21

I have noticed a trend that questions (even good ones) that have multiple answers are not being voted on.

When a new question comes in it's on the front page for less than a day (around 16-20 hours) unless it has votes then it's pretty much forgotten. The problem is much worse on SO where front page time is less than an hour less than 12 minutes.

If a question has votes it will stay towards the top and will generate more answers and get more views. I'm curious what others criteria is for clicking the up vote on a question?

If it is worthy of you take the time to answer it shouldn't also be worthy of a vote?

Out of our 5,550 questions only 41% have at least 1 vote which leaves around 3,000 with 0 votes and a few hundred with negative votes.

When we were in Beta our questions got a lot more votes and the good ones stayed on the front page for much longer.

I think our community as a whole doesn't vote as much as some of the other SE sites. For instance on scifi I asked a question that I thought they might close and it got 5 or 6 votes the first few hours and the top answer got 16 votes. When was the last time an answer got 16 votes here?

Is it the quality of our Q&A's or that we are really stingy with our votes and only reward them for really good stuff?

5
  • See also: Why did the old questions get more upvotes?
    – fuxia Mod
    Apr 28, 2011 at 21:15
  • Did you ask this question after listening to the second SE podcast? Joel talks about the same issue: not enough people vote on questions, which is a shame, because you don't have to be able to answer a question to judge whether it is a well asked question.
    – Jan Fabry
    Apr 29, 2011 at 8:20
  • I listened to the podcast then really started noticing it.
    – Chris_O
    Apr 29, 2011 at 8:33
  • Colour me guilty.... First up, and I may shame myself here, where are the Q&As? Also, I didn't really know what the standards were for the Q voting. I'll have to get more active..... Apr 29, 2011 at 17:37
  • @jan see blog.stackoverflow.com/2011/05/… May 6, 2011 at 9:12

3 Answers 3

9

This is not a new concern.

Why aren't people voting for questions?

Note date on question -- originally asked August 5, 2008. I tend to agree with the top voted answer by Orion Edwards:

As I see it, the 'workflow' of reading a question/answer goes like this:

  • User opens page

  • User reads question.

    Unless the question is abnormally good or bad, or otherwise provocative, this isn't likely to elicit any emotional response. It's just a question, carry on.

  • User scrolls down and begins reading answers

    As there are many answers, and good answers are rewarded by being 'accepted' and also with increased reputation, this puts the user in the mindset of 'make the answers better'

    The emotional response behind having your answer accepted or upvoted is "I know stuff, I'm smart, I feel good." Likewise, conferring that reward on someone else is quite a powerful thing too. This provides a very strong motivation to rank and provide answers.

  • Because of this motivation, people will put a lot of effort into writing answers (like me with this diatribe) and ranking them.

This works very well for providing and filtering good answers, but there's no such motivation behind voting for questions. For most questions, the strongest response they are likely to elicit is "I have that problem too", which while it's strong, is only going to apply to a small portion of the viewers/answerers.

See my blog response with details of the changes we made:

http://blog.stackoverflow.com/2011/05/vote-for-this-question-or-the-kitten-gets-it/

3

One thing that strikes me is, why do questions even have votes? Questions are already "rated" by the answers it gets, so give the question an amount of points that consists of the sum of the points given to the answers, or some other fancy algorithm.

For handling bad questions, stick a "this question annoys me"-button next to them instead of a down-arrow.

BTW, I did upvote this question, but it didn't feel very good.

3
  • That's a good point. Voting for questions seems unnecessary if questions are solely rated by the value of their answers. The question is, if a question doesn't have any good answers, is it still a good question?
    – Dan H
    May 6, 2011 at 14:41
  • 1
    this also implies questions like "What's your favorite wordpress cartoon" would become the most valued questions on the planet.. lots of funny answers beloved by many and upvoted strongly. Bad questions also sometimes produce great answers, that does not mean the question was a good one. May 6, 2011 at 22:07
  • @Jeff, is that so different from how it works today? Look at stackoverflow.com/questions?sort=votes , the two top rated questions are subjective. May 11, 2011 at 8:48
2

I would think that questions are most accurately rated by how many people click the link to view the question, as that is the greatest indicator of their interest in reading more about the question itself and discovering possible answers.

The idea that a good question is determined by the number of answers completely ignores the audience - it shows how many people are willing and capable of answering the question, not how many people are interested in knowing the answer. That metric is best represented by viewers of the question, as there could be an enormous amount of views on a question that has no answers at all (yet).

I know the entire SE system is built around voting, but could viewing stats (clickthroughs) possibly be considered as well? Instead of changing the underlying architecture that sorts items by votes, maybe an additional function could be added that increases the "votes" on a question automatically in response to views. These "auto-votes" would be weighted less than "intentional-votes", maybe 1 up-vote for every 10 views.

I would think such functionality could be easily added to the existing SE architecture, as votes would still be the primary weighting mechanism - we'd just be extending the methods for votes to be recorded.

The system of rating items purely by manual user votes might work on reddit or digg, where users are emotionally expressing their response to a random meme (and thus more likely to vote/comment more frequently), but in the Q&A paradigm, measuring a question's value only by manual user votes, or by the number of answers (like comment karma), we are missing entirely the value of the question to those with interest in the question's answer.

As previous posts here indicate, this is a common problem in Q&A sites precisely because users don't intuitively think to vote up the questions they're interested in (but aren't able to answer). But those users do intuitively click on the links and read the questions, especially if there are answers, but even if there aren't. They'll vote up the answer that solves their problem because they have a positive emotional response, but it simply doesn't occur to them to scroll back up and "thank" the original poster of the question...

1
  • Views already are a part of the system: they are displayed prominently on the overview pages, and there are some badges for questions that get a very high number of views. I don't think we need to add an extra system that adds reputation based on views. This would be needed if the viewing pattern would be very different from the voting pattern, but if you look at the questions ordered by votes, you see that they are also the popular questions
    – Jan Fabry
    May 16, 2011 at 13:58

You must log in to answer this question.

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