What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 134 Stack Exchange communities.

Question: How can users be motivated to vote more?

Please do not consider the following a ‘personal’ belief – but rather an observation of human nature (likely nothing revolutionary at that).

My understanding of the design of this site is that it is predicated upon the notion that a game like design fosters participation. One of the attributes of a game is keeping score – which this site accomplishes through reputation. While one may play the game for ‘fun’, if you are not trying to ‘win’ you really aren’t into the game.

This leads to an apparent conflict between two ideas which appear fundamental this site:

  1. Voting has no cost to the voter, but is necessary for the community as it ‘grades’ answers.
  2. Voting (in a game scenario) rewards (advances) your ‘opponent’ with no gain to yourself.

Arguably, with respect to the voter, the negative second point outweighs the neutral first point – implying that there is an overall resistance to voting.

In biology, true altruism is a non-existent concept – organisms do not tend to cater to greater good, unless it has some benefit to the individual.

For instance, bats that hunt for blood are not always successful. In the event of a bat returning from an excursion without finding food, another bat will often regurgitate some of its meal to feed the unsuccessful hunter. At first glance this seems like a selfless act – doing with less to help one less fortunate. The reality is that bats expect this treatment reciprocated.

  • A parallel to the above scenario might entail the following: a new user (1 rep) asks a question – they do not have sufficient points to up vote an answer. An answerer up votes the question, with the (implicit) expectation that the person asking the question will up vote the answer. While this may be an edge case, it illustrates a case where voting may be directly beneficial to the voter.

Consider another scenario: corporations donating money to charity. Certainly they all do so from the goodness of their hearts – but they do get a significant tax break in return.

  • The only times that points are ‘lost’ on the site is when a user hits their daily reputation cap, when they receive a down vote, or when they offer a bounty. A system where casting one up vote allows a user to ‘not lose one point’ may provide some (minimal) incentive for voting, without actually giving points for the vote itself (or alternatively may increase the use of bounties if the poster has accrued sufficient points to offset their value).

Voter turnout for elections is typically between 40-55% - for something that actually has an impact on the individuals in question.

  • On this site, voting (or not) has no actual effect on the voter (badges are nice and all – but if my place of work decided to pay me in stickers I might not be so thrilled), so why should there be an expectation that people will vote?

People are always willing to do something for the possibility of a gain – filling out a survey to be entered in a draw, disclosing spending habits in return for a few points towards future purchases, etc.

  • Perhaps employing the same kind of model may have some merit. The gain however, must be indirect – so that people do not vote without merit simply to earn the reward. Something to the effect of ‘if you cast more than 20 votes per week (at least 2 per day), you will get a 1% bonus on the points you earn’.

A few points about the above:

  • 20 votes/week is a low - asking for 40 votes in a day will result in the clause either being mostly ignored or voting without merit.
  • The requirement is prolonged participation (i.e. over a week, not just a single day)- the objective being to foster a certain type of behaviour over a longer period of time.
  • No points are directly awarded for a vote – you still have to earn the points – this is just a ‘bonus’, and 1% is fairly small.
  • On the downside, it provides little incentive to exceed the minimum number of votes (20/wk in this example) (although, if most people just attained that, it would be a marked improvement).
    • Alternatively, a formula where each subsequent vote has a smaller impact on the bonus gained may be preferable
      • For example: b(v)=(4/(1+2.5-0.04v))-2,
        • Where b(v) is the bonus as a function of the number of votes cast for the week
      • 30 votes/wk gets a 1% bonus
      • 100 votes/wk gets a 1.9% bonus
      • There is an upper limit of 2% for the bonus

To motivate people to vote, I believe that there must be a tangible benefit to them – and unfortunately, badges do not appear to be perceived as such a benefit. An effective solution must affect people’s points – without directly giving points for a vote.

share|improve this question

migrated from meta.serverfault.com Jan 8 '12 at 20:19

This question came from our discussion, support, and feature requests site for system and network administrators.

+1 for regurgitating bats. –  jadarnel27 Jan 9 '12 at 4:22
Interesting that you classify badges as '[having] no actual effect', but reputation as a 'tangible benefit'. –  AakashM Jan 9 '12 at 9:05
@AakashM: a) I think that rep, being incremental and more quantitative provides an immediate and continual measure of progress - while badges are more akin to noting specific achievements. (If you will, a high numerical mark on a test vs a sticker or 'Good job' - both are nice, but I'll take the 95% with no comment vs an 80% with a 'Good job'). b) SF has 740 users with over 1k rep, but only 60 who have earned the 'vox populi' (40 votes in a day) badge. c) I think some of the answers might have convinced me that rep alone is unlikely to be a sufficient stimulus to bring about a change in voting –  cyberx86 Jan 9 '12 at 12:11

4 Answers 4

up vote 10 down vote accepted

Voting (in a game scenario) rewards (advances) your ‘opponent’ with no gain to yourself.

I think you might have a flaw in the reason people don't vote.

I have no opponent here. I'm not in it to "win" and I think you'll find none of the high rep users are here to "win".

We're here to help others, so that we can get help when we need it too, by fostering a smart, intelligent, helpful community.

Yes, there's a strong element of gaming to the SE network, and in fact Jeff is constantly talking about how the gaming aspect of the site is the way it encourages users to stay active. But I don't believe that the gaming is what keeps people here.

I believe that what keeps people here is the fact that they get great answers to their rubbish questions, and occasionally they'll get a great answer to a great question.

However, I do agree that not enough people vote on Server Fault. I think I'm the 4th heaviest voter and I've (only) voted 4,345 times in almost 3 years, and I actually don't mind your idea of increasing the value of an upvote.

Because of this, I'm going to ship this over to meta.stackoverflow where it will get a lot more eyeballs and feedback.

share|improve this answer
Thanks for the response. (I realize I come off as a bit self-centered caring only about rep, hopefully that isn't my true nature.) Perhaps the word 'win' wasn't quite right, but I do expect that all users want to earn (not just get) rep, and most do track their progress and see how they stack up against each other - friendly competition isn't a bad thing. I think the appeal of the site comes down to learning - by asking questions, reading answers, and also by answering questions - few places offer such an opportunity. I'd suggest there are 2separate aspects to the site - knowledge and rep. –  cyberx86 Jan 9 '12 at 6:05
When I answer a question I don't actually care about the reputation it may earn me - it is the simple enjoyment of solving a puzzle, possibly learning something new in the process, and helping someone out. After I have answered though, then I certainly want some reputation. The two are not as connected as it seems. I'd suggest that the real reasons behind not voting are lack of clarity to the meaning of an up vote (correct vs. exceptional), most users only reading a few questions/answers; and most users having a limited knowledge base. –  cyberx86 Jan 9 '12 at 6:06
"...great answers to their rubbish questions, and occasionally they'll get a great answer to a great question." - This. I feel like SF is fast on way to becoming tech support for people that arrive via Google and not a site for professional systems/network folks to ask questions of each other. I don't want to log on every day to downvote/flag to close everything I see or try to resolve a "HALP! It doesn't work!" for lots of effort and no upvotes. –  kce Jan 9 '12 at 8:40

I vote as a public service. I do it to tell others who come after me "yes, this answer is a good one" or "yes, that is the answer I would have written." Occasionally I vote down, to say "do not use this answer, it's not good."

Like many people, I don't vote much on questions. I try to vote up everything I answer, to thank the asker for giving me the opportunity. I can usually flag or VTC bad questions, or edit them to be a bit better, so downvoting them is rare. When I have a problem, do a search, and get my solution, I vote up the question since that person is the reason my problem is immediately solved without my even asking. I also vote up any answers that helped me with the problem.

At no point in any of this do I consider the other answerers my opponents. They are my colleagues, and I cheer them on when they do well. If you think more people should vote, I recommend adopting this thought process and trying to pass it on.

share|improve this answer
Thanks for the response. I actually fully agree with all your positions (and I don't in fact consider other users my opponents). I still consider myself rather new to SE (only 6 months), but I have only recently started voting (I think I cast more VTC than other though, which needs to change). I always find it surprising that almost all the searches I do lead me to questions (and answers) already on SE. I do think still believe that some form of rep based incentive would increase voting. (Also, hello to a fellow Ontarian). –  cyberx86 Jan 9 '12 at 6:19

This is the second time this week that someone has tried to "fix" the voting problem by making complicated changes to the system. In this answer, I wrote about how we had a similar problem on gardening.se (admittedly, a tiny site when compared to SF), and by demonstrating to the community that there was a paucity of upvotes, we were able to improve our voting stats to some extent (baby steps).

My analogy there, which made sense to Kevin was:

If only 30% of the electorate in a country turn up for the polls, do you think the correct solution is a change in attitude of the society or a lowering of the voting age by an amendment to the laws?

Let me change that last part to "increasing the weight of an individual's vote" and post the same question to you. Do you think that makes sense? I ask this seriously, and not to sound rude. If you do, I'd like to know why that makes sense to you.

You say that in real elections, there is some impact on the individual/community, whereas here, there isn't. Guess what? There is an impact here too. What people need to realize is that while upvotes serve the explicit purpose of rewarding the question/answer it has an underlying hidden role in keeping the fabric of the community intact.

If upvotes are hard to come by, eventually it will lead to frustration in the community that their efforts are not being recognized. This in turn leads to users either quietly leaving or looking at alternative ideas such as inflating the value of an upvote. A good example is webapps.se which, although being the first to graduate in the wave of graduations last year, has only a single 10k user and 14 above 3k. In comparison, Gardening has three 10k+ users and eight 3k+ users although it's a beta site and has been around for just 7 months. The stats are a lot worse if you compare users at different privilege levels (although I don't think this is fair to them, since we're a beta site).

In the end, it's upto your community to resolve this problem. How you do it, is upto you. Perhaps you might be interested in compiling the stats and writing a meta post. Perhaps the mods might do it. Perhaps no one will, and that's fine too. Eventually each one will go their own way, off to greener pastures.

share|improve this answer
The weight of a vote and an incentive to vote are very different things. weight: the vote has more value (i.e. your vote and mine count different); incentive: there is a demonstrable gain for the voter. (So the altered analogy doesn't make sense - I don't want to open voting to more users, but to get eligible users to vote more) Governments do regularly (although not always successfully) try to 'change the system' to increase voter turnout (proxy, mail-in, online voting, etc.). I fully agree that voting is essential to the continued existence and growth of the community. [...continued] –  cyberx86 Jan 9 '12 at 6:59
While I don't think that the meta sites get nearly enough pageviews to bring about a wide spread change, I do think your idea has merit (and your statistical analysis looked good). We are both trying to solve the same problem - just using different approaches. Every game has rules - rewards and penalties - sometimes we can hope that everyone will do as we ask - but I think that having something tangible to back that up goes a long way. (An interesting article which suggests that voting is socially motivated) Thanks for your response. –  cyberx86 Jan 9 '12 at 6:59
My apologies for writing the analogy in terms of "weight of a vote". I fully understood your point, but accidentally wrote something else as I was recollecting it from memory and mixed it with other MSO requests for high rep users' votes to count more. The essence of it still remains – replace it with "cutting you a check for voting" :) Your point about incentives is valid, but I guess after a certain stage, badges and reputation don't matter – I certainly don't care for either. At that point, it's the interest in the community that keeps you going. Anyway, good luck and I hope SF improves :) –  Lorem Ipsum Jan 9 '12 at 7:12

I vote a lot (the most on SF) because I find it educational to read a lot of questions and answers. I don't answer as many questions as I'd like to, so I contribute to the site by voting up the questions and answers that I think are good and down the bad ones. (And lots of voting to close, to get rid of the really crappy ones.)

I've previously suggested a badge for people who consistently vote a lot as one way to encourage more voting, but there's no sign of SE, Inc. interest in that idea so far. I think a non-rep incentive like this could help. Certainly anyone who's worked on getting the maximum flag weight would likely go for a hard-to-get badge.

There are some, limited reputation bonuses for other activities (e.g. suggesting edits), but they're a bit more straightforward (suggest an edit, if it's accepted you get a small bonus) than what you've proposed and I think that type of bonus would make more sense. Maybe something like "Every month that you exceed 1000 votes, you get a 50-point rep bonus, up to a maximum of 1000 points."

share|improve this answer
Thanks for the response. I am still rather new to SE (6 months), but am trying to vote a bit more often (including VTC) - at least a few votes a day. I fully agree that reading over other people's answers is a great learning experience. Part of the problem for me is why I view questions. I'd break my question views down as: I have answered (15%), I think I could answer (50%), found searching for information (10%), and curiousity (25%). Most of the time I simply browse the unanswered questions. This isn't so much because I want rep as that I like problems. [...continued] –  cyberx86 Jan 9 '12 at 7:50
Reading over the responses to my post (and with a bit of introspection), I am no longer convinced that rep would be a sufficient motivating factor - perhaps having two levels of votes - one to say 'correct' and one to say 'exceptional' might help or granting additional privileges to voters. Given that there are only 273 'suffrage' and 60 vox populi badges (out of 69k users) I don't think that asking for 1000 votes a month is likely to impact a many people - something to the effect of 5 votes everyday for a month might be more realistic. –  cyberx86 Jan 9 '12 at 7:50

You must log in to answer this question.

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