It's long overdue that I make this post revisiting our policy on "check-my-work" questions. These are questions, often (but not necessarily) homework-like, that present a complete mathematical or logical derivation and ask whether it's correct.

Historically our homework policy has rendered check-my-work questions off topic. However, we're currently in the (very long) process of revisiting that policy. According to the poll on homework close reasons (answer 1, answer 2), the community seems to be generally in favor of keeping such questions off topic, although not obviously so.

It was brought up in a chat session a while ago that perhaps people would support making only certain kinds of check-my-work questions on topic. When deciding whether a given check-my-work question is on topic, we might take into account factors such as

  • the level of the question (basic or advanced)
  • how much prior research the asker has demonstrated (have they tried asking colleagues?)
  • whether they have a reason to believe they have made a mistake at all
  • whether they have an idea of what the mistake might be

and others to be suggested.

So the point of this question is to resolve the issue for whenever our homework policy gets revamped.

Should any kinds of check-my-work questions be on topic?
If so, how do we distinguish the on-topic ones from the off-topic ones?

Return to FAQ index


Check my work question should always be off-topic. Those that can be rephrased should be rephrased. "Am I right?", "Is this correct?" or something else is always only of use to people who did the exact same derivation, and this is definitely too localized.

To answer the bullet points in order:

  • The level of the question should be utterly irrelevant. I have no more desire to correct a multiplication error in a basic kinematics derivation that I have to hunt sign errors in an advanced QFT calculation. Aside from my desire, neither will be of use to people who have not committed the exact same error, so it is definitely too localized either way.

  • Asking colleagues or anything else like that is nothing we could read. The close-worthiness of a question must not be influenced by a statement like "I asked my prof and he didn't know either" since we have no way to know whether that is even true or not, or whether that prof even should have known, etc. Prior research that I would expect is finding out the correct answer (preferably with derivation) or, if that cannot/has not been done, finding similar problems or techniques and briefly explaining why they are of no use in this case.

  • This is crucial. If it is simply "Am I right?", a probably full and complete answer is "Yes.", which is too short to even submit it as an answer, which shows unequivocally that we as a SE do not want such questions. Asking "Where's the mistake?" is better, but...

  • The only kind of "check my work" I think we should allow is the one where a derivation is presented, leading to a wrong result, and the question is "It seems as if step X is wrong? But it should be right because of Y, so why is this not the case?". There must be a reasonable explanation (by established physics, of course) of why the derivation is expected to work in the eye of the asker, and then the answer pointing out the flaw in the reasoning can actually be useful, since the question is then essentially "Why is the physical principle Y not applicable here?" The question should also be edited to reflect that.

Note that it is still possible that questions of the latter type are essentially based on a sign error or somesuch, because we all are sometimes blind. Point it out in a comment, VTC the question (or not, that must probably be left to individual judgement, but that's what the close/reopen queues are for), and move on.

  • 7
    $\begingroup$ Good argument. I would consider the kind of question mentioned in bullet point #4 to be a conceptual question, not a check-my-work question. $\endgroup$
    – David Z
    Aug 28 '14 at 16:55
  • 3
    $\begingroup$ I like the "why is principle X not applicable here" argument: we are not looking to find the missing factor 2 in the derivation, we want to explain why there is a term missing in the equation... $\endgroup$
    – Floris
    Aug 29 '14 at 14:43

I know the stated goal of the site is to provide a knowledge base, and I know that every time I rant on about the importance of teaching I get told that isn't the main purpose of the site.

Nevertheless I think teaching is important, and doing worked examples is an important part of teaching. I'm not saying I want the site to fill up with posts asking how to do a worked example, but I think answering such questions can illustrate important concepts.

A blatant "spot my deliberate mistake" question is always going to be off topic, but I would urge site members to consider whether the mistake is conceptual rather than arithmetical. If it's a conceptual mistake I think there is some justification for answering the question as long as your answer is mainly concerned with the concepts involved.

  • 1
    $\begingroup$ Hi John, I can't tell if you are advocating some "check my work" questions, or if you're advocating some "check my concepts" or something else. I agree with your answer but I can't figure out how it translates to site policy. $\endgroup$ Sep 6 '14 at 7:08
  • 7
    $\begingroup$ @BrandonEnright: I think site policy needs to be flexible. My view is that if your answer explores some interesting physical concepts then the question is OK. If the answer is "you missed a minus sign on line 57" then the question should be closed on the grounds. $\endgroup$ Sep 6 '14 at 7:26
  • 3
    $\begingroup$ "I know that every time I rant on about the importance of teaching I get told that isn't the main purpose of the site." The funny thing is that it is the purpose of the site. It's just that, as experts, we forget that when we discuss things with each other and answer one another's question we are teaching one another. That being the case, and in full agreement with you that worked examples are a fantastic teaching method, I agree that examples should be allowed and I further think that they should be used liberally in answering even purely conceptual questions. $\endgroup$
    – DanielSank
    Dec 28 '14 at 6:37
  • $\begingroup$ Late to this party, but toward @Daniel's comment the pedagogical literature suggests that worked examples are a surprisingly poor teaching method for all but the most diligent of students: it is the working of examples that teaches and most people only read prepared worked examples (there may have been point in your education where the way you read textbooks changed so that you got more out of reading after the change; there certainly was on in mine). Perhaps visitors to a stack exchange site are biased toward the exceptional group, but I can't think of any way to test that hypothesis. $\endgroup$ Sep 4 '17 at 18:50

I'll offer up my other view on this since I am undecided in some respects.

Checking of mathematical derivations only should always be off topic. We're not TA's here, we shouldn't need to hold office hours for people.

Having said that, when people ask a question that really can be rephrased as "Did I include all of the relevant physics?" then I think it's okay. In that sense, even basic questions like the equation of motion for a mass-spring system could be verified -- does gravity matter in the model for example. Obviously higher-level questions would fit into this naturally as they involve more physical concepts that need to be included.

In this way, we aren't separating based on "high" or "low" level, which I strongly object to, because these are entirely subjective to the background. I've said it previously -- I never encountered Lagrangians until grad school so in my background, even "basic" questions about deriving one is "graduate" level.

Instead, we are deciding based on whether we are checking that the relevant physics is included in the work and we go no further. If all the physics is present but the derivation made a mistake/sign error/math error along the way, take it to office hours.

  • $\begingroup$ Agree to subjectiveness of "high" and "low" level. You may never have seen lagrangians until grad school but I never saw maxwell's equations until grad school (nor the principle of least action, nor heard of the term "metric". My supervisor had his hands full, to say the least). Would somebody like to try to top that? $\endgroup$
    – Jim
    Aug 28 '14 at 13:56
  • $\begingroup$ I tend to believe that questions which effectively ask "Did I include all the relevant physics?" are too broad (because really, does anyone ever include all the relevant physics? what does "relevant" even mean?). But questions asking "How large of a contribution does [physical effect] make to this system?" or, in most cases, "What effect could account for this amount of inaccuracy?" I'm okay with. $\endgroup$
    – David Z
    Aug 28 '14 at 21:57

"Check my work" type questions should remain off topic for two reasons:

  1. They do not ask for clarification of physics concepts, they ask for clarification on the mathematical solution to a particular problem

    • In the "How do I as homework question" post, it states that Physics.StackExchange

      is not primarily a homework help site; it's a place to get specific conceptual physics questions answered.

  2. The question will likely only help the original poster and not future visitors to the site

    • This is likely more true for the more advanced topics.

It is likely that some of these "Check my work" questions can be reformulated to be an on-topic question, e.g. reworded so as to fit the "I was working on X and didn't understand why Y isn't the case" criterion. However, note that changing the question ourselves would constitute a drastic change in the intent/meaning of the question, which should be left for the OP to do.


Let me quote part of the homework-like close reason... Ahem...

We want our questions to be useful to the broader community, and to future users.

Powerful words. So when it comes to checking someone's work, most of the time the question and answer will be specific to that user and not very useful to the broader community. This includes all levels of questions, research-level too. Little Susie's question about what mistake she might have made in question 16.8 of Scherrer's textbook might be off-topic for more reasons than Billy's question about where he went wrong in deriving the equation of state parameter for a universe filled with dark matter that is weakly coupled to gluons and photons, but the fact that they are both too specific to be at all useful to the broader community or future users makes both of them equally off-topic (it's really a binary state, thus "equally").

That said, I did say "most of the time". Based on that phrasing, it stands to reason that some of the time these questions are on-topic. And that's where it get's into a grey area. There's really no easy way of making a catch-all statement for when they are on-topic. If we feel that the question describes a commonly seen method and that the mistake is one that we might expect a lot of people to run into, then I think it is useful to the broader community and that we should give an appropriate answer to what they've done wrong at least once (that is assuming, of course, that the question isn't off-topic or closeable for some reason other than its specificity). But we should make sure to reword the question such that anytime someone else comes along having run into the same problem, we can link to that question as a duplicate (obviously).


You must log in to answer this question.

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