(This is perhaps pretty close to Question/answer ratio etiquette in Stack Overflow? and maybe a few others...but it seems at least a smidge different than an "exact duplicate" to me...)

It seems that "a lot" of "stupid" or "bad" questions (asked only for reputation points?) come from people with few answers. While I'm by no means any type of an expert on Stack Overflow (with a measly 1000+ reputation points), I answer a whole lot more questions than I ask--by a ratio of about 10-answered to 1-asked.

Is there anything on Stack Overflow that is driven by this ratio? Should there be? If so, what would be a "good" question/answer ratio?

While we all want people to feel free to ask questions, there is a lot of benefit in both answering questions and taking the time to write a decent question.

  • That's why when dupes are closed, they're marked as Possible duplicates:
    – random
    Oct 26, 2009 at 1:12
  • 1
    Any above 0 and below 99 :)
    – perbert
    Oct 26, 2009 at 1:25

6 Answers 6


The exact answer to Dan is that, unfortunately, there's absolutely no benefit or even recognition (no extra perms, no badges, no stats displayed outside of the Q and A counts on user's profile) for people who provide answers.

@Bill - I beg to differ. It is significantly easier/more rewarding to ask a question than to provide an answer:

  • The pool of questions is very large - easy to find a Q to ask.

    As was noted before, some people succeeded in acquiring a fair amount of rep points by merely doing what amounts to typing in every question from a textbook (I'm exaggregating but not much) - and as long as the questions were fairly well worded and not GimmeTheCodez, they don't get voted down or closed due to Joel and Jeff's attitude about newbie questions and overall purpose of the site becoming an "encyclopedia".

  • On a related note, you don't really need to know anything to ask a question, so your range of topics for asking is nearly infinite, whereas you need to posess at least some knowledge to be able to answer all but completely trivial questions.

  • The risk of posting a question is much lower. Especially risk adjusted by amount of work/time spent.

    Very few questions aside from really awful get downvoted much and many get upvoted. Answers get down-voted much more, and for many more reasons - someone didn't like the answer, someone thinks it's wrong, someone wants to improve their own answer's placement by tactically downvoting others, etc...

    Even leaving aside spurious things, the risk of having your answer be not correct and thus downvoted exists unless you're Jon Skeet.

  • It's much more work to produce a good answer

    You need to make sure it's correct, it often needs some working code, you should have some links to references, and you should format it nicely. Most questions can be a lot sloppier and will be shortly edited to better form (much more often than asnwers).

  • There's no time pressure on questions.

    You can post at any time, take as long as you want. Answers, you need to catch the question you can answer on time (before someone else posts 100% good answer and before someone posts 100% same as yours answer - hopefully the same thing :) ) and you need to try and answer it as soon as possible due to TFGITW bias (one of MSO posts had very telling statistics about TWGITW).

  • Asking a question has its own rewards aside from rep points and quite likely greatly more important than them (from at least satisfying burning curiosity to helping with making your paying job easier and your work faster). Answering, aside from a dicey chance of earning some rep, merely allows you to feel better about yourself and/or show off a bit.

So, by treating questions same as answers, you merely create a bias towards rewarding questioners (they earn easier/more/faster rep on average due to factors above) and thus a pressure against answerers. And while you need questions for Q&A side, the questions will ALWAYS come even if there's no rep awarded for them - they will come from people genuinely wanting to know the answer (or people wanting to better the site - just as people answer wanting to better the site/community). Yet if you drive away the answeres (or lower their motivation - which personally already happened to me), you lower the quelity and potential of the site.


  • "@Bill - I beg to differ. It is significantly easier/more rewarding to ask a question than to provide an answer..." How does that differ from any of the points that I made? Oct 26, 2009 at 14:45
  • 3
    I downvote questions all the time because I find poor grammar/wording/English, failure to RTFM, etc. to be "unclear or not useful". Of course, that usually does more harm that good as someone comes along and does a "sympathy upvote".
    – Ðаn
    Oct 26, 2009 at 17:18
  • 1
    @Dan: Yeah, which means that I'm very reluctant to downvote questions. I'd only downvote if I thought a decrease in rep was warranted, and one downvote plus one sympathy upvote = 8 rep. I hate sympathy upvotes. As a geek, I like to do things according to known rules, not to try to outguess other humans. Oct 26, 2009 at 18:12
  • Maybe if the tooltip text was changed, I wouldn't downvote so often. I figure I'm doing my part to flag "unclear" or "not useful" questions by downvoting. But maybe I'm just taking "unclear" and "not useful" too literally.
    – Ðаn
    Oct 27, 2009 at 2:57
  • @Bill - the main point of your answer seems to me: Uber-good Qs are great, uber-bad questions are bad and we will police them, but anything in between (vast majority) is as good as answers - e.g. there's no reason to drive anything by Q/A ratio. Please correct me if I misread it.
    – DVK
    Oct 27, 2009 at 14:47
  • 1
    @DVK: That's a fair enough summary of my first and last points. The reason Q/A ratio is worthless is because it doesn't in any way measure quality of either questions or answers. Good answers tend to get far more upvotes than good questions, so the community has already responded to the disparity in difficulty of asking good questions vs. answering them. Answers already get rewarded far more than questions. My main point, though, was that there are other ways to give back to the community than answering questions. Q/A ratio can be ignored. Oct 27, 2009 at 17:36
  • 1
    Actually, now that I've given this a few more minutes of thought, it wouldn't be that hard to build up a Q/A ratio that does take quality into account (we have the vote information for both right there). If you only counted positively-voted questions and answers (or weight them by votes somehow) a Q/A ratio could be made to contain some useful information. Oct 27, 2009 at 17:58
  • 1
    @Bill - sounds like a good idea. However I'm not sure I fully agree with your assertion on earlier comment that good Qs get up-voted less than good answers - on the more trivial/popular Qs it seems to be vice versa since people up-vote the question as a sign of "I wanna know this too", and all those people either don't vote for the right answer because they saw the Q before the answer was provided, or many deem themselves not knowledgeable enough to judge the quality of the answers (I know I sometimes tend the latter way on topics I'm not too familiar with).
    – DVK
    Oct 29, 2009 at 18:08

I don't think there's any one right answer to this, and as far as I know nothing on the site is driven by your Q/A ratio. If you look at the top users you'll see some tiny Q/A ratios. Question quality is a lot more important in this regard, though. I don't think anyone would fault you for asking 100 high quality questions for every answer you gave.

We do like people to contribute back to the community, but there are other ways to do it than answering questions.

  • Ask good questions
  • Edit other people's posts for grammar/spelling/accuracy
  • Flag spam for moderator attention

Finally, there have been a handful of cases where a few members asked such a huge number (hundreds) of dreadful questions, and the community flagged them often enough, that a warning email was issued to inform them of the problem and what to do about it. In other words, we will let you know if it's a problem. ;)


Asking a good question can take a lot of effort -- you need to clearly lay out what the problem is that you are trying to solve, document your efforts to solve it, specify constraints that disqualify some solutions, etc.

But you can easily ask questions you haven't put effort into, and someone else will clean it up for you, and you will get the rep for the improved question.

As it stands today, the rep system is rewarding inquisitive idiots with increased control of SO, and I humbly submit that is not a good thing for the SO community.

Down votes tend to indicate that people think the question should just be closed. Up-voting a question seems to me to say "I want to know the answer to this question too." So instead of awarding the rep to the original asker of the question, award the rep to the accepted answer instead.

A simpler implementation that just eliminates voting on questions themselves would also be an improvement to the status quo.

  • at the very least, eiher award less rep per Q up-vote, or keep that rep in a separate bucket.
    – DVK
    Oct 27, 2009 at 14:57

Ah well: an instance of a not good Q/A ratio is 63 questions, 1 answer (and that one down-voted twice), with a 28% accept rate. The user in question has made 22 up-votes and 3 down-votes too. And has a reputation of 578 - enough to go retagging. (Now (2009-11-09) nearer 750 reputation, and currently 'nameless'; the name has been removed from the user entry.) Frankly, I find that unsatisfactory; the user has not demonstrated competence and should not IMNSHO be allowed the perquisites that go with the accumulated reputation.

I've come across another 'not good' Q/A ratio - 26 questions: 2 answers (both of which should be edits to a question by this person, both showing 0 votes); 0 up-votes; 0 down-votes; 0 accepted answers; 245 reputation points.


There are those who:

A. Ask a lot more than they answer (if any at all)

B. Answers a lot more than they ask

C. Quite balanced in asking and answering

The ratio actually doesn't bother me. What really bother me is those whose acceptance rates are low and the asking ratios are high. Having said that, it's my personal opinion.


I think people need to stop caring whether or not people answer questions.

It's not important. It's only important whether they ask good questions. If they ask bad, vague and easily searchable (on SO, it's not important that they can be found on Google) then that's different.

You don't have to earn the right to ask questions. Let it go.

  • 4
    Cletus - Joel explicitly stated at some point that SO rep is used in job hunting (either is, or he aims it to be - don't recall). As such, one's relative SO rep becomes more important. So anything which unfairly penalizes people who are providing a service to the community by answering the questions (as opposed to those who mainly provide service to themselves by asking) in essense diminishes the value of answerer's rep by moving the curve way up. So yes, it's important
    – DVK
    Oct 26, 2009 at 6:58
  • 2
    BTW, my own suggestion to fix this is to at least merely separate Q from A rep. 100% fair - you don't change site's behavior/functionality in any way, you merely allow people who do answer question to have a separate grading curve - see my reply in this thread as to WHY the curve needs to be different.
    – DVK
    Oct 26, 2009 at 7:01
  • @DVK: I'm not sure, if you think that recruiters are idiots. If they find a high rep user the next thing they will do is checking his user profile. If they find hundreds of questions and no answers, what do you think they will do? Hire him? Oct 27, 2009 at 8:15
  • 2
    @John - if the recruiter needs to balance between 2 similar resumes with pretty much the same qualities, and one has rep of 5k and one has rep of 8k, he might tilt to 8k one without bothering with Due Diligence. I've seen people decide on much flimsier reasons than that.
    – DVK
    Oct 27, 2009 at 14:56
  • 2
    @DVK: If they do that, I don't think you want to work for them anyway. Look at it as a QA for the job seeker ;) Oct 27, 2009 at 20:55

You must log in to answer this question.

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