I love how OKCupid has turned into not just a dating site, but a huge source for aggregating data.
Apparently white men are more likely to like Tom Clancy, golfing, and The Big Lebowski (why do none of those surprise me?), while white women are more likely to like the Red Sox (that one does surprise me), Nicholas Sparks, and flea markets.
Black men are more likely to like basketball, Mos Def, and ESPN (which I would have thought would be race-neutral). Black women are more likely to like soul food (A LOT), The Color Purple, and “the Lord.” Apparently black people mention religion in their profile more than any other race.
There’s more, but click through to see it. This is fascinating stuff.
OkTrends always has some very awesome data analysis. Besides the obvious statement of saying Alicia Keys pretty much wins the internet. I wanted to see profile complexity broken down by gender, damn, maybe next time.
With how much analysis that OkCupid has shared with their users - as well as the marked comparison of how statistically likely it is to get you chatted with and/or dates you have to question how much their souring their own data pool for future stats. Then again maybe they’ll do the analysis on profile change when related to oktrends postings.
On a side note: the Coleman-Liau index is an interesting thing - yes, complex sentences usually indicate a larger vocab and what not but sloppy, poorly written sentences and slang and related can create false scores (not necessarily false high or low.) I suppose aggregate volume is expected to normalize the data set…
