> They found that an individual’s reputation increases with age, at least into a user’s 40s.
I really hope that's some bad paraphrasing, because I take that to mean for user X at time T, rep(X, T1) <= rep(X, T2). Unless you're very actively malicious (for which you'll get banned) your reputation should never go down by a meaningful amount over time.
Honestly I think it's more poor interpretation than bad paraphrasing. They were sampling reputation and correlating with age, not tracking reputation scores of single users with time. So a more precise, if not clearer statement would be "They found that median reputation increases with poster age at least into a user's 40's". Better?
Whether this constitutes good science or not is not something I'll speak to, having only skimmed it. But certainly stackoverflow.com looks like good data for this kind of research to me. I'd hope they'd have done things like control for different fields, different activity times for the accounts, etc... But I don't know.
> They found that an individual’s reputation increases with age, at least into a user’s 40s.
I really hope that's some bad paraphrasing, because I take that to mean for user X at time T, rep(X, T1) <= rep(X, T2). Unless you're very actively malicious (for which you'll get banned) your reputation should never go down by a meaningful amount over time.