15
$\begingroup$

As a student, I find this site to be really useful when I need help with an assignment or a concept I can't understand completely. I can often find what I need in some of the existing questions, but in cases where I can't find a good answer, the existing answers deal with much simpler examples than the one I'm working with or don't explain the matter in enough detail, I need to post a new question myself.

When I do, and I'm asking about an assignment containing tasks which can be solved, I often try to solve them myself and arrive to a result, but I'm not sure about the methods I used to get it. In this case, I have the choice between just posting a question asking someone to solve the task or posting a check my work kind of question. These have several benefits:

  • Users answering the question can get a sense of my level of knowledge
  • I've included everything I can in the question itself, leaving no space for ambiguity
  • I get to practice my $\LaTeX$ by writing down all the steps of my progress, even if I do end up copy-pasting most of it

I can also see a couple of downsides:

  • They are boring: checking someone else's work is most definitely less stimulating than solving the task yourself, and if you check the work by solving the task yourself, you might solve it in a completely different way, limiting its usefulness in checking the validity of the OP's progress
  • They result in awkward answers if the work was done correctly

Still, since they aren't actively discouraged, as I could see from several similar questions I could find here, I've decided that I'm going to post check my work kind of questions when I can. That leaves me with another question, which I feel none of the other similar questions answer very well: should I post my attempt at a solution as an answer or include it in the question?

My thoughts on posting the two parts separately, with some of my doubts and questions in the parentheses:

  • The question itself remains clear, making it simpler to understand for both users trying to answer it and users looking for instructions on how to solve a similar problem (Can be solved in either case with some nice formatting?)
  • The question and the attempt at a solution can be upvoted/downvoted independently of each other (Should they be?)
  • If the attempt contains errors, it might get treated more harshly than it would if it was part of the question(Only if other users ignore the fact it was posted by the same user as the question itself)
  • The question doesn't appear under the unanswered category, providing it with less exposure, making it less likely someone will actually check the work (Do self-answered questions get any kind of special treatment regarding their ranking on question lists?)

What are your thoughts on the points above, and do you think I should post the question and the attempted solution separately or together?

$\endgroup$
  • 6
    $\begingroup$ I'm coming around to the opinion that if it's an assignment question then the way to check your work is to submit it on your assignment. After all, the person marking your assignment is being paid to do that job. If the marking isn't thorough enough, send your work on that question to the lecturer and ask him/her to let you know whether you've done it right. $\endgroup$ – Gerry Myerson Oct 23 '15 at 0:34
  • $\begingroup$ You might have a look at this older post: Best way of asking check my proof questions. And maybe also some other questions tagged (solution-verification). It seems that there is no clear consensus about this. But still you can find some useful advice there. $\endgroup$ – Martin Sleziak Oct 23 '15 at 5:37
  • $\begingroup$ Not so long ago there was again some discussion whether solution/proof verification questions should be allowed at all. See here. I will quote from mixedmath's question (emphasis mine): The meta-tags proof-verification and solution-verification are increasing in popularity. These tags are a bit against the organizational grain of the site. See meta.math.stackexchange.com/questions/6636 and linked meta threads. (While this isn't actually a pressing issue, I conjecture that in time will come back and burninate these tags). $\endgroup$ – Martin Sleziak Oct 23 '15 at 5:43
  • 2
    $\begingroup$ @GerryMyerson I completely agree with you. In my case, the person marking my assignments is the lecturer, and I can even ask him for help without submitting the assignment. While I do have his mail, and he checks it regularly, it often takes him several hours, if not a couple of days, to reply. When he does, considering I'm asking about an assignment, he will often just suggest arranging office hours. On the other hand, when I post a question here, it often gets replied to within minutes, by multiple people, in great detail, and I don't even have to leave my chair. Which one would you prefer? $\endgroup$ – akukas Oct 23 '15 at 14:49
  • 1
    $\begingroup$ It is standard practice here to look unkindly at questions that do not display any attempt by the asker to find a solution. Also, the act of writing and formatting your answer, trying to make it look neat and readable, will often force you to look more critically at what you have done. This is a good thing. Also, the more presentable your attempt is, the more likely it is to be examined. So, Yes, do make your solution part of the question. $\endgroup$ – steven gregory Oct 26 '15 at 12:10
  • 1
    $\begingroup$ You wrote: The question doesn't appear under the unanswered category, That is only true if you accept your own answer or if your answer is upvoted, see here. In fact, if your question only has answer which are not upvoted, it might be bumped by the community user, which means a bit more exposure. $\endgroup$ – Martin Sleziak Oct 28 '15 at 5:51
16
$\begingroup$

I don't think it's a good idea to post your attempt at an answer. If you're not sure about it, then it is more a part of the question. If you figure it out after asking, though, go ahead and post your new solution.

I include my attempt in the question. I start by stating the problem, which I highlight with a > to make it stand out. Then I give my general thoughts about the problem ("I think I should approach it like this but I'm not quite sure about this other thing") and put bullets in front of any additional specific questions that brings up. At the end, I post my attempt below a horizontal rule labelled with "My Attempt." In doing so, I mean to imply that it's okay if you'd rather just post your own solution instead of 'debugging' mine, but that my solution is there if you want it. I also find it to be good practice to outline what I'm going to do in the first paragraph of the proof, in the spirit of the old public speaking adage, "tell them what you're going to say, say it, then tell them what you said." This applies to both MSE question asking and proof writing in general.

One important thing I should add is that for every question I have asked like this, there are about 10 questions that I was going to ask, but figured it out while attempting to write a presentable question. In other words, my primary reason for encouraging that users show their work in homework questions is that I know from experience that it teaches. (Similarly, at least a third of my education was gained from trying to not look stupid while emailing the professor for help.)

$\endgroup$

You must log in to answer this question.

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