Hacker News Reader

> Makes me sorta want to see a breakdown by sex. aka "male software developer has a 6% chance of marrying a software developer

But this data is specifically "among married couples". The male software developer's chances of marrying a software developer are much, much higher if we just assume he's married in the first place.

View on HN

I don't get it. Wouldn't anyone's chances of marrying any group be higher if we assume they are married at all? Because "being married" is a super set of "being married to a person from X group"?

Because "being married" is a smaller superset of "being married to a software developer" than "being alive" is.

It's the same reason there's a higher rate of being married to software developers among people who are married to software developers than there is among the general population.

I think an extreme example might help: if there are 100 software developers but 98 are single, and 2 are married to each other, then 100% of software developers(that are married couples) have married software developers, even if only 2% of all software developers are married at all.

Reply on news.ycombinator.com