It seems very common for someone to ask a homework question along with some variant of the phrase "I am completely lost" or "I have no idea what to do", and therefore violating the guideline that self-study questions should include some effort on behalf of the student.

What if the student genuinely has no idea how to begin/approach the problem? Obviously this is often the fault of the student for not paying attention in class, but in this instance are we just supposed to move on and let the question die if they don't understand hints?

How are we supposed to answer questions by students who are completely out of their depth?

  • 14
    There is concrete advice about this at Avoid no-clue questions, part of the canonical "How to ask a good question" thread. In general, the student can at least provide the source and motivation of the problem. If it comes from a book, the textbook likely has discussion in the same section on how to solve the problem. – Carl Mummert Oct 3 at 12:55
  • 25
    If the student truly is out of their depth, and cannot even say why a question is interesting or propose any nontrivial way to solve it, I think it may be better to just let the question die. This is particularly the case if the student is unwilling to provide a source where someone else could find the problem. IMO, the best use of this site is to address questions that people encounter when they are engaged with mathematics - not to answer mere mathematical problems that the OP happened to find somewhere and is unable to even begin by themselves. – Carl Mummert Oct 3 at 12:58
  • 1
    I often ask for clarification as to whether the OP needs help with definitions of the terms used in the Question, or whether they can understand the definitions but do not see a ready connection with material for which the exercises are intended to reinforce learning. If no clarification is forthcoming I would recommend backing up and asking a Question which the OP does understand well enough to verify when a correct Answer is given. – hardmath Oct 3 at 15:17
  • 26
    @Isabella The purpose of MSE isn't to help students pass their classes. The StackExchange format is extremely bad for a long-term, personalized study plan which is what someone who genuinely had "no clue" would require. Such people should be seeking help from the people responsible for their education. People who don't care about a topic except to pass a class are often very frustrating and unrewarding to (attempt to) help. Someone with "no clue" about a subject should not be passing a course on that subject. – Derek Elkins Oct 3 at 21:00
  • 13
    @Isabella If a student is taking a course that has nothing to do with their field, and doesn't wan't spend time learning the material, and doesn't have any idea about what's going on... then why should they pass? Passing a course is a measure of competency, and such a student doesn't have it. – T. Bongers Oct 4 at 2:04
  • These past discussions seem also related (at least to some extent): Homework, reasonable to have no clue? and How to prevent “no clue” questions? – Martin Sleziak Oct 4 at 4:57
  • 3
    @DerekElkins I disagree that someone with “no clue” about a question should not be passing their course. Oftentimes I’ve seen professors give (as bonus problems) challenging questions that some bright students will flounder against. Sometimes a question comes down to an obtuse trick, or a new perspective, and isn’t directly tied the results they’ve been taught up to that point. This bright student can still have “no clue” as to how to approach the problem, since their usual tools just don’t work in that context. – Santana Afton Oct 4 at 10:46
  • 4
    @DerekElkins I take your point about how MSE isn't designed to give people the first clue of how to begin solving a question, but two points: 1) the first clue can be (helpfully) given through this site (example). 2) I would caution against reasoning on the basis that a student "should not" pass their course, as with a bit of help, that fact can change. As educators, we are (in principle) supposed to be bringing people up to a standard, not excluding people who don't meet it. – Theo Bendit Oct 5 at 2:28
  • 4
    @TheoBendit All I'm saying is that when a student passes a course they should not (at that time) have "no clue" about the subject of the course. I'm clearly not saying that if a student doesn't have "any clue" on the first day of a course, they shouldn't pass. The whole point of the course is to make them "have a clue". I also don't think that most people asking "no clue" questions on this platform actually have no clue, and I recommend the advice in the answer referenced by Carl Mummert as prompts. – Derek Elkins Oct 5 at 5:23
  • Well, I think one must have some clue if one needs help solving a problem and is asking others for help. Eg I have absolutely no clue what the heck "algebraic geometry" is. Then I would not be asking a direct question here from that subject. Rather I would try to learn about it from books, online material (including past stuff from this website), and if I stumble somewhere I would ask others for help. – Paramanand Singh Oct 5 at 14:54
  • @SantanaAfton: In that situation it would be better to mention the approaches tried. Perhaps one of the approaches might work with a minor modification or one of the approaches is closer to the right approach. Such context often helps the answerer to provide exactly the kind of help needed by the asker. – Paramanand Singh Oct 5 at 15:05
  • 2
    I think this question is contradictory because it claims the user who "has no clue" is a user who "has made no effort". These two do not equate, and I think the issue is a little different. The user claiming to "Have no clue" is claiming to have made an effort, but is unable to demonstrate as much because their efforts have yielded nothing. I haven't checked just now but I don't think making effort is sufficient to comply with policy. I think policy requires that users demonstrate effort. Therefore if they only "claim having no clue" they need to do better and demonstrate it somehow. – Robert Frost Oct 9 at 12:16
  • 1
    By the law of conservation of energy, let us not create something out of nothing. – Alvin Lepik Oct 11 at 6:05

This answer might be contrary to official opinion/community consensus on SE, but I still think that there are a few options.

If you have some energy and willingness to work with what might even be a defensive, frustrated, or unmotivated student, then perhaps the following might be helpful.

  1. In the comments section demand to know the source of the problem, statements of definitions of major terms, and perhaps what is making the problem difficult/challenging/interesting.

  2. Provide in the comments a quick hint ("if $f(x) \in X$ and $X \subseteq Y$, then $f(x) \in Y$. Can you see why this helps?")

  3. Direct OP to similar questions (these "no clue" types are usually duplicates) and ask that they respond as to whether or not the other answers were helpful.


In general, I'd say one of the three methods works roughly $40$ percent of the time. Usually at further solicitation, OP will mumble something (usually wrong) in the comments section, at which point you can ask them to put it in their original question. If that doesn't work, then you gave it your best and you can "let the question die."

If you choose to give a full solution, I think that it's worth mentioning that the solution is probably only useful for future readers. It adds to the collection of problems/solutions on the website. Depending on how highly you estimate the utility of this, it might be worth doing. I've done it if I find the problem interesting for whatever reason and wanted to share my solution. (I'm not a mathematician, but a student, so I don't think that's so evil.)

One thing I like to do is say "Let's see if we can help without giving you the answer, which won't really help anyone." I then suggest that they type in the definition of the relevant terms, and I tell them how to do so, namely "You can add this by clicking on the word 'edit' just below your question." I sometimes also indicate that we'd like this for two reasons: to reinforce their knowledge of the definition, and to let us know what definition they're using.

Today, for instance, someone asked a question about proving something was a binary operation on a group, and when pressed for a definition of "binary operation" said "commutative, associative, closed." I personally suspect that "commutative" wasn't really part of the definition, but it gave a starting point for useful discussion of how to approach such problems.

You must log in to answer this question.

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