14
$\begingroup$

I've been discovering DataScienceSE for about two months and although I kind of like the topic, my impression is that this SE is not doing great. I've been active on a few other SE sites for some time and one can really tell the difference. Apparently there's just not enough regular active members for the flow of questions (68% answered). Crucially there are very few votes being cast, and this is quite bad given that the SE model relies on votes.

To illustrate my point: I gave 50 answers in the past two months, half of which have zero votes. In average I collected probably around 0.5 votes per answer (not counting OP's choice; without any negative vote as far as I know). I don't mind as long as I enjoy giving some insight here and there, but with very little feedback from the community it's not very encouraging and I guess I'll eventually get bored with the site.

So what I'd like to know is:

  • Why so few people vote? Is it because there are not enough "avid members", too much turn-over, too many questions, questions being too specialized (or not enough)... ?
  • What can be done about it? (if anything) More precisely, as a modest active member, what should I prioritize? For instance I've seen other sites which encourage:
    • Vote early and vote often: should I vote on questions/answers even if I don't feel very confident about the topic?
    • voting to close out-of-scope/unclear questions early (and harshly) in order to force the user to clarify before (possibly) re-opening. But of course that's not very welcoming.
    • be more strict about duplicates: many beginners questions are variants of existing questions, should I vote to close as duplicate even if it's not exactly the same question?
    • any other suggestion...?
$\endgroup$
12
$\begingroup$

Let's start with some basic context. If you hover over the voting buttons, and you will note they say:

Upvote:

This question shows research effort; it is useful and clear

Downvote:

This question does not show any research effort; it is unclear or not useful


With that context in place, let's move on to your questions:

So what I'd like to know is:

Why do so few people vote?

I wish I knew the answer to this. I have a general rule, that if I answer a question, I will upvote it. Why are you answering a question if you do not think it is helpful? In addition, if someone took the time to answer my question, and the answer was worthwhile, why not upvote it?

FYI: I previously made some similar points here.

What should I prioritize?

You should vote. Up and down. The stack exchange sites are organized around this. As an example, the roomba process make some decsions based on votes on the post:

Why is the script for deletion called Roomba?

Enable automatic deletion of old, unanswered, zero-score questions after a year?

Turbocharging the Roomba: solutions for premature deletion

Voting to close out-of-scope/unclear questions early (and harshly) in order to force the user to clarify before (possibly) re-opening. But of course that's not very welcoming.

Why do you say "of course" not very welcoming? Why do you feel welcoming is a key here? Why should anyone be welcome to put up anything they want, and be welcomed for it? Why should someone be able to "expect" that experts will give them time freely and yet not expect those asking for the experts time, to construct good questions?

Be more strict about duplicates: many beginners questions are variants of existing questions, should I vote to close as duplicate even if it's not exactly the same question?

Questions are duplicates if the answer for one question would work for the other. So yes, if it is a duplicate, vote to close.

$\endgroup$
  • 1
    $\begingroup$ Thank you Stephen, that helps clarifying things a bit. As a newcomer to the site I'm often hesitant to vote down questions when I'm not sure if they are in-scope or when they are in the grey area in terms of usefulness/clarity. Knowing that very few people vote makes me even more reluctant to vote when I have a doubt, since an error of judgement would have more impact. In borderline cases, would you say that it's better to be bold and vote up/down to the best of my knowledge or leave it to more knowledgeable people? $\endgroup$ – Erwan Jul 2 at 9:36
7
$\begingroup$

I guess we can only speculate about the reasons. This site is more "practical" than crossvalidation. So questions necessarily are relevant to a smaller audience. And we have a lot of "beginner" questions that sometimes feel repetitive.

One simple, practical thing I'd recommend is this: If you go through the "First Posts" (and we have a lot of first posts) and "Late Answers" review queue and you see a good/on-topic question/answer instead of simply clicking "No Action Needed" upvote and click "I'm done". Just as quick.

Another is making edits to questions with uninformative titles (like "quick question about classification" or whatever) so people know what it is about and attract more traffic.

$\endgroup$
  • $\begingroup$ Thanks for the nice tips! $\endgroup$ – Erwan Jul 3 at 0:12
2
$\begingroup$

Why so few people vote? Is it because there are not enough "avid members", too much turn-over, too many questions, questions being too specialized (or not enough)... ?

I think it has more to do with the questioners. I haven't been here long, but I suspect many people make an account to ask one question related to a school project or something. When they don't get a quick answer or even if they do, they never return or bother to upvote/accept. (This is why I don't usually answer calculus homework questions on the math SE). In other words, they're not interested in contributing (voting, asking, answering). For people interested in contributing, this may be frustrating and drive them away. All SE sites have many such people, but the ratio must be small for anything to work, I think.

Another answerer mentioned practicality, but I would have expected that to actually be helpful in terms of motivating people to participate, though many such problems can be quite localized, which may be what he/she meant.

What can be done about it? (if anything)

For sure, more voting would be nice (at least for me, I'll probably give up contributing without that improving). It's ok if questions are basic I think (I have basic questions too), but at least they should be interesting and well-written (grammatically, syntactically, coherence-wise, and semantically), which often isn't happening. I am not generally "harsh" on questioners and want to be helpful to beginners, but if they can't even take the time to formulate a well-written complete question with context and details in proper English, then why should anyone take the time to answer it? Sometimes I answer these anyway if the underlying question is ok, and don't even get a +1 or any response hahaha. Then again maybe my answers aren't good.

I think this site is quite different than cross-validated (which I expect to be more theoretical), but I actually don't really know the difference between this site and the AI SE. Maybe coordinating to better define the difference would help, since the overlap seems substantial.

I don't mind as long as I enjoy giving some insight here and there, but with very little feedback from the community it's not very encouraging and I guess I'll eventually get bored with the site.

I have rather concurring sentiments.

$\endgroup$

You must log in to answer this question.

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