Is there any guarantee that the accepted answers or the ones with the highest score on StackExchange are the correct ones?

On the other hand, couldn't an answer with a low or even a sub-zero score be correct?

Is common sense the only reference? and if so, how can we solve that?

4 Answers 4


Unfortunately, there's no justice, there's just us. The community is all there is, and the community often gets it wrong.

  • @Shinto Sherlock, that's what I thought, and now that you said it, it became a sure thing to me. Anyway, there is always a hunch about whom you may trust! Thanks!
    – Dia
    Sep 7, 2010 at 14:07
  • Reminds me of the despair.com poster that says "None of us is as dumb as all of us."
    – Chris
    Sep 10, 2010 at 19:25

There's two different things going on here: the short-term and the long-term. Here's an example:

(note: the following has exaggeration for effect; live with it)


ESLguy comes in and posts something like:

qestion my:
"we go store Yesterday n bi meets" rite iz ?
teecher sez rong but i rite i no.

  • Answer #1 says no, that's wrong.
  • Answer #2 says no, and here's a better way to say it.
  • Answer #3 says, "itZ Go0d 2 mEe."
  • Answer #4 says no, and here are the particular language/grammar rules you're breaking.

Unsurprisingly, ESLguy prefers answer #3, and accepts it as the best. Oops.


For a variety of reasons, people wander into the site and run across the question. Over the next few years, answers 2 and 4 get a bunch of up votes and 3 gets even more down votes.

End result (?)

The theory is that, in the long run, answers 2 & 4 will have accumulated scores that are so much higher than the accepted answer that it will be obvious to all that—while neither is the accepted answer—they are plainly better answers.

  • remarkable analysis! I agree with you.
    – Dia
    Sep 7, 2010 at 14:12
  • 5
    I like how the scores of these answers are exemplifying exactly what you are talking about.
    – Kosmonaut
    Sep 7, 2010 at 16:34
  • 1
    The condition for this nice story is that the pool of problematic questions remain shallow to a large subset of the community. The Unanswered tab is a good start, some of the review and moderator tools as well, but we can perhaps think about even more ways to interest users into the long tail of questions after they have plunged behind the first pages.
    – ogerard
    May 20, 2011 at 9:19

I try and cite my responses whenever possible. The only problem is that I'm normally citing webpages I found with google which usually wouldn't pass the academic standard for a quality reference. My responses are at least as good as my references.


One thing I've noticed on this site in particular is that there are two different - but equally valid - answers to some questions.

There is the technical answer - e.g. which is of these examples is grammatically correct, which is likely to cite a dictionary or a Wikipedia entry.

Then there is the "common usage" answer, which may contradict the technical answer, and which may cite no source other than the experience of the answerer.

Given a choice of which one to mark as the accepted answer, the questioner may be in a quandary. Whichever one he or she may choose, this is a good example of where the community's votes and comments are most useful to any subsequent visitor.

  • 1
    I call false dichotomy here. An answer that’s either of these at the expense of the other is a bad answer. A bad answer may claim “technical correctness” based on some bloviating pundit or unsourced style guide, or claim “common usage” based only on personal experience. But a good “technical” answer will refer to serious, well-informed dictionaries or the like; and a good “popular” answer will be based on real usage data; and these will tend to agree, or at least to not conflict.
    – PLL
    Dec 13, 2010 at 4:14
  • @PLL: Are you sure about the "not conflict"? In Czarist Russia, the common people spoke Russian while the upper classes spoke French and in many cases could not even speak Russian any more. Similar things, though with much less distance between the opposites, must exist in many languages, to various degrees. Jan 5, 2011 at 1:51
  • 1
    @Cerberus: “If you wish to be thought well of in court, you should say J’habite à Saint-Pétersbourg, as may be read in the letters of the finest aristocrats. If, contrariwise, you would be understood by tradesmen and peasants, you would be better served by Я живу в Перми, in an idiom that has been scrupulously observed by some of our more progressive novelists.” There is a conflict in the language, between what is correct in different contexts; but there is no conflict in what an ideal answer should do: describe what is standard in which contexts, ideally with some justification.
    – PLL
    Jan 5, 2011 at 2:58
  • 2
    @PLL: Good quote. I think you are right about the different contexts. Perhaps the problem lies in this ramification of answers into different contexts, since it is not always easy for the reader to match the right branch with the right context. An answer might be intended to mean "x is said in group y" but interpreted as matching x to a slightly different or much different group of people, especially since these groups are often not mentioned explicitly. What is more, speakers are often not aware of the standards of a group that they are not in. Jan 5, 2011 at 3:29
  • 1
    The best "answers" contain both actual answers, really..
    – SamB
    May 4, 2011 at 4:54

You must log in to answer this question.

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