2 ms·

I worked at a dating service, pre-Web, in many capacities. After seeing various listed preferences, matches, relationships, and so on, I began to do my own (am

I worked at a dating service, pre-Web, in many capacities. After seeing various listed preferences, matches, relationships, and so on, I began to do my own (amateur, although I did take rather a lot of statistics for my major) probes of trends. First, you can take away the "online" component. None of the trends OKCupid had blogged about (and were since deleted) were new to my experience. Online magnifies trends through sheer numbers but that's about it. The men tended to make a lot of requests; the women tended not to. Most of the men were rejected; most of the women were not. Women tended to have more "must haves" than men. And so on and so forth. It was all very disheartening. I wonder if I could still pry the data out of dBase III.

roel_v 6y ago
"I wonder if I could still pry the data out of dBase III" Is that in .dbf file format? I happened to be working with that format today, there are 2 up to date and complete libraries for Python for it, it's trivial to get such data out.
- at_a_remove 6y ago
  Well, it is a bit more than that. I would have to remember the field mappings for something that used dBase III as its underpinning for all kinds of preference questions, which I would also have to find the template for. And then there's the decades-old floppies ... In any case, I think the deletion of the old OKCupid statistical analyses is fairly telling: most people do not want to look at the evidence, and ignore it when it goes against the grain. You will get a lot of pushback when you conclude "men have a harder time dating than women" from what you gather. I've often considering "doing something" with it, but it is likely unethical and, in any case, wouldn't really change any minds.