Whenever I submit an answer (a real answer, not some @someone comment) I upvote the question. I generally do this because if the question is good enough to take time to answer it, it's good enough to be brought to someone else's attention.
Do you do this? What are your guidelines for upvoting a question that you answer?
My opinion is the same as yours. If a question is good enough for me to stop what I'm doing and answer, I obviously think it's an interesting/useful question, so it deserves a vote.
That's really my only criteria: interesting question, or something that would be useful to other people.
Sometimes the question even deserves both an answer and a downvote.
Or an answer and no vote at all.
For me, answering questions and voting on questions are independent. I tend to view giving an up vote as something like making a statement like “Hey, great/useful question.” (likewise for answers).
Just because I am able to decipher a vague, ill-formed, and/or confusing question and supply an answer does not mean that I think it is a “great question”. Likewise, I almost never consider questions that are “RTFM-answerable” worthy of an up vote (even if I post an answer that summaries the relevant bits of documentation).
On the other hand, if I find a question or answer concerning a topic that I find important or interesting (roughly “useful”), I might give it an up vote even if it lacks (a bit of) clarity.
The real question in my mind is whether there's a point to rating questions at all.
Obviously the purpose of rating answers is to find the best one for a given question. The value of having a list of the "best questions" is much more dubious, especially given the issues people have noticed with voting on questions.
Someone looking for an answer to their question is going to find the right question using search, whether internal or external. People who want to keep up with new questions on Stack Overflow I think are going to be much more interested in sub-communities within the site, whether via monitoring specific tags or some other mechanism, than tracking some kind of reddit-like hot questions page.
Question threads should be displayed to users based on relevance rather than popularity.
Some people ask great questions. They provide necessary context; the question is short and clear and easy to understand. It's well defined. And then they've done a bit of research - they give a link to some man page or a support site or whatever. The question will provide an interesting insight into use cases or conflicts in different documentations or whatever.
Those questions get a solid up vote. I'll try hard if I answer those questions to provide good quality answers, with links and quotes and etc.
Some people just bang out a question. There's a bit of thought, but the question is reasonably easy to answer. I'll answer those questions (if I know the answer) and I'll try to give a good quality answer. I won't upvote the question, but I'd want an upvote for my answer. And a thank you comment.
I always wonder about people when I see a question with one or two favourites ...
... but no upvotes.
"I love this question! But I don't think it's good enough to vote for."
Constantly. I have a high bar for upvoting questions. Most of the ones I answer don't deserve either an up or a down in my opinion.
There are three possible points of view:
If I am bothering to answer it, it's a good question, so you should upvote.
Sometimes I answer bad questions, so I may not upvote or may even downvote.
I answer questions to strategically earn rep. Therefore, I may provide a patently obvious answer to a bad question in order to get a couple of upvotes. Also, I may answer a good question, but reserve my upvote for a later time in order to bump the question on the recent tab, and earn my answer more attention and possible upvotes.
I do not vote on a lot of questions I have answered. IMO a lot of questions dont deserve a upvote or downvote but I'd still answer them. Heck I sometimes even answer low quality questions just to get it an answer(of course a correct one).
Also there are people who just started on SO who make low quality posts but still deserve an answer.
I totally disagree with the accepted answer.
I do upvote most questions I answer, but for me, there's an important exception on Meta sites: a feature-request I disagree with or a discussion where I disagree with the point of view presented in the question. Upvoting those would, in my opinion, give the wrong signal; here I am intentionally extending the well-known Help Center text about voting on Meta a bit.
On posts tagged feature-request, voting indicates agreement or disagreement with the proposed change rather than just the quality or usefulness of the post itself.
I think I upmod most questions that I answer. If I didn't think it was a good question, I wouldn't waste my time answering it.
I've answered lots of bad questions. I do feel a little funny about answering a question I'm voting to close, though.
I'm very bad about neglecting to upvote a question I answered. I'm working on it though, so don't blame me if I forgot to upvote your question. It's just me being lazy.
Or something.
Purely theoretical: A dedicated badge whore would of course downvote every question that he answers, while also editing the questions to remove important information about context, introducing new spelling mistakes and destroying formatting.
I will go ahead and downvote this question right away. Please help me get that [• Reversal] META badge.
All I need is another 21 down-votes on the question and 20 up-votes on this answer!
The upvote button is titled "This question shows research effort; it is useful and clear".
It's not obvious whether the ";" means "and" or "or", but it is possible to encounter a useful question that is both unclear and showing little research. One can't honestly upvote such a question.
But, regardless of its clarity or the efforts of its poster, a question can still be considered useful if it leads to useful answers.
I might very well answer such a question, believing that my answer will help other people.
Yes, you might say that I, myself, should edit the question to be more clear, or even edit it to add research to the details. But perhaps I don't have sufficient privileges to do that, or perhaps the question has several answers that would become incorrect if the original question were reworded.
Badly worded, poorly researched questions are still capable of generating good answers.
