What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 130 Stack Exchange communities.

Hindsight on acceptance percentage related to the hypothesizing seen in this question: Will the “Accept rate” score encourage unwanted behavior?.

I've had a couple of accepted answers now where it was clear that they really shouldn't have accepted the answer -- they wound up doing something else than what my answer said to do, and they said so.

How prevalent has this behavior become since the implementation of the acceptance percentage?

Feature request: Is there a way we could indicate that an accepted answer shouldn't have been accepted so that those interested in data-mining SO can figure out if the acceptance percentage thing is a net win or loss?

EDIT: To clarify: Would it be worthwhile to allow us to mark an accepted answer as "erroneously accepted" so that data mining could tell us if we think we're getting too many of these? I don't think it's something that needs to be shown to all users or anything; perhaps implement it similar to flagging for moderator review. Just have it available in the data dumps so we can try to figure out if it's a net win or not.

share|improve this question

6 Answers 6

up vote 13 down vote accepted

It encourages people to go back and accept inappropriate answers to old questions that didn't get answered.

Even if they did find an answer, they still may mark SOMETHING as accepted, rather than writing out what they found out, and then marking that as the answer.

I know I've done it.

It is potentially misleading to future surfers who find the question, and raises another problem: If someone asks a duplicate question, and the accepted answer on the existing question is wrong, there's currently no good way to encourage a fix.

share|improve this answer
4  
I agree, I don't like the acceptance rate either, some times none of the answers is correct, the question goes to oblivion and it affects our acceptance rate when it really had nothing to do with us. –  Carlo Oct 23 '09 at 1:27
2  
I agree as well. To me, the acceptance rate doesn't seem to make one bit of difference anyway, people will still rush to answer people with 50 questions and 20% answer rate because of upvotes from other users. And frankly I have found myself considering my upvote rate and being tempted to select some crap answer just because it has 1 vote and there are no other answers with any. (especially on SU, where I almost never seem to get answers that I am satisfied with). –  TM. Oct 23 '09 at 19:13

The problem isn't the accept rate score on its own, but how it affects the dynamics of the site.

The problem is being compounded by too many people leaving comments like "Work on your accept rate", effectively making a veiled threat that their question isn't going to get answered until their older questions get accepted answers. Those comments should be left on the older questions, not the new one, and only if it's obvious that the old question was solved; otherwise it encourages the wrong behavior.

I've had a few instances recently where my answer didn't solve the problem, but got accepted anyway long after the question became stale. Here's the most recent example.

share|improve this answer

It would be nice if moderators had the ability to uncheck the 'Accept" box, or mark a different answer as Accepted. It should be used sparingly, of course.

EDIT: This is for the cases as described by the OP: "...it was clear that they really shouldn't have accepted the answer -- they wound up doing something else than what my answer said to do, and they said so." In that case, the moderator could correct the question/answers state to reflect reality. I do not mean to impose more work on the moderators.

EDIT2: Why so many downvotes? Are people really that distrustful of the moderators that giving them this ability is seen as so horrible?

share|improve this answer
1  
I don't think that's the right approach... it's more work for the mods for one. –  retracile Oct 22 '09 at 22:42
4  
I can't see a sensible way for me (or any other mod) to arbitrarily dictate correctness. –  Marc Gravell Oct 22 '09 at 22:46
2  
I had an answer accepted which actually solved the guy's problem, then six individuals downvoted my answer because it wasn't "politically correct". Having a way for a third party to un-accept an answer which the person who asked the question had accepted may lead to some problems, I think. –  delete Oct 23 '09 at 1:44
    
@Kinopiko: yeah, I was kind of thinking in terms of the person who owns the accepted answer making the call that it shouldn't have been accepted, and flag it in a probably non-visible way as a failure of the acceptance metric so we could understand the scope of the issue with some actual, you know, data. –  retracile Oct 23 '09 at 2:09
    
You know data is probably better than you don't know data. –  delete Oct 23 '09 at 2:58

I've seen some old questions where people answer with "I just want to close the question" and than accept that (like here or here). I don't think it's a huge problem, but it's adding useless noise to these questions.

share|improve this answer
    
That second example doesn't really apply. He asked and answered his own question in the OP, then made a stub answer to accept so the question could be marked as solved. He wanted to put his problem and solution on the internet so it might help someone. –  RandomEngy Nov 13 '09 at 23:28

When the acceptance rate was first introduced, it was made clear that not having a 100% acceptance rate was perfectly appropriate. If your question wasn't answered, don't accept an answer. If you solved it yourself, post what you really did to solve it and accept your own answer.

The only reason I've got a 100% acceptance rate on SO is because they all have real, working answers.

share|improve this answer
    
and less than 100% doesn't stop people from answering your questions. –  Paul McMillan Oct 22 '09 at 23:15
    
I usually get concerned if it's less than 50%, personally. –  Gabriel Hurley Oct 22 '09 at 23:20
3  
What acceptance rate are you talking about? Oh! That percentage number that I never look when answering a question in its own merits? –  perbert Oct 23 '09 at 0:03
2  
Regardless of how clearly the intent was stated, or how reasonably-minded people view the importance (or lack thereof) of the acceptance rate, I think it is fairly certain that it is encouraging non-constructive behavior from at least some users. To quote a good friend of mine, "It is a question of degree." –  ベレアー アダム Oct 23 '09 at 1:32
    
@Adam: I agree with you, I was trying to sarcastically say that acceptance rate really doesn't mean anything on the greater scheme of things. What should matter is only the quality of the answers and questions. –  perbert Oct 23 '09 at 2:39

Don't forget - your answer, though not 100% correct, may have been enough for them to work out the solution themselves. In that case I can see why they might accept your answer.

In fact I did this very thing with one of my questions - well before the acceptance rate came in.

share|improve this answer
    
Hm. Hadn't considered that... –  retracile Oct 22 '09 at 22:29
    
This is definitely a case where the OP should explain that in a comment! –  squillman Oct 22 '09 at 22:33
1  
@squillman - agreed. In my case I posted an answer as I needed to include the code I actually used. –  ChrisF Oct 22 '09 at 22:48
    
@ChrisF exactly! if you solved it yourself, post what you did! That'll help future readers more than a hint from another answer. –  Gabriel Hurley Oct 22 '09 at 23:14

You must log in to answer this question.

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