33

I wanted to blog/tweet our list of unanswered questions, which appear to be sorted by vote. However I noticed that given the amount of users on this site that the amount of voting on the questions isn't so hot.

Is this how it's supposed to be or are we still too small to expect more voting? There's a ton of good questions on the front page with only 1 vote. I try to vote on answers and questions every day to get the quality stuff on top, this is what I'm supposed to be doing right? :)

Update: Here's a blog post on the importance of voting, and some new badges to boot!

0

4 Answers 4

32

Voting should be encouraged but voting for the sake of voting should be dissuaded.

Voting on answers is pretty self explanatory

It's easy to notice a good answer amongst less-good answers even if it's competing with your own answer.

I know it's very easy for the person with the most points on the site to say "be honest with your votes" but I mean it. We're all here for the same reason: to help people. Not voting the best answers up stops the site working like it's supposed to and that hurts its whole efficiency.

But what makes a good question?

  • Is it something that affects us?
  • Something we think may affect lots of other users?
  • Something that doesn't have enough attention?
  • Something you don't understand (for technical, not semantic reasons)?
  • Something you can answer?

I personally vote for the first four. I don't think the fifth is enough on its own especially if it's done for the reason of getting the question asker over the points threshold so they can give your answer a +1.

Ultimately this isn't something we can enforce. People will do what they want, when they want. But yes, including something in the FAQ might get people to vote more freely.

5
  • 6
    I've voted this just to be more meta
    – tutuca
    Commented Sep 26, 2010 at 2:01
  • 1
    i'd vote you down if I could for confusing effect with affect :D j/k
    – tamale
    Commented Oct 21, 2010 at 14:38
  • @tamale Good call. Fixed.
    – Oli Mod
    Commented Oct 21, 2010 at 16:03
  • 1
    fix the second one too ;)
    – kounryusui
    Commented Oct 25, 2010 at 21:21
  • 1
    I don't like point 2 (Something that doesn't have enough attention?). It's kind of dishonest. If you really care that much about the Question receiving attention., why not put a Bounty for it?
    – tshepang
    Commented Dec 12, 2010 at 20:22
18

Yes - voting is a very very important act on this site. It's what drives the content - and it needs to be stressed - especially as the majority of this communities user base comes from outside the existent StackExchange community. The system of this style isn't one that's frequently used (Think of it as most users here are coming from the blog/IRC/Forum world).

Voting on great questions and even greater answers helps to drive the quality content that we strive for on this site. Almost equally as important is the downvote power which contributes to what is good and what is considered not good by this community - a community which should round out as Experts in the field of Ubuntu.

In addition to voting - the community here needs to start taking user moderation (that is the powers given to user for closing, editing, reopening, etc) a lot more seriously. It's hard to just type a great question or a great answer at first shot. But having feedback from users and having users directly edit questions and answers to contribute and even better answer or question further helps establish ourselves as a community of experts providing phenomenal answers to compelling questions.

3
  • 1
    Should we add this to the FAQ? Also, maybe a "vote!" link or something on the top banner of the parent site linking to this so people realize it's important? Commented Sep 19, 2010 at 1:15
  • 1
    Also, perhaps talking about the importance of voting down as well? Commented Sep 19, 2010 at 1:16
  • 1
    We only have the ability to edit the top most portion of the FAQ (What questions are acceptable)
    – Marco Ceppi Mod
    Commented Sep 19, 2010 at 1:55
6

Consider the following data - feel free to play around with the SQL queries yourself (Links at the bottom)

Average number of ______ by site

                    upvotes/question    upvotes/user        answers/question        answers/user
Unix&Linux          8.71                5.19                3.05                    1.81
Stackoverflow       6.38                12.76               3.17                    6.34
AskUbuntu           6.34                3.93                2.61                    1.62
Linguistics         8.74                4.74                3.25                    1.76
Superuser           6.45                4.47                2.98                    2.06

My reading of the story is not upvotes in general, more like upvotes in the context of the overall userbase. Fewer of us answer fewer questions than any of the other 4 communities I picked on a whim. Not only that, but we seem to reward questions less than the other communities. If I was to make a rash judgement, I'd say we prefer to just harvest what the community has to offer but feel less inclined to contribute and to reward those who contributed.

https://data.stackexchange.com/superuser/query/614178/answers-per-user https://data.stackexchange.com/superuser/query/614175/answers-per-question https://data.stackexchange.com/superuser/query/614068/upvotes-per-user https://data.stackexchange.com/superuser/query/613862/number-of-upvotes-per-post

PS: I am tempted to make a few more queries - if you have one in mind, let us know in the comments and I'll see to it that I find the time to run the best ones as an update to this.

Update 20170113 This is what voter ranges look like on AskUbuntu:

    The number of users in the entire community:       404632
+-------------------------------------------------------------------------------------------+
  Number of users who never voted:                         328852       81.27%
  Number of users who upvoted between   0   -   5:          45342       11.21%
  Number of users who upvoted between   5   -   10:         10825        2.68%
  Number of users who upvoted between   10  -   20:          8255        2.04%
  Number of users who upvoted between   20  -   40:          5453        1.35%
  Number of users who upvoted between   40  -   80:          3148        0.78%
  Number of users who upvoted between   80  -   160:         1535        0.38%
  Number of users who upvoted between   160 -   320:          635        0.16%
  Number of users who upvoted between   320 -   640:          322        0.08%
  Number of users who voted more:                             113        0.03%

And this is what Stack Overflow looks like close up and personal:

    The number of users in the entire community:      6539900
+-------------------------------------------------------------------------------------------+
  Number of users who never voted:                        5594471       85.54%
  Number of users who upvoted between   0   -   5:         331193        5.06%
  Number of users who upvoted between   5   -   10:        126725        1.94%
  Number of users who upvoted between   10  -   20:        125488        1.92%
  Number of users who upvoted between   20  -   40:        109279        1.67%
  Number of users who upvoted between   40  -   80:         87916        1.34%
  Number of users who upvoted between   80  -   160:        65310        1.00%
  Number of users who upvoted between   160 -   320:        43647        0.67%
  Number of users who upvoted between   320 -   640:        28872        0.44%
  Number of users who voted more:                           10490        0.16%

You can experiment with the numbers at: https://data.stackexchange.com/askubuntu/query/615470/upvotes-per-user-ranges

The query is pretty easy to tune to your liking.

4
  • It would be nice to get some idea of what the distributions of these variables look like.
    – edwinksl
    Commented Jan 12, 2017 at 19:52
  • Can you give me an example for the user and the question "range"? Or you mean like users with no votes vs users with 1 to 10 type of data? Commented Jan 12, 2017 at 20:04
  • I mean "users with no votes vs users with 1 to 10 type of data". We may want to not bin the data to see how it looks like first before binning them.
    – edwinksl
    Commented Jan 12, 2017 at 20:38
  • I ran a quick errand to see what serial voters look like and updated the answer. You can also check more communities at data.stackexchange.com/askubuntu/query/615470/… Commented Jan 13, 2017 at 14:13
3

Yes. IMHO, even if one is not personally interested in a particular question (if he is, he will usually not only vote, but also star it), he should vote it if he finds it interesting and/or probable to be practically useful.

I think for each n votes one should to be given a reputation gain. The only issue here is to resist reputation-increasing mass voting robots in this case, but I think it is not a hard problem.

Also I think voting button should be available on questions listing pages, so one can vote without browsing to question details.

You must log in to answer this question.

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