.) Ok Trends looked at “the messaging habits of over a million people.”?Men (including African-American men) write back to African-American women at about a 20% lower rate.? famous resume study done by Marianne Bertrand and Sendhil Mullainathan, which found that employers who place want ads were less likely to respond to resumes from people with African-American sounding names.But in some ways the Ok Cupid result is even more depressing than the racial disparities found in employment.?They wanted to find out what happens after one user sends a message to another user.? The basic result is that the race of the sender matters a lot – even after taking into account the user-defined level of compatibility.
For example, what happens when the sender is a black woman and the recipient is a white man??But there are so many permutations (regarding sex, race and sender/receiver status) that there would be many, many lines to analyze.Still, the following chart shows that the average quality of the match for different racial permutations is roughly similar: As you can see, the races all match each other roughly evenly: good news.It means all other things being equal, two people, of whatever race, should have the same chance to have a successful relationshp.
It seems that Ok Cupid doesn’t match couples where the match would be inconsistent with an explicit racial preference of a user.?So these racial disparities persist even after excluding users who have stated an explicit racial preference.