From https://serverfault.com/election:

2011: 3,834 voters were eligible, 566 visited the election, and 310 voted (8% of eligible users voted)

2012: 6,350 voters were eligible, 1,361 visited the site during the election, 474 visited the election page, and 208 voted (3,2% of eligible users voted)

So - we have more users, and more people visiting the election page, but less votes - even with nearly twice the userbase. What's going on?

  • That may only be counting the "election round" - which was mostly during the weekend. That may skew things. Jan 24, 2012 at 20:22
  • 2
    Clearly, the banners at the top of the page weren't in your face enough. Jan 24, 2012 at 20:25
  • From SO - Early 2011: 11,3% participation, Late 2011: 6,8% participation
    – pauska
    Jan 24, 2012 at 20:29
  • I noticed this too... Jan 24, 2012 at 20:36
  • @Holocryptic i die a little inside when those system messages are used for anything but the most important messages "We are going to break something, sorry" type stuff
    – Zypher
    Jan 24, 2012 at 20:38
  • One might wonder if we have annoyed some people enough as described here (meta.serverfault.com/questions/2737/…). Maybe being overly specific about questions results in us killing the potential for a larger community.
    – Zoredache
    Jan 24, 2012 at 20:43
  • 1
    @Zoredache I'm not sure that really applies here. This year, over double the people visited the election page, but 1/3 less voted. It doesn't seem that growth is the issue.
    – MDMarra
    Jan 24, 2012 at 20:50

6 Answers 6


I think this is an unfortunate byproduct of a growing community. Politically speaking things usually divide up into a few camps:

  • The Very Involved
    People who run for public office or make a lot of noise trying to get things changed.
    On SF this would be folks who run for mod positions, post on meta, flag actively, etc. These people definitely vote.

  • The Somewhat Involved
    People who stay generally informed about political affairs.
    On SF this would be the folks who read meta, flag occasionally, throw around close votes when they have 'em... These people probably vote.

  • The Generally Uninvolved
    People who might be bothered to turn up to a presidential election.
    On SF this would be the folks who skim the candidate bios and vote out of a sense of obligation.

  • The Apathetic Masses
    People who don't care how the "government" works as long as the streets are plowed.
    This is pretty much "everyone else". They typically won't vote.

  • The Disillusioned
    People who think "Meh, my vote doesn't count".
    Given the low turnout of the SF election I would hope there aren't many of these people.

  • Note that this begs the real question: How do we drag people out of "The Apathetic Masses" and get them to be at least "Somewhat Involved"
    – voretaq7
    Jan 24, 2012 at 20:31
  • Do you really want those who don't care about the voting to cast votes? What sort of results might that produce? Over here voting in political elections (all levels) is compulsory and the outcome generally reflects the negative attitude of the majority. Not a good thing. Jan 25, 2012 at 3:59
  • 1
    @JohnGardeniers I'd rather they become interested and involved in the process. If they're just showing up to pull the proverbial lever I would much prefer they stay home: Compulsory voting with a disinterested electorate is an awful idea.
    – voretaq7
    Jan 25, 2012 at 4:28

I think you need to remember that the user base each year will include an increasing number of inactive users. Those who posted a single Q or A right at the start and haven't been back since are still part of the user base. This means the numbers will always appear to deteriorate.

I suspect that if we were to count only those who have been active within the last few months the figures would look very different.

  • Interesting approach. Are there any numbers about that?
    – Nils
    Jan 24, 2012 at 21:48
  • @Nils, I don't have any numbers but just looking at the posts, or even the user list, show that the vast number of people come to the site, post one or a few questions and/or answers and then go away. Those who have remained active, even as little as myself, for extended periods are a minority. Jan 24, 2012 at 23:13
  • I can sort of agree with this -- The reputation barrier to entry to be considered an "eligible voter" is incredibly low (150). If we consider only the users above 1000 rep (using the numbers from the Reputation Leagues) our voter turnout numbers are much better: 2/3 visited the election page, and around half actually voted.
    – voretaq7
    Jan 25, 2012 at 4:35
  • There are more people eligible to vote (1.65x as many as last year) but the number of eligible people who visited the site during the election period was up by a larger amount (2.4x as many). So it's not that the people who could've voted didn't visit SF, but they didn't bother going to the election page. Jan 25, 2012 at 5:20

So - we have more users, and more people visiting the election page, but less votes - even with nearly twice the userbase. What's going on?

The problem seems to be that we didn't get more people visiting the election page: 474 this year vs. 566 last year.

Maybe the banner coming and going had something to do with it? It was always there on meta.SF, but on SF itself sometimes there was an election banner, sometimes there wasn't.

Or maybe people are spread too thin with multiple SE sites? I think most people who would bother to vote at all are the type who are conscientious enough to only vote if they really felt they actually knew the candidates, the site.

  • I did have an election banner every day of the election (though I mentally tuned it out - it could probably disappear and I'd never notice :-x). I agree with the dilution idea though: A lot of what was once concentrated here is now on U&L, AskUbuntu, DBA, & Security -- might be reducing the number of people who feel "involved enough" to vote.
    – voretaq7
    Jan 24, 2012 at 20:34
  • 2
    It was definitely on and off for me. Sometimes it would disappear for a few hours and come back. Very annoying. Jan 24, 2012 at 21:40

Don't forget personality:

  • People didn't know or like the candidates.
  • I doubt this would be a significant factor. As the candidates are all active members of this community the "didn't know" part is less likely and the "like" part is essentially covered by the primary voting, where you could downvote as well as upvote. Jan 24, 2012 at 23:16
  • You could try factoring in how the candidates did compared to their participation on the chat server for Serverfault and then reference it to voting...although I doubt enough of the people are in chat that it would sway voting much. Jan 24, 2012 at 23:52
  • Could you downvote? That's interesting, it wasn't immediately obvious to me. I thought I clicked people, and their badges came up on the right hand side. Potato, potaaahto.
    – TristanK
    Jan 25, 2012 at 0:23
  • That was in the final phase. In the primaries the voting was the same as it is for questions. Maybe you missed that stage. Jan 25, 2012 at 4:02
  • So there were three stages? Self-nomination (I saw), primaries (?), then actual voting (I did)? Ah well. Maybe it's not that the people weren't liked, maybe it was the commitment requirement!
    – TristanK
    Jan 25, 2012 at 4:42
  • Dug up the relevant threads on the process... See MDMarra's answer below :)
    – TristanK
    Jan 25, 2012 at 4:49

I don't think this is unexpected. Just because we have a lot of users, doesn't mean they are active users and it doesn't mean they have to give a f*** about who gets to be a mod.

  • 1
    But, if you do the math (which I didn't bother), you'll see that there was a MUCH higher percentage of voters in 2011 versus 2012.
    – pauska
    Jan 24, 2012 at 20:21

People be reel stoopid. Obviously.

You must log in to answer this question.

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