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.

I think it could be a good idea to have the ratio of your answers with 0 votes vs. your answers with 1+ votes on the profile page.

It could give further insight into the quality of a persons answers.

I'm generally thinking of this because I see quite a solid amount of people with relatively high reputation who have 1000 answers with 0 votes (I'm just making the numbers up of course ...) and 100 answers with 10+ votes.

I'm not saying this is bad, I'm just saying that if we can see the 'answer ratio' we can maybe make a slightly better distinction between people that just answer everything to try and 'grind reputation' and people that actually put time and thought into their answers.

This could however lead to people being afraid to answer questions because they might loose ratio...

Thoughts ?

share|improve this question
    
possible duplicate meta.stackexchange.com/questions/27261/… but I still think this is relevant –  Simeon Mar 25 '11 at 16:31
4  
Trivial answers to popular questions usually get many votes, and answers to less popular questions get few votes if any. So this rate tells mainly about how many unpopular/specialized questions you answer. So it'd discourage people from answering unpopular questions which are exactly the questions where answers are most needed. –  CodesInChaos Mar 25 '11 at 16:45
    
@CodeInChaos ... agreed, will think about this :) would you mind posting this comment as an answer so we can discuss it easier –  Simeon Mar 25 '11 at 16:49

3 Answers 3

up vote 7 down vote accepted

Trivial answers to popular questions usually get many votes, and answers to less popular questions get few votes if any.

So this rate tells mainly about how many unpopular/specialized questions you answer. So it'd discourage people from answering unpopular questions which are exactly the questions where answers are most needed.

In my experience one of the fastest ways to get reputation is quickly answering a popular but easy question in a tag like C#. You usually get 5-10 upvoted for that if you're fast enough. And that for a simple look at this function + link to MSDN.

On the other hand when I answer a specialized question I usually get close to no upvotes. And that answer was certainly worth more since there are not 5+ people with a near identical answer.

share|improve this answer
    
As I can type fast enough for the quick answers, this problem is making me question how many answers it is worth posting. –  Ian Ringrose Mar 25 '11 at 17:14
    
although I completely agree with this, I still think that even in a very specific question if your answer gets at least 1 vote or is accepted with 0 votes, this could still increase ratio. When I ask a very specific question for example I tend to vote up any answer which is good and I think in a greater percentage of the cases a good answer gets at least 1 vote. –  Simeon Mar 25 '11 at 17:19
    
the ratio would also not be dependent on how much votes your answer has, rather just if it has any votes at all or is accepted .... or even maybe not negative ... however as a put more and more thought into this I really think you're right and this would have a negative effect rather than a positive one ... I'll think a bit more though –  Simeon Mar 25 '11 at 17:21

If you look at this query you'll see the following

Top 100

  • Lowest Ratio: .5%
  • Highest Ratio: 38.9%
  • Mean Average : 18%

Bottom 100 above 10000

  • Lowest Ratio: 8.9%
  • Highest Ratio: 49%
  • Mean Average : 27.4%

Bottom 100 above 1000

  • Lowest Ratio: 0%
  • Highest Ratio: 100%
  • Mean Average : 40%

So there's a trend that as you go up in reputation you can expect to see a lower variance of ratio of Zero vote answers as well lower average.

This isn't very surprising or interesting since in my experience, the more questions you answer the better you are at 1) identifying which questions will yield answers that get up-voted 2) how to provide questions that get up voted.

So as per the faq

Reputation is a rough measurement of how much the community trusts you. Reputation is never given, it is earned by convincing fellow users that you know what you’re talking about.

I think the above data demonstrates that at best the Zero answer metric would provide a rougher not a more precise measurement.

share|improve this answer
2  
+1, a good metric either is not subjective to or finds a way to mitigate the problem of small sample size. –  NickC Mar 25 '11 at 17:58
    
this is what I mean by good answers you put time into :) this should definitely increase your ratio :D –  Simeon Mar 25 '11 at 18:01
    
although the best answers are not those with the most time put into them of course .... –  Simeon Mar 25 '11 at 18:02

As well as not answering they could always delete answers with 0 votes after some time has passed, which will increase their ratio. I don't really think that it would be a reliable, or even very good indicator of answer quality.

share|improve this answer
    
depends on the idea, deleting 0/negative rated answer should not increase ratio, also a time buffer could be set so you can't cheat like that, ratio can also be calculated on weekly/monthly basis rather than instantly like rating –  Simeon Mar 25 '11 at 16:33
    
It will definitely not be reliable, I will argue however that rating is also not completely reliable. The question is will it be beneficial ? –  Simeon Mar 25 '11 at 16:35
    
I don't think that it would provide any substancial benefit, but you are right about rep. –  Belinda Mar 25 '11 at 16:39
    
It would provide benefit IMHO (at least to me :D). So if you have to say yes or no, do you prefer having it or not ? –  Simeon Mar 25 '11 at 16:42
2  
After reading @CodeInChaos's answer I'm going to say no –  Belinda Mar 25 '11 at 16:56

You must log in to answer this question.

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