# Why do Cross Validated and Math SE have different policies for homework questions?

Why is it that CrossValidated requires homework or the like to be tagged w/ , but Math SE did away w/ homework tag a long time ago? Why don't the arguments for no more homework tag in Math SE apply to CV? Why not treat all questions on CV the way self-study questions are treated?

Math SE arguments for not having homework tag:

• What are the arguments for not having a homework tag in Math SE? Dec 2, 2015 at 14:00
• @Scortchi Edited ^-^
– BCLC
Dec 2, 2015 at 14:21
• @Scortchi: see here meta.math.stackexchange.com/questions/14981. Some arguments are compelling but seem to stem from the fact that homework questions are much more widespread on math.SE than on CV, so here we simply do not seem to have many of the issues raised there (that might be part of the answer to BCLC's question). However, it's not very clear to me how math.SE deals with homework questions now that they do not have the homework tag. Dec 2, 2015 at 14:21
• @amoeba I don't really see why self-study and non-self-study questions should be treated differently. Why can't we just pretend everything is self-study? Askers give effort. Answerers give prompts. Self-study or not.
– BCLC
Dec 2, 2015 at 14:22
• Okay, but is it the current policy of math.SE too (apart from being your personal opinion)? Or are they still trying to restrict themselves to only providing hints if the question smells like homework? Dec 2, 2015 at 14:24
• @amoeba afaik, math SE treats homework and non-homework questions the same. If the user shows little to no effort, in general, don't give a full answer. If the user has shown sufficient effort, answer as you see fit. I don't see how this doesn't apply to every question on SE, homework or not
– BCLC
Dec 2, 2015 at 14:25
• I see. I don't have any opinion on that but I think this discussion is very welcome and am looking forward to it. But I wanted to share my major quibble with our [self-study] tag, and that is that many (most?) of the questions tagged with it are actually not at all homework. I have now looked at the [self-study] questions ordered by vote count and scrolled down the first several dozens; almost none of them is homework! Something is very wrong with this tag. Dec 2, 2015 at 14:49
• @amoeba self-study is not a different word for "homework" but a broader class of question that includes homework. Dec 2, 2015 at 15:00
• @Glen_b, yes, sure. The tag wiki excerpt says "A routine question from a textbook, course, or test used for a class or self-study", and that's what I think it is. But I still claim that from the first several dozens of the highest upvoted [self-study] questions almost none is a routine question from a textbook. This, however, might only be true for the highest upvoted threads; I do not know. Dec 2, 2015 at 15:02
• @amoeba, some self-study posts may not be homework, because the tag is self-study, not homework. That means self-study is broader than homework. If I want to start learning on my own I can ask here and label the post self-study in order to not receive a full response, or to show fellow users I am studying this for 'self-learning' (It is not a homework received from someone else). Now if this policy works, or is misused is another story. I don't know how homework word(ed) in Math, but the discussion here should consider that conceptually 'self-study' is different from 'homework'. Dec 2, 2015 at 15:03
• @amoeba some people who are trying to solve a problem themselves (i.e. undertaking 'self-study' add it because they're seeking the kind of answer the tag requires. I think it's reasonable to use the tag that way. The excerpt won't necessarily describe every appropriate use of the tag Dec 2, 2015 at 15:03
• @Glen_b: This is fair enough but I would simply encourage to you look at the list and inspect the questions yourself. "How to interpret a QQ plot" and "Pitfalls in time series analysis" (two highest upvoted questions) is just not something I would call a routine textbook question or a problem that people can be trying to solve themselves. If that's self-study then everything is self-study. But this is a bit of an aside to the topic raised by OP. Dec 2, 2015 at 15:06
• @amoeba It's quite possible for some posters to use the tag incorrectly (that happens with any tag); since most legitimate self study questions get few votes it's hardly surprising that the highest voted ones might tend to be mislabelled ones. That's not a problem with the tag itself. Dec 2, 2015 at 15:11

Statistical questions are unlike mathematical questions. On the math site, a question will quickly be closed if it is not immediately clear what it means and what it is asking. All other questions on that site might as well be textbook questions--there is really no basis for distinction, not even the level of sophistication.

Our community has learned to work within the SE framework to help people articulate the questions they come with. These questions typically are vague, ill-formed, use non-technical language, and often are laden with inherent contradictions and misunderstandings. It is remarkable that we can elicit an answerable question from most of these posts--and that makes our site uniquely valuable.

In this context, a question that is quoted from a textbook, test, or homework set is not the norm, for two salient reasons:

1. It usually has been clearly and precisely framed in technical language. It requires little or no context to understand and often needs only routine application of purely mathematical operations or elementary statistical thinking for its solution--all of which likely have been illustrated many hundreds or thousands of times in similar threads that vary only in minor details.

2. It is almost never a real world problem, which is specifically the kind of problem people are invited to pose on this site. The two-minute tour states:

Get answers to practical, detailed questions. Focus on questions about an actual problem you have faced.

The second point is dispositive: it suggests that the artificial, unrealistic, naive kinds of questions typically posed as homework and tests really don't belong here at all. Perhaps we should redirect them all to the math site, but because we share an interest in the statistical interpretation, we not only tolerate them here, we even welcome them when they invite clear, illuminating, clever, and well-articulated solutions.

• These are very good points, but we do have a relatively vibrant [mathematical-statistics] tag with around 1.5k questions (and there are many more questions where this tag could in principle be applied too; it's not used very consistently) that are certainly considered on-topic, and that probably are closer to the math.SE modus operandi. So even though textbook exercises are indeed not the "norm" among the bulk of the questions here, they seem to be pretty close in spirit to a substantial part of questions that are on-topic. Dec 2, 2015 at 19:59
• @amoeba I am having trouble equating questions about computing Normal probabilities, using Venn diagrams, calculating with logarithms, etc, with any form of "mathematical statistics."
– whuber Mod
Dec 2, 2015 at 20:38
• Sure, but I imagine that folks at math.SE also enjoy answering questions on differential geometry or Galois theory more than about how to solve quadratic equations or transform some sines and cosines... In fact, I took our [self-study] policy for granted until this thread came up today and was surprised to see that math.SE decided to deprecate [homework] tag. But this did make me wondering if we really need it too. I am not disagreeing with your point of view, by the way; I don't really know. Dec 2, 2015 at 21:17
• @amoeba I understand. Although I have not followed the math.SE debate, I suspect one key issue would have been where to draw the line: is a basic Galois theory question "homework"? What about a question of advanced calculus? Etc. On CV I think we have a clearer criterion to apply. In some sense, I think of most math-like questions--at least those with known answers or answers that are readily developed--as being of the "self study" type and the rest of the questions here as not being self-study because they deal with actual applications--something that is irrelevant to a pure mathematician.
– whuber Mod
Dec 2, 2015 at 22:37
• This is an interesting point of view. Almost all questions that I ever answer here (about the math of PCA, LDA, and related methods) are self-study under this broad definition. However, we are not tagging them with [self-study] tag and not trying to provide "helpful hints" only, but instead readily go into all necessary mathematical details. So I agree that there is this clear criterion, but we are not using this criterion to assign [self-study] tag and to use associated with it policies. That I think is the crux of the discussion. Dec 2, 2015 at 22:46

Why not treat all questions on CV the way self-study questions are treated?

People come to CV to get real help on real problems, and they often succeed. I don't see why we should stop providing this help and instead start just giving hints and observing how the asker is progressing with the help of the hints (which is how I understand your suggestion of treating all questions as self-study). It would take much more time for those who are answering and would also take valuable time from those who did not come here to dig deep into statistics but who just need an expert's advice.

E.g. if I am writing an academic paper and just need a quick clarification of some technique that is not central to my topic, should I spend all the time being educated hint-by-hint how this technique really works? If I had infinite time, then it might be worth it, but everyone has to prioritize.

I strongly prefer the system as it is working now.

• If the topic is that advanced, then give an answer right away. If the topic is relatively basic, give hints. Why is a self-study tag is needed?
– BCLC
Dec 4, 2015 at 13:49
• I say, give hints for self-study questions but give straight answers to all the rest (regardless of how advanced). What is basic for you may not be basic for me, and I might not have the time for playing a teacher and a student, so to speak. Dec 4, 2015 at 13:54
• @BCLC "Simple" and "advanced" is a poor distinction in this context. A homework question that's done as preparation for an exam at graduate level may be quite advanced, but is still basically bookwork (even if it's bookwork done at a high standard), and this isn't a site to do people's homework for them. On the other hand a conceptual question might be asked at a very basic, pre-university level, but merit a clear and didactic answer rather than a hint-dropping conversation. Dec 4, 2015 at 13:55
• @Silverfish 'this isn't a site to do people's homework for them' but Math SE is? :|
– BCLC
Dec 4, 2015 at 15:11
• @BCLC Maths SE is whatever its users want it to be. But to a greater extent than on Stats SE, question-askers there seem to want to ask homework questions, and answerers there seem quite happy to provide complete solutions (often with an emphasis on the "neatness", "elegance" or speed). Relatively few questions on Maths SE come from issues users have encountered in the real world, and that proportion is higher on Stats SE, so it is unsurprising the cultures are quite different on the two sites. Dec 4, 2015 at 15:35
• @Silverfish why don't you post that as an answer?
– BCLC
Feb 13, 2021 at 23:03