Not-a-particular dataset request – still not kosher?

I think I know the answer to this, but as a naive newbie I posted a plea for "some interesting datasets" for a particular purpose. I see the rule about asking for "a particular dataset", and I suppose that applies to my posting as well. But before I delete my question, I want to hear some opinions. The question is here and I will delete the question if there is a consensus that I should. Thanks.

Update

Based on the discussion here, as well as on the question itself, which has not given a clear "delete" signal so far at least, I intend to leave my question there. I think there is (and has been) a clear methodological context for the question, and it is my hope and belief that its presence will not open the floodgates for the types of data requests we don't desire.

• By the way, welcome to CV! I've noticed you around the past couple days & haven't had a chance to say hi. I linked your site (which I appreciate) in my question here: Free internet or downloadable resources for sample size calculations. If you know of any others, please add them. Jul 29 '14 at 20:33
• Hi back to you. Thanks for linking my site - I will add links there if I know of them. Jul 30 '14 at 13:08
• Russ -- One thing to note here in meta is that the meaning of upvotes is different. Here it doesn't necessarily mean "good question" - if you're asking "is X okay" and you get upvotes, that signals agreement that more people think it's okay than not. Perhaps even moreso with answers - downvoted answers are not bad, it's just their opinions are not agreed with, and upvoted answers are ones some people agree with. I don't suppose you can see the vote breakdowns yet (I think that's at 1000 reputation), but as I type nobody has disagreed with Scortchi's answer... (ctd) Jul 30 '14 at 18:49
• (ctd)... Between all the upvotes on your question and Scortchi's answer, I think the indications so far are that your question is widely regarded as on topic by the people who participate here in meta (which largely determines how it runs -- at least within the broader constraints of how the network-wide rules are applied). There's no indication here that you should delete, at least not for now (and I don't expect that to change). Jul 30 '14 at 18:59

My take on the rule is here: I think we only want to ward off (the many potential) questions about getting data ("statistics") about particular subjects—economics, baseball, or medicine.

I generally believe that questions about datasets should be on-topic on CV (see my answer here: What is a good place for "how/where do I get data on X" questions). Your question fits the category of asking for datasets that I think is acceptable, so I wouldn't mind leaving it open. However, the community seems to clearly disagree with me on this. So I suspect your question is off-topic for CV under the agreed-upon rules.

• Doesn't look like clear disagreement to me...Your answer has no downvotes, 4 upvotes (now including mine), and only 1 comment (upvoted 2× though it is) that seems more ambivalent or reflective than clearly opposed. Similarly, the other answers don't strike me as disagreeing directly. Scortchi's answer on the thread he linked seems indirectly supportive, and @RussLenth's question now has 4 upvotes (again, including mine) and 0 down/close-votes. Glen_b might disagree, but it's not clear to me if he wants these questions to be off-topic or is just trying to represent the community's interests. Jul 30 '14 at 5:11
• @Nick I was simply trying to interpret the explanations of what's on and off topic on the on-topic help page, my personal preferences didn't enter into it (at least not in any conscious sense); I'm interested to read people's thoughts here -- I hope to be convinced one way or another. I lean somewhat toward thinking this question probably should be acceptable here*, but I don't read the words in the help that way. $\quad$ $\quad$ *(though I also want to more clearly have a way to figure out where the grey line should go) Jul 30 '14 at 18:36
• Oh, good point! I hadn't seen anyone refer directly to the help page yet. Seems the relevant quote is, "Questions about obtaining particular datasets are off-topic (they are too specialized)." I suppose we wouldn't want to elaborate that too much, but if more good questions raise confusion about the limits of the rule, maybe inserting a link to this meta-question would be an efficient way to acknowledge reasonable exceptions. Jul 30 '14 at 18:45