1

"Reasonable" is of course highly subjective. But even granting that, there must be a range of acceptance rates that most users would agree is reasonable.

For example, I have asked 64 questions and accepted answers on 56, for an acceptance rate of 0.88. Some answers I accepted didn't answer my questions as well as I had hoped, but they were very helpful and it seemed unlikely a better answer would come along. In other cases, I wished I could accept more than one answer.

Does an acceptance rate in the range of 0.75 to 0.90 seem reasonable to the rest of you? If not, why not?

  • 1
    87.3951% or you're a jerk. – Olin Lathrop Apr 8 '18 at 13:31
  • 1
    Related: meta.stackexchange.com/q/136951. The original Stack Overflow used to display this percentage under your name in your questions. Others used it decide whether it was worth giving an answer if the asker was unlikely to accept it. Controversy ensued, long story. – cobaltduck Apr 9 '18 at 15:35
  • @cobaltduck Thanks for an informative comment. I didn't know the AR used to be posted; I agree that it shouldn't. I'm still interested in what users here think is a reasonable range for AR ---- obviously 6 digit precision is nonsense. – ab2 ReinstateMonicaNow Apr 9 '18 at 18:19
5

Total opinion piece following, admittedly without longer reflection:

I don't know my acceptance rate and don't care - if there is an obvious answer to accept, I do, if not, I don't. Sometimes I revisit my unaccepted questions to see whether there is a good candidate on a second read. I don't understand why accepting should be a big deal. For me voting is the main instrument for quality questions. Sometimes accepted answers are even annoying, as they put worse answers above better ones - I can't remember this on TGO, but I come across it quite frequently on other SE sites.

To answer your actual question: I consider 0 to 1 as reasonable. So yes 0.75 to 0.9 seems reasonable, but not a potential inverse statement (that <0.75 and >0.9 is unreasonable).

Lets bet I am going to get to know my acceptance rate soon ;)

  • 1
    You asked 20 questions, 2 of which were duplicates and had no answers. You have accepted 9 answers, for an AR of 0.50. – ab2 ReinstateMonicaNow Apr 10 '18 at 1:16
  • 2
    Wow, I am perfectly centered within my personal reasonable acceptance interval. Lets see whether I can get unbalanced :) Ah and thanks for the info! – imsodin Apr 10 '18 at 10:48
3

It's a tricky thing. I would say that many a times, it depends on the question one asks as well. There are questions which can be answered to the point (a lot of rope work and knot related questions seem to be having pin point solutions). In which case, you can accept the answer. There are questions which cannot be comprehensively answered (many of my questions fall into this category eg: breathing techniques, walking techniques et al). With those, it depends on the person asking the question to accept or to hope for a better answer to come along.

Personally I've had many questions where I felt the question in itself cannot be answered wholly to cover everything. However, I do go ahead and accept the one which I feel is best describing the solution for my problem.

In short, a higher acceptance rate means nothing for me. Unless of course, the OP doesn't ever want to accept any answer, in which case I'm just intrigued by the person. I would still go ahead and answer the question to the best of my abilities.

  • I agree. The self-answering crew here will obviously reach acceptance rates close to 100%, while people who only ask difficult questions they do not know the answer to (hopefully after investing some time on Google) will have a much lower acceptance rate. I think it matters not. – fgysin reinstate Monica Apr 12 '18 at 9:49
2

Since the acceptance rate of someone asking a question isn't immediately obvious, it doesn't matter. In other words, since those considering answering the question don't know your acceptance rate, there are no penalties for a low rate. Don't worry about the numeric rate.

What you should do, is accept the answer that helped you most, if one did. If they were all off the mark, so be it. Accepting basically says, "Thanks, that solved my problem". If none of the answers were all that helpful, then it's OK not to accept any. If at least one of them was, then you should pick one and accept it.

Each question is a separate case. Obsession over the numeric acceptance rate is pointless because it's based on a the false premise that all your questions not independent of each other, and should fall into a nice and neat pattern.

That is not to say it's OK to ignore accepting. You should when it's appropriate. Questions that don't have accepted answers are considered unanswered, and are automatically bumped to the top of the questions page occasionally. That's useful if the question really isn't answered, but a distraction and rude to everyone else if it is answered.

Another point is to not accept too early. Leave at least one day so that everyone around the world has a chance to chime in. No matter how good any one answer seems, you can't really know what brilliant revelation or unexpected insight someone else might add. I don't bother entering questions, let alone answer them, if they already have accepted answers. My limited time here is better spent elsewhere.

  • +1 "Another point is to not accept too early. Leave at least one day so that everyone around the world has a chance to chime in. No matter how good any one answer seems, you can't really know what brilliant revelation or unexpected insight someone else might add." – James Jenkins Apr 16 '18 at 17:57
1

The upside of high accept rate is that people are more likely to answer your questions and here is why

  • Some people are very generous with their upvotes and accepts and are highly involved in the process.

  • Some users are less so.

I am way more excited to answer questions posted by people in the first group, and will admittedly put more effort into my answers.

With that said, I don't have a spreadsheet of users and their accept rate, it's more of a fuzzy thing.

You must log in to answer this question.

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