Did the reduction from 10 points per upvote for a question to +5 points have any effect on community voting patterns?

  • Are a higher percentage of questions being upvoted?

The definitive answer lies somewhere in the datadump.

share|improve this question
    
and no, I will not be writing this query for at least 24 hours ... I want others to have a go first :) –  waffles May 13 '10 at 6:58
6  
perhaps, but IMO the "Electorate" badge has a larger effect. –  Nick Dandoulakis May 13 '10 at 8:38
    
If I were going to query the data dump, I'd query the time the Electorate badge was introduced until the time the +5 was introduced. –  George Stocker May 13 '10 at 11:55
1  
@George, +7 you need to look at 7 day chunks whenever you look at patterns as its cyclical –  waffles May 13 '10 at 11:58
    
the +5 questions, not +5 timeframe. –  George Stocker May 13 '10 at 14:25
add comment

4 Answers

Looking at the percentage of question votes (in comparison to answer votes) it is totally clear something happened late December.

However very little changed when the new rules kicked in.

See this query

Looking at the period after march 21st, it seems the ratio remained pretty stable. The volume went down a week later (but that was probably due to holidays)

once exported to a CSV and graphed you see this: alt text

Its amazing the effect the new electorate badge had. (unless I'm missing another change in late December)

share|improve this answer
add comment

I think that directly speaking, no, decreasing the value of votes on questions did not affect the amount of votes that questions would get. It would be lying to say that there is no meta-voting where people will judge who gets their vote based on their current rep... but I hardly think that anyone who uses such bias really cares too much about whether they're giving the person 5 rep or 10 rep. Voting on questions more because they give less rep is... baffling. It's not petty or negative or anything, because it simply makes no sense. It is not as if the voter has a limit on the amount of rep to grant, just implicitly through how many votes are allowed. It otherwise seems meaningless to allot your votes based on how much rep they give other people.

However, it may incur extra voting on questions, by virtue that votes on answers have greater value. Some people vote up questions in order to make them more attractive to look at (higher voted questions probably get better attention) and so if you answer a voted-up question, then it'll be more likely that you'll receive upvotes for your good answer by proxy. And those votes, in turn, are worth more than if you asked your own questions.

Of course, this is an analysis on reasoning more than actual data trends.

share|improve this answer
add comment

I don't think that the change in rep was supposed to increase voting for question. There was some notion that it might reduce downvoting. The primary goal, as I understood it, was to make it harder to pile up giant amounts of rep with large numbers of muppet-chow questions. This was in turn theorized to reduce tendency of people to vote down these questions, resulting in less plaintive emails to the team.

share|improve this answer
add comment

Graphed the same query run on Super User. Wonder why the ratio drops so low for June. When did SU private beta end ?

alt text

share|improve this answer
add comment

You must log in to answer this question.

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