What *is* a 'homework question'?

Question

This is going to be a bit of a long post, please bear with me. (And no TL;DR's ಠ_ಠ)

A decaffeinating story

Once upon a meta post, a legend from the realm of Manly soil rose and said: Let's decide on a homework policy. His voice echoed, and other heroes followed. But there were problems$\,\ldots$

Years later, IUPAC¹ reminded us that we don't have as much as consensus as we thought we did.

Current situation

So, basically, back then the legends didn't act on what they should have since they couldn't. We now can though. Let me be less mysterious: These things (i.e. enforced site policies and stuff) should be decided by a community consensus based on evidence on how things turned out. This was impossible in a newborn chem.SE since we didn't know how things may fit well together and we decided to adopt some of the policies in other science SEs and math.SE.

But it's different now. Chem.SE is old enough to have rules that stand on their own. It has grown its own culture, and its own core of avid answerers and moderators (as in people who help clean up the site not mod mods). It has had its own ups and downs².

The question

So out of all of this topsy-turvy, Mart the mod made the best sense:

In general I agree, that there is an issue with the homework policy. I don't think it is done by including waiting periods and/ or new rules. I think we have to rebuild homework guidelines from scratch.

Ooooh, since we still have to see if we agree there's a need for this reconstructing, I'm not asking for that yet. The aim of this post is to gather ideas from y'all on one important question:$$\color{#bb3333}{\mathrm{What} \mathbf{~is~} \mathrm{a~homework~question}?}$$

What should the answer look like?

I wish it was this easy to answer this question. But it's not. Well, of course

Calculate how much weight I will lose trying to understand this Wikipedia page.

is a homework question. But the problem is when you get practical. The way of asking and answering HW questions should be such that benefits the asker and the answerer and stops future unwanted consequences of accepting homework questions.

Practice has always been harder and more exotic than paper. For instance, the difficulties currently surrounding reaching an easy set of rules on HW questions is whether or not hints should be allowed in answers, or how much explanation would suffice.

You can mention one of these things in your answer in order to make it more useful and subjective to us:

• What do you imagine an ideal Q/A of homework to look like? The more detailed you present us with your ideas, the better. For example, should it have lots of comments? Should the answer be complete, or a hint, or a simple link to a tutorial? Should the answer be posted after the question gets comments? etc etc.
• Why should there be policies about homework? I could think of reasons like "Since I don't want to do homework for lazy all-nighters³." or "since I don't want to see them flood the site" (My reason this is, actually) or "since I want the OP to learn thoroughly" etc.
• There's recently a semi-natural boost in the number of homework dumps illegitimately and deceivingly queried low quality posts that do not demonstrate the real question the author has. (What? Blame this guy for that) Do we need something extra to prevent these posts from coming?
• You tell me.

And one final request, $$\Large \text{We'd love to hear all ideas please.}$$

---

_{1: So he is called, for he's one of the first that actually understood IUPAC rules.
2: And some pop-corn-ish dramas. He-who-must-not-be-named, anyone?
3: All-nighter, if you're not familiar with the term, means someone who doesn't sleep the night before the exam because!}

Answer 1

"What is a 'homework question'?" may be the wrong question to ask.

This comes out of the thought that 'homework' is the wrong word for it. "Homework" questions don't need to be actual homework. You can be a 60-something person who hasn't seen the inside of a school for 40+ years and still ask a "homework question".

Better Question: What is it about 'homework questions' that we don't like?

Or more generally, what is it about 'homework questions' which require them to have a special policy?

It's not "homework" per se that we're cautious about, it's more other features of the question which make it a bad one for the site - which is why questions which aren't literally homework are still considered to be "homework questions". The key is not so much figuring out what is "homework", but determining what it actually is that we're against when we're against "homework questions". If we can define that, hopefully the definition of what is a "homework question" (and if there is a better name for it) should fall out naturally.

To that end, let's compile a list of reasons people are against/wary of "homework questions" and a short discussion of each. (I've made this post a community wiki.)

• Answering the question defeats the purpose of the question

  One definition of "homework question" is "any question whose value lies in helping you understand the method by which the question can be solved, rather than getting the answer itself." ref

  The concept here is that because the intellectual value in the question is working it out, any answer that just gives the final result deprives the questioner of the intellectual value of the problem. You can "solve" a hedge maze with a chainsaw, but that defeats the purpose.

  So in this view, "homework questions" are special because the real question is not the literal question posted, but rather "what's the key piece of information that I'm overlooking when working this that would help me solve it on my own"?

• Copy/paste questions

  Questions which are a literal copy-paste from the text book or worksheet. People don't like these because answering is likely a waste of time for everyone. The questioner is simply going to copy-paste the answer back into the worksheet they got it from, without any knowledge transfer. No one gains from the answer being posted.

• Low effort

  These are the questions where someone has written the question in their own words, but hasn't made much effort to solve it on their own. People don't like these for much the same reasons as the copy/paste questions - if someone didn't expend effort on writing the question, how much effort will the spend on understanding the answer?

  On the other hand, if the goal here is to build a database of Q&A that is useful for future visitors, having too much asker-specific stuff in the question is a drawback. In several years, someone coming to the site from a Google search isn't going to care that the OP unsuccessfully attempted to use the Henderson–Hasselbalch equation to solve it - they're only going to care about the best way to solve it.

• Repetitive

  "Homework" questions tend to be ones which are asked frequently, and which are covered in other questions. Answering them again just wastes everyone's time. On the other hand, if it's a duplicate, it should probably be closed as a duplicate, rather than being closed as a "homework" question. But is there a reason people aren't finding these other questions? Is it because the "duplicates" are only duplicates if you already know what the answer is? Should there be a better way of finding the canonical FAQs which answer the repetitive homework questions? Perhaps the "repetitive question" question should be split from the "homework question" question.

• Too complex

  "Homework" questions are often asked by people who don't know what they're doing. So often they may have multiple layers and contradictory information and require complicated answer and a bunch of back-and-forth in order to tease out the key sticking points and formulate an answer that actually addresses the key issue. Perhaps there is space on the web for Socratic dialog, but StackExchange, with its simple question/answer format, is probably not the place.

  On the other hand, often the muddled questions have simple answers: "You're trying to use VSEPR to solve this problem - this is a situation where VSEPR doesn't work." Perhaps instead of bothering with the "homework question" label, we should give them the simple but unsatisfying answer, and tell them to open a new question with a refined approach.

• Too specific

  A "homework question" asker wants the answer to their specific question, tailored to their specific state of knowledge. But the StackExchange philosophy is to make a database of questions that are useful for future visitors. Ultra-specific questions don't really fit the site's purpose. (This goes back to the Socratic dialog issue.)

• "I know it when I see it" (or "A homework question is a question voted to be closed as such by 5 members of the community.")

  While capturing the practicalities of the matter, this doesn't help either editors who wish to learn what the consensus about "homework" questions is, nor the people asking questions - especially those which fall in the not-literally-homework-but-still-"homework" category.

Conclusions?

"Homework questions" have issues that span a range of issues. Having a single "homework question" policy might be oversimplifying things. Perhaps there should be different policies which address each of the how-should-I-solve-this/low effort/repetitive/too complex/too specific issues.

Perhaps there could be a "homework policy", but instead of a separate policy, it's simply a re-iteration of the various other policies, tailored to address people who are literally asking about their homework.

Answer 2

After being awfully quiet about this for a long time, I think it's time to give this another push.

What is a 'homework question'?

I don't know. And I don't want to know. I really think it is the wrong question to ask.
The term itself comes with the perception that it is an assignment, see Wikipedia. Often it is not. It might be part of some self-study, just some exercise from a book, a riddle in a magazine, or whatever else. Things like that are covered with our current policy, but not reflected in the name. This is why one of the common replies is: "This is not homework."
I want to get rid of this. I want to throw anything and everything under the bus.

I agree with R.M. in the point that we need to ask the question more along the lines:

• What kind of question would we prefer to not answer?
• What should be a quality criterion for every question?
• What are the questions that are in general not helpful for future visitors?
• What are the questions that usually get closed due to the homework policy?

We would have to do a lot of data mining to answer the last one. There are more than 500 questions closed as off-topic currently, not including deleted ones (reason 102 off-topic for the query). But this would give us the idea about what question are actually getting closed.

Anyhow, there have been already a few points brought up, that I really want to have included in a new policy:

• What to do about rudimentary questions?
• Policy on questions in basic mediums
• How much effort is "enough" with homework problems? We would like to hear all ideas! (please)
• Homework vs. home-experiment: double standards?
• "Is my calculation correct" is back again
• What should be done with "Is my calculation correct ?"-type questions?

And now, just to set something on fire:

• Should we stop using the homework tag?

---

This has been dumping a lot on the plate so far. And it did not help a nano metre in the discussion. Constructing a new policy from scratch is difficult, we have been aware of this. And the folks that created the site, were aware of this, too. Now the community has aged and we need to address this.

What I am trying to say is, we need to decide, what questions are too broad and what questions are too narrow. When is a question and its corresponding answer helpful to future visitors and when not.

Why do we, as a community even want to close questions that deal with exercises? What could be the value of such questions for the site?
I don't even want to start and talk about the issues that the others have brought up. I want the new policy to be selfish. I don't care if the asker is gaming the system, I don't really care if s/he learns anything from the answer. What I care about is, whether or not the asker has put some effort into his/her question. Whether or not the comments are read, understood, answered. I don't want people to argue with me. I want to know that when I bother to read a question, edit a question, comment on it, maybe even answer it, that I am really helping someone. And the only way I can see that is, if the question has a minimal quality standard.

Wow, that turned into a rant.

---

It will be hard to construct a policy based on something that is not easy to define. It is even harder if we are not all on the same page.

In general I like the idea how the some of the language sites address the problem. They call it general reference, and it requires in principle to include the research you have done. As an example I am using the one from german.se:

This site is about the usage and rules of the German language. It is not well-suited to replace dictionaries, grammar books or similar. If you have already consulted such general references and still have questions, please edit your question to explain what you found and why it did not help. See this post on Meta for more information.

I think we could benefit from something along the lines of we cannot be a substitute for a textbook, please explain where you have troubles in understanding.

Okay, that's enough for now, tell me more in the comments please.

Answer 3

‘What is a “homework question”?’, is definitely the correct thing to ask

But we shouldn’t restrict ourselves to thinking ‘school/uni homework’ while doing so. First, we need to have a clear idea of what our opponent actually is before we can name it the correct thing and finally take appropriate banhammers action.

What we want to do is reduce the amount of questions that are, in the best case, of mediocre quality so that the ‘real’ high-quality questions gain the showcase they deserve. But at the same time, don’t forget that there are good homework questions out there and that these should prominently feature a homework tag so that later users know what to look (out) for. Thus, in an ideal world, we not only understand what we don’t want, but also how we want the stuff we don’t like to be transformed into stuff we like. But let’s get some things straight out of the way first.

homework is always going to be a meta-tag

There is just no way in the world that the final product will be anything but meta. The tag basically means:

This question did not originate from scientific curiosity but aims at understanding concepts.

And maybe this distinction is very important because it can help us to encircle the opponent.

Homework is generally given for, and I think most if not all homework-type questions will reflect one of the following reasons:

• rote memorisation and execution of specific rules (example question: How to name these alkenes according to IUPAC nomenclature?)

• application of previously learnt rules to new situations (example question: We did the oxidation with $\ce{KMnO4}$ in class but in our homework we are to use $\ce{K2Cr2O7}$. What changes?)

• transferring of bits and blobs of unrelated knowledge to access a new domain (example question: If unhybridised carbon were present in methane, what implications would that have for the resulting carbon-hydrogen bonds?)

• writing a report to an experiment of some kind (I’m going to put this separately even though it generally encompasses the former points; see later, why)

Self-study questions are currently included in the homework policy as being homework questions. This is a good thing, since they (if the self-student is a good self-teacher) will also revolve around one of those principles.

How does the classification help us to define ‘homework questions’?

I already hinted above that homework should aim at understanding concepts, which is also the gist of our current homework policy (linked above). This will mean:

• A homework question will build up on previously acquired knowledge in some way.

• It will hopefully grant the asking person access to a new, more general or more accurate concept and maybe also help them understand the principles of past concepts better.

• It will revolve around a specific issue, but the issue is not that specific, and it could, in principle, be asked and answered similarly for every occurrance of said specific issue.

  • To state the last point in a different manner: We could create canonical questions and answers for certain types of homework question and link others to them as duplicates. If there is a high amount of possible duplicates with completely different examples or wording, we are facing a genuine homework question.

• It should never be answered by giving a direct answer alone. At least the working/calculations/methods should also be given, better would be giving reasons why this method is preferrable over another. In many of my exams in later school years including the A-levels, the correct solutions to exam questions were provided so that the examinee could both verify their result to be correct and continue further even if they did not get the correct result — this exemplifies how little the end result helps to understand the true concept behind a homework question.

Maybe I should have rewrittten the final point to read:

• The actual answer does not necessarily help the asker in any way.

What about all the maths questions we’re getting? Are they homework?

Tricky. Some probably are, some definitely aren’t. I think we should express our new homework policy without maths questions in mind and then decide whether they fit in or whether we should handle them separately.

Conclusions

Presented in no particular order.

• homework questions are there to learn. If no effort is involved asking the question, we should close- and banhammer it politely remind the asker that s/he is here to learn and take appropriate action as seen fit.

• Questions that are merely of the rote memorisation type should be closed instantly, unless a full solution is presented by the OP, in which case we can comment/answer:

  Correct!

  (And nothing more.)

• Homework dumps Entire sets of homework copypasted here should be closehammered instantly, reminding the OP of guidelines.

• Questions that can be generalised, because they somehow involve new applications or new concepts, should be left open as long as they adhere to our general and vague ‘high quality’ rule — unless, of course, a duplicate or maybe even a canonical answer to said question already exists. We may choose to do either of the following:

  • Hint towards the solution in comments; or

  • Provide a full and complete solution that can serve as a canonical example as an answer.

  Never limit the answer to the specific case the asker is addressing. Never just present them with a solution but explain it. If you are not comfortable with your answer becoming a canonical example don’t post it!

  Correct answers that do not adhere to the above principles should be commented on to remind the answerers of said principles. If no action is taken they may be downvoted. Do not flag.

  Incorrect answers that adhere to the above principles should be treated like incorrect answers always are: Comment and downvote. Do not flag.

  Incorrect answers that do not adhere to the above principles are a special case. We should attempt to get rid of them quickly. Downvote and comment. Mention in the chat. However: do not flag! Flags are not meant for inaccurate or wrong answers and thus should be declined if they are used to indicate one. (But hands up who is not guilty of clicking recommend deletion in the low quality review queue for a wrong answer.)

• Sometimes, one might not immediately see the issue an asker has with their homework question. This could be a variation of the X-Y problem, where they are already asking about their perceived solution while they should be asking about some earlier step (which they did wrong). But, no worries, such questions should be closed as unclear, explicitly asking the OP to tell us how they arrived at the point where they are. This is probably mostly a problem for the third type of homework as per earlier points.

• Lab reports are very, very likely to not adhere to our high quality rule. As such, they should usually also be closed. However, sometimes a question disguised as a lab report may be one of the good questions noted above which would mean it should stay open. If, however, another good question pops up that does not include the lab report part, the lab report one should be closed as a duplicate of the later one.

• R.M. mentioned a type of question too specific, tailored to the individual OP’s knowledge. This can mean:

  • It is truly a rote memorisation/execution question and should be closed as one; or

  • it is one that is generaliseable. In such cases, it should be answered in a generic manner. It is then the OP’s job to re-reduce it to their specific homework.

• If a homework-like question sounds like it could have stemmed from scientific curiosity and OP explicitly states so, they should be given the benefit of the doubt and the question treated as such. However, that is no guarantee that their question remains open; it could still be a bad one.

I reserve the right to amend to this answer as I see fit.

Answer 4

What is a homework question?

Any question containing an exercise from a textbook (workbook, course-ware) or a blitz copy-pasted is a homework question. The questions are typically very specific and can be answered without resorting to arcane knowledge of science papers.

Issues of answering homework question

• An answer to homework question must not serve as a way to cheat.
• It must help the asking party to improve his skills and/or understanding.
• • Unfortunately, skills and understanding are two different things, requiring separate approach to improve.
• It should not do the work that the asking party can do in reasonable time.

Anticheating

A quarantine of 2-4 hours should be a decent anti-cheat for blitzes and class-works.

The time frame for homework is typically larger, (up to several months) and typically consultations with textbooks are implied. So the issue is not so much as cheating as to not de-stimulate independent work.

Improving skills

A lot of exercises are meant to improve skills. In this case a canonical solution of the kind of work is given in textbook/lecture notes, so providing one in the answer is not a good idea, and requests for that should be closed. On the other hand, asking to check for errors in solving is a legitimate request.

The key point is that skills are acquired by effort of doing something, not from effort of seeing how someone else do somethings. The role of a teacher is to ensure that the solution process is carried without logical errors and possibly to point at possible ways to simplify the task.

In tutoring it is often done by supervising step-by-step solution of a problem. I'm not sure if we can adopt this approach. However, providing solutions (even incomplete ones) is not beneficial for the asking party long-term.

A secondary issue is that to apply skills, one needs to understand concepts lying in the foundation of these skills. If such understanding is not present, it is legitimate to provide an explanation of this concepts. However, doing so may be counterproductive, as textbooks/lecture notes should provide it already, so providing separate explanation degrade asking party's skills to use textbooks.

Improving understanding

There are several issues at work here.

Improving understanding: misconceptions

There is a number of misconceptions taught in school, either not working in wider environment, or simply proven wrong.

A common example is 'extended octet with d-orbital involvement'. It is more or less proven wrong in 1990th, but is still taught at school-level.

Another such concept is hybridization, that may work reasonably for predicting geometry $\ce{BCNO}$ atoms' neighborhood, but fails miserably when one-electron properties are involved.

The issue here is to show clearly where said concepts fail and where they work. Also, arguing with a school teacher about what is proven right or wrong is usually counterproductive, so sometimes it is easier to go with misconception.

I'm not sure what is ethically preferred here.

Improving understanding: blurry areas

There are areas where two or more approaches may be adopted, based on mutually exclusive axioms, but giving qualitatively similar results. One or more of them may be a misconception or oversimplification. Ideally, all approaches should be presented and area of applicability of each of them should be discussed.

For that matter, ideally area of applicability should be discussed for any model used/presented.

Improving understanding: state of art/active research ares

There are areas under intensive research, that simply have not a good, unquestionable model. It should be plainly state so. For example, for a lot of catalytic processes the detailed mechanism is unknown. Unfortunately, such questions may infiltrate school-level...

Another trouble here is that sometimes the proper model is beyond current understanding of the asking party. A good idea would be to list the topics that should be researched before return to this question.

Independent literature search

It is the age of the internet. For a lot of questions the answer can be found using google search. Even if not, textbooks often provide answers as well. Ideally, if the answer can be extracted from several first pages of the google search, the question should be closed with the search query provided. Similarly, concepts extensively described in common textbooks and/or sites should not be described here.