Hopefully everyone here can recognize that Ta-Nehisi Coates' post simply presents a hypothesis and contains absolutely no supporting data.
it is at least a plausible hypothesis
Racial views in the US are absolutely filled with (and arguably dominated by) gross misconceptions that seem plausible to many people despite being untrue.
Blacks do look seriously underrepresented based on the response rates to the two questions. (Suggesting that blacks account for around 4% of OkCupid's users.) That's hardly proof of anything, but it does offer some credence to Coates' argument.
In the real world, it's often very hard to find compelling data, so we sometimes have to rely on compelling argument.
Coates is pointing out that there are good reasons
to suspect strong sample bias.
OkCupid didn't address this -- they just assumed an unbiased sample.
What kind of data do you absolutely need...?
My statement was broader than intended. Obviously OKCupid's data is not necessarily reflective of all dating sites or interracial dating patterns overall, but Ta-Nehisi Coates is only presenting a hypothesis and some speculative claims, not data, and his speculation shouldn't be treated as anything more until someone presents data that supports it.
However, if I remember OKCupid's posts correctly, they explicitly pointed out that they weren't trying to generalize, and acknowledged that all of their conclusions applied only to the users of their site. I think it's everyone else who've tried to generalize their conclusions.
The author of the article from this thread is actually practicing skepticism. He is skeptical of the general conclusions drawn from the data set, and he proposes an alternate, simpler theory - selection bias - that would explain the behavior seen.
The author of the article from this thread is actually practicing skepticism