Referencing this proposal on Area 51: http://area51.stackexchange.com/proposals/37195 and this post here: https://stats.stackexchange.com/questions/20788/what-is-the-income-distribution-in-the-u-s-in-1-increments

We have at least one mod (mbq) suggesting its a duplicate of CrossValidated (and thus would belong here) and IIRC comments to that effect from CV users and mods on the proposal before they got swept up by an Area 51 mod.

whuber then closed the question because its off topic.

So is it, or isn't it?

  • $\begingroup$ Thanks for starting this. $\endgroup$
    – user88
    Jan 9, 2012 at 11:00
  • 2
    $\begingroup$ Please note that "off-topic" is just one of several reasons for closing that particular question. Another is for lack of research: the answer is easily obtained through simple means such as Google searches. This is of interest because it would also place the question off topic in the "Data" SE proposal, which specifically focuses on "statistical data that is publicly available but is difficult to find with a Google search." IMHO, that proposal (in its current form) won't go anywhere because it is far too broad and unfocused to be sustainable or build a community. $\endgroup$
    – whuber Mod
    Jan 9, 2012 at 14:28
  • $\begingroup$ @whuber: Lack of research isn't a reason for closure that the SE network supports. See "Introduce a “general reference” close reason" and "Should a clear lack of research be grounds for closing a question?" (Don't mind me if you already knew this; I realize your comment is 2.64 years old by now...) $\endgroup$ Jul 31, 2014 at 17:28

1 Answer 1


I thought we'd settled this one a couple of times, but it's a tough restriction to enforce when it isn't listed in the FAQ.

I still think these should be off-topic because of their potential to clog up the feed with questions that are likely to be irrelevant for many of our users. I know we have tags for filtering questions, but I dislike that approach - within the domain of statistics and ML, there are all kinds of things that I don't yet know will interest me. I love being able to open a random question here and learn something new. I don't think I or anyone would get that same kind of surprise learning from data sourcing questions that aren't at least somewhat limited by field.

I think questions like these are best handled by the ever-expanding subject matter StackExchanges - even more so than a Data SE (in my opinion), because that strikes me as too diffuse a topic to really lure the multiple experts for each topic that you'd need for a high-quality community.

  • 3
    $\begingroup$ I agree with these thoughts, including those about a proposed data SE. The wording in the FAQ is carefully chosen--"collecting" data distinguishes it from "finding" data--but of course we can make that distinction stronger at any time. Suggestions are welcome. Note, though, that we explicitly decided that questions about how to go about generating (and, to an extent, managing) your own data are on topic. I probably should add that (a) mods don't often have to close off-topic "where do I find data" questions and (b) if you think that's inconsistently enforced, please flag the violations. $\endgroup$
    – whuber Mod
    Jan 9, 2012 at 23:58
  • 3
    $\begingroup$ @whuber I'm completely in agreement that the generation and management of data are on-topic. I've only seen one or two "where do I find" questions in quite some time, so I don't think it's a pressing matter at all - but I do think a more direct prohibition of "where do I find data" questions would be nice to have. Realistically, I doubt most people check the FAQ before asking - but I'd feel a bit better if we could point authors to that exact question. Overall, though, this hasn't been nearly the problem that I feared it would be when we were starting up. $\endgroup$ Jan 10, 2012 at 0:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .