13

This is something that I'm very guilty of but I'm trying to improve. Somebody asks a question and I answer it to the best of my ability but I don't upvote the question.

If it's worthy of my time to answer it, then it's a good question, and I should be upvotng it, right?

EDIT: 55% of questions have fewer upvotes than highest voted answers. Source: https://stackapps.com/questions/8505/percentage-of-questions-upvoted-lower-than-the-highest-upvoted-answer/8506#8506

This has been discussed before: Vote early, vote often

I hereby resolve to go through my answers and upvote their questions.

After I made the resolution I tried to upvote all the questions I answered but was finally stopped by "Daily vote limit reached; vote again in 10 hours." I would like to suggest that for at least one day, we all spend our quota of votes upvoting questions.

| |
  • I have looked at my own profile and it states that I have not up-voted any questions. I have a vague recollection of having up-voted several. Is the reported statistic correct? – Francis Phillips Nov 24 '19 at 17:36
  • 2
    I've wanted to bring this up for a while, but I kept second-guessing myself. Thank you for bringing it up, because I think this is a big problem on our site! – Richard Nov 24 '19 at 19:40
  • 2
    This community has had a voting issue for a long time now. Current "power" users and most active members rarely vote. No idea why. – Von Huffman Nov 24 '19 at 22:46
  • 3
    Sadly, the only questions that get upvotes are the ones that get featured in the hot networks questions. See all those questions with one upvote? Those are mine. The situation is worse than it looks like. If i didnt vote, most questions would stay at 0. There are questions that have 5 or 6 answers with ZERO upvotes until I vote on them, and it's a VERY common sighting. – Von Huffman Nov 24 '19 at 22:50
  • 1
    @lyd this has been my experince as well. Practically all of my questions have much lower upvotes than the highest ranked answer. Even on questions that have been closed for poor quality. And on this site, once a question is closed, it is never reopened, even after editing. – pro Nov 25 '19 at 5:12
  • 1
    We also have a bad trigger happy issue with public moderation votes, and not enough attention to reopen scenarios. It's not as bad as it used to be, but it's still there. – Von Huffman Nov 25 '19 at 5:15
  • 1
    @lyd it's still pretty bad. I'm the proof of it. – pro Nov 25 '19 at 12:25
  • 1
    Yes, this needs to happen - especially on any questions that one cares to take the time to answer... And, perhaps any question that one can't deem as 'stupid' or 'unworthy,' if one is even taking the time to simply read it... As in, "this question is in some way thought provoking and/or entertaining. It does not make me want to slit my wrists," then it should be upvoted. Especially by the older/more experienced members that might otherwise think a novice-type question to be 'beneath' them – Tim Burnett - Bassist Nov 29 '19 at 13:19
9

I do recommend us voting more as a community, however, up votes and down votes both should be used more. Because voting is tied to quality, we should use it but be aware of what it means for the site.

Up voting good well asked questions on the site will make them stick out and show off what useful questions we have have here. Down voting questions that are confusing, roundabout, or not useful helps us monitor, inform, and refine the question we have and display to the outside world. We can also always steer the OP into a better quality question by explaining our down votes.

Note if questions are not answerable, closing them makes more sense than down voting. We should still answer these questions, but showing quality helps us a lot. There will also always be questions in the middle. In a perfect world, every question would get a vote, but there is grey area. Also note that voting while important is still in the individual's hands and as such there's not a blanket statement we can give with voting on specific questions and answers. We can only really state guidelines.

While we're on it, off topic questions and answers are kind of outside the scope of voting and we should keep that in mind. We should deal with them though other means like closing, migration, or deletion. Down or up voting off topic content doesn't do much for us as a site.


We're also fighting something else when it comes to voting which is the hot network question effect in which the whole network will see questions on the sidebar and come here and vote. When they do, typically they vote more on answers than the questions themselves and do not have the ability to down vote due to them needed 125 rep on the site. This leads to a lot of the lopsided voting on newly asked question that seem to get a ton of votes and views out of nowhere.

| |
9

After thinking about this for a bit, I decided to actually dig up data so we can be informed. There is already a SEDE query we can use to examine site data. The only thing this does not show is up vs down votes, but this gives the idea of what we are actually doing. The raw graph can be seen below at the time the post was written:

weekly Q/A data

The actual number of votes is less, but we're also generating more answers than questions which is generally desired and wanted on SEs. Looking at the ratios between the votes and posts per question and answer get the following:

  • Answer vote/post ratio: 4.39
  • Question vote/post ratio: 6.156

Which seems ok. We do generate about 2 more votes (up or down) per question posted than answer posted. The bigger problem I'd say is we only generate 4 votes per answer and 6 votes per question and our question rate is pretty low only about 51 questions were asked last week.

To give an idea of other site's ratio, here's a sample of several other graduated SEs for comparison using the same query above:

  1. WB SE:

    • Answer vote/post ratio: 7.25
    • Question vote/post ratio: 10.8
  2. CR SE:

    • Answer vote/post ratio: 4.887
    • Question vote/post ratio: 5.069
  3. RPG SE:

    • Answer vote/post ratio: 10.836
    • Question vote/post ratio: 12.190
  4. The Workplace SE:

    • Answer vote/post ratio: 15.673
    • Question vote/post ratio: 22.6

The lowest rate of the sample SEs above is Code Review, but they have much more intense questions than most SEs. I may play around with the data more, but this alone gives us an idea what we need to do which is just in general vote more.

| |
  • 1
    55% of questions have fewer upvotes than highest voted answers. Source: stackapps.com/questions/8505/… – pro Nov 25 '19 at 22:43
  • 1
    @pro just to also point out of all the the examples posted above, only the code review SE has that number is under 50% (37% to be exact) and we're the next lowest. We do have to face that a lot of our votes are from people outside this site and the general pattern is answers get more votes than questions. I do think that a 55% stat for question vs answer is pretty good and I don't think that score alone should be used for what we're talking about especially since one question generates many answers. Remember up and down votes are both important to SEs. – Dom Nov 25 '19 at 23:14
  • 1
    I'm afraid I'm confused. What do you mean that votes come from people outside the site? You can't vote unless you're logged in. – pro Nov 26 '19 at 7:02
  • 2
    "our question rate is pretty low only about 51 questions were asked last week." Given my experience asking questions, even in the last year, that does not surprise me. And I'm in the top 2%. I've had questions closed because they were too broad and questions closed because they were too specific. As a result, the first place I ask questions is often in other forums. music.meta.stackexchange.com/questions/3302/… – pro Nov 26 '19 at 7:05
  • 1
    @pro anyone with the association bonus across the SE can come here and vote and leave without ever posting an answer or question. The HNQ always has at least one of our questions on it and it very much shows on those questions. As for the other meta post, let's keep the conversation on that question and answer. – Dom Nov 26 '19 at 12:48
  • 1
    that kind of problem with the HNQ is a good problem to have. :-) – pro Nov 26 '19 at 19:14
  • 1
    @pro I disagree due to quality. It's very common on those questions to have low quality or out right wrong answers up voted and drown out question and answers that better show what this site has to offer. – Dom Nov 26 '19 at 19:25
  • 1
    I see. Thanks for the insight. – pro Nov 26 '19 at 19:50
  • Huh, a double answer. I always wondered if I'd ever see a good one (two?). Thanks for the insight! – user45266 Nov 27 '19 at 22:35
  • For the top 5 users on Music.SE this year, their lifetime ratios of number of votes cast on questions-to-answers written are: 0.2, 0.2, 0.07, 0.06, 0.8. It increases some after that--continuing with the top users this year, the lifetime question votes-to-answers ratios are: 0.9, 3.8, 0.3, 2.0, 1.3, 0.9, 3.3, 1.5, 0.05, 0.2, 0.04, 0.2, 2.1, 6.1, 0.9. Just another angle on the same question. There are some highly active users who vote heavily on questions, but there are many who very rarely vote on questions. – jdjazz Apr 12 at 16:19
  • I feel like this sort of thing is really common: as of right now, this question has 56 views, 1 upvote, and 4 answers--all from really active users (Michael Curtis, piiperiReinstateMonica, ttw, and ggcg). I like the question and will upvote it, but how doesn't it have more upvotes?? – jdjazz Apr 13 at 20:01
  • 1
    @jdjazz Votes show quality, I don't really personally feel like it's a high quality question as it is quite broad some other may feel differently. I do feel like we need to use votes more up and down especially our core users, but I don't think everyone should just upvote a question they answer. – Dom Apr 13 at 20:05
  • Thanks for the insights! – jdjazz Apr 13 at 20:17
7

I don't know whether I'd agree that a question worth answering is always deserving of an upvote, but I've noticed the same thing you have about questions having lower scores than their answers.

Sometimes, I've noticed that there are questions that attract a bunch of downvotes but then get a good quality answer or two. Ultimately, upvoting is a personal decision, and I think every user on this site has their own unique criteria for what they'll vote on. For example, I vote up or down on nearly every post I come across, and some people only vote when something's really good or really bad.

A lot of our questions come from new users, too, and especially with questions that are really simple to answer, often the answers will be of much higher quality than the question itself.

And I know you have access to vote breakdown info on each post, but questions tend to attract downvotes more often than answers, since downvoting a question does not decrease one's own reputation. This could potentially be adding to the observed effect of questions having lower scores.

So I think at least some of that disparity in post scores is normal. But that certainly doesn't explain every scenario, and I'd be interested to hear the opinion of the mods on this one.

And yes, before you ask, I did upvote this meta question. :)

| |
  • 1
    Whilst folk from other sites come visiting on the strength of the questions (usually posted on the right hand side), I can't decide whether they actually really come for the questions, or those answers. Also, when I answer some questions, I think they're good as they are about something confusing, so up vote them, but other times, I may still answer even though I consider the question facile, and those I won't upvote. Not answered and dv'd yet though..! – Tim Dec 3 '19 at 16:34
  • @Tim I've answered and dv'ed on a few meta sites, but I can't remember the last time I did that on a main site. Maybe ELL, b/c sometimes the question is so poorly written that I end up dv'ing, but I still answer if I know what's going on in the post. – user45266 Dec 3 '19 at 18:19
4

I think that's a very valuable consideration. I wouldn't say that it's always one-to-one. Sometimes a question may be unclear, and you answer on the off chance that it addresses what the OP was going for. But generally, I think you're right. Our questions often have much lower scores than their answers.

BTW, at 10k you will be able to get into the site tools through the review page.

| |
  • 1
    For historical reasons, it is unlikely that I will reach 10K any time soon. And from what I recall of the moderator tools that I had access to when the site was in beta, there was no way to answer my question as a humble 10K-er. I think it requires diamond moderator tools. But you're over 10K, perhaps you can prove me wrong? – pro Nov 25 '19 at 13:10
  • Yes, I suppose you're right. The site tools show questions and answers with high upvotes or high downvotes, but not detailed statistics. – luser droog Nov 25 '19 at 16:26
  • 1
    There might be something on stackapps.com to get more detailed info. – luser droog Nov 25 '19 at 16:31
  • 1
    There's always data.stackexchange.com. Most likely someone already made a job to do this. – Dom Nov 25 '19 at 19:13
3

There are times when I am annoyed enough at some question that a downvote+comment are insufficiently satisfying! So feel compelled to "answer".

Admittedly it has never happened for me on music.SE and rarely elsewhere. Here is a recent example.

That said I agree with this suggestion and try to make it habit to upvote the questions I answer.

| |
-2

What you're basically proposing is to have a system in place that automatically upvotes the questions you answer.

Since there's a limit on upvotes, you would then impose a limit on answers.

If you're comparing different SE sites, and wondering why there's a big difference, well, it's because they are different. We are the one manipulating those numbers, depending on what we prefer to discuss.

The data you're getting should be "weighted" further to accommodate the fact that some subjects are more quantifiable than others.

| |
  • The limit on votes is very generous. – Von Huffman Nov 30 '19 at 11:52
  • @Lyd my point is if you're going to upvote every question you answer, you are limiting # of answers by definition. – Hatebit Dec 1 '19 at 14:00

You must log in to answer this question.

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