What all we know, we (i.e. academia.SE) are hacking the system for objective verifiable Q&A for the purpose of advise questions. (And, IMHO, we have hacked it successfully).

However, I have some doubts when it comes to accepting answers. On StackOverflow it's obvious if an answer solves your problem. Then, out of such, you can choose the approach you actually took.

For soft-questions you don't have "I copied and pasted your code, it works, thanks". Usually there are piece of advice and wisdom in many "answers". And none of it "solves" the problem (actually differences in opinions are often fruitful and show academic landscape).

Of course sometimes there is an answer which is worth to be singled out among other. But in other cases, when there are more compelling answers - what to do?

3 Answers 3


What were you looking for? When you ask a soft question, you're definitely not looking for a unique actionable “that solved it thanks” solution, as you say. You are usually looking for ideas, advice, viewpoints different from your own, etc. So, I would suggest to:

  1. Mark as accepted the answer that provided you with the most useful advice. It may not be easy to decide, but probably one of them has a point of view that you wouldn't have considered by yourself, or an answer backed by quotes or statistics.
  2. If you can't make up your mind, choose a good answer amongst the later ones: late answers tend to receive less exposure, and thus less votes.
  3. Possibly write a comment below it, explaining how you found it useful, and that you really appreciated insight given by the others.

An alternative would be not to accept any answer. I think it's not very satisfactory, because it sends the message that “none of this helped me” (of course, if that's true, then don't accept any answer) and, less importantly, it lets a good +15 rep go to waste.


Additionally to F'x answer, I would like to mention another point of view. The point of view of a person who will look at the Academia site for some answers and will read your question in the future.

If the person is very interested, he or she will read all the posts and think about their value by himself (herself), not just the accepted one. In this case it does not matter which answer you will mark.

Most of the people will probably read just few top rated answers. In this case, you can proceed according to the point 2. of F'x answer. Give the privilege to an answer which is bellow but you find it valuable and help others to learn more in this way.

(if the person will read just the accepted answer or the first one, he or she is probably not that interested in that matter and again it does not matter what you will mark out.)

So, If you have doubts which answer to mark, you can think about which answer will be the most beneficial for other possible readers in the future.


If there is no clear right answer possible, then the question doesn't belong on the site. It should be closed as "not a real question" or "off-topic", and then possibly deleted.

If a clear right answer is possible, but it is split across several existing answers, then the thing to do is to write a synthesis of all the right things into a single answer.

  • Writing a synthesis has some downsides: in particular, if other answers are added later, the synthesis become outdated. I would argue that the whole page (question + upvoted answers) is accessible anyway for people to read and make their own mind.
    – F'x
    Commented Nov 10, 2012 at 15:21
  • 4
    @EnergyNumbers With all due respect, if you were right, then academia.SE wouldn't exist. It's good to be aware of the fact that most of questions here would be closed on other scientific Q&A sites as subjective. And even there the questions are less objective than on StackOverflow itself. Commented Nov 10, 2012 at 16:56

You must log in to answer this question.

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