Is there any way to filter questions by a certain number of subjects (e.g., real analysis and differential equations)? Additionally, is there a way to filter the questions by difficulty?

The latter question, I feel, is more important. I'd like to participate more in Mathematics SE by answering more questions, but far too often I encounter questions in areas I haven't studied or are at a much more advanced level, making the entire process of questions somewhat difficult.

I propose that questions could be labelled as "high school", "undergraduate", or "graduate". (This would not be the same as a tag, however, since tags aren't always comprehensive by nature.) Filtering questions this way would allow people to find questions that are more suitable to their level of understanding, and potentially mean more questions could be answered. So, for example, if I were to post a question, I would tag it with differential equations, and click one of three boxes labelled as "high school", "undergraduate", or "graduate".

I'm not suggesting this is the only solution; my intention is just to find a way to contribute more. From my experience, most people who answer questions are professors or Ph.D candidates.

What do you guys think?

  • 2
    $\begingroup$ This has been discussed, but without a solution that would satisfy you: meta.math.stackexchange.com/questions/7039/… $\endgroup$ – user53153 Feb 4 '13 at 20:15
  • 2
    $\begingroup$ I wonder how the message "This question may already have an answer here" got there. There's no edit history. Divine intervention (cf. virgin birth)? Well, the question may have an answer there, but as a matter of fact, it does not. $\endgroup$ – user53153 Feb 6 '13 at 4:33
  • $\begingroup$ An older similar discussion: Sophistication level tags? $\endgroup$ – Martin Sleziak Apr 27 '13 at 6:50

I'm not opposed to the idea, but fail to see how the specifically mentioned advantage would be true and important. It is suggested, in my words, as a shield from questions at a higher level of a potential replier's educational background. How many lower level questions remain unanswered at the end of any given day, and how long does it take to browse those among them with no answers? Not long. I could see how people with very high mathematical ability might be more interested in more challenging questions; but from informal observation, those who are good typically, and more often than not, appear to enjoy to also work on easier questions (be that just for reputation gain; or from the pleasure apparent in some people's answers to mentor online). That's not to mention mathoverflow for the true experts, who probably enjoy time on MSE to relax. 

That is in addition to what has been remarked before: the rather significant differences in educational background when you strip this from a US-centric focus. My Polish math homework buddy in Germany had encountered induction in 9th grade; and a German Vordiplom covers (in the old days at least) a U.S. math major background. As another example, the French system is, for someone not intimately familiar with it, rather arcane (how even to compare an universite with an ecole d'ingenieur, within the system, in terms of 'level' of a topic? A course in a (pre-'college') class prepa might be what you see in the U.S. at a senior level)..

  • 1
    $\begingroup$ +1 for pointing out that not everything in the English speaking world revolves around the Unites States of America. $\endgroup$ – Asaf Karagila Feb 5 '13 at 0:54
  • $\begingroup$ "from informal observation, those who are good typically, and more often than not, appear to enjoy to also work on easier questions". Observation bias at its finest. The experts who do not enjoy digging through the steaming pile of homework tend to not participate in MSE. As a result, graduate-level questions they could have answered sit around with no answers, no comments, and 15-20 views. $\endgroup$ – user53153 Feb 5 '13 at 1:10
  • $\begingroup$ @5pm: certainly a valid point for some, and - from your comment - probably you. I also have noticed names answering category theory one day (say), and pre-calculus the next. I guess the vote count on the question will give an (informal :)) hint about the relative size. I just chipped in what came to my mind. $\endgroup$ – gnometorule Feb 5 '13 at 4:10
  • 1
    $\begingroup$ @5PM Adding to this is the issue that frequently short answers to simple problems receive far more votes than long answers to difficult problems. Creating a negative incentive to work on the more difficult problems, depending of course on how much reputation is a motivator. $\endgroup$ – JSchlather Feb 5 '13 at 7:33
  • 2
    $\begingroup$ @Jacob, 5PM: It seems to me like a complaint of a McDonalds executive "Why do people who like quality meat don't come more often? Sure, some of them pop in for a bite, but not many actually have their meals here...", as history can no doubt teach us -- you can't be both popular with the masses and the elite. If you open your gates for the masses to post homework questions and vote, it's no wonder that short and palatable answers are highly voted, and the Batman question is highly voted; and it's no wonder that many experts would avoid the site not wanting to wade through bogs of homework. $\endgroup$ – Asaf Karagila Feb 5 '13 at 9:05
  • 2
    $\begingroup$ @Asaf But if they had direct access to advanced questions (e.g., RSS feed for a tag designating questions at certain level) they would not have to wade through homework/batwork, or even see it. $\endgroup$ – user53153 Feb 5 '13 at 11:48
  • 1
    $\begingroup$ @AsafKaragila By the way, my impression is that the extremely popular questions on Math.SE were upvoted more by established SE users outside of Math (led there via news aggregators and the SuperCollider) than by the homework-motivated transient users of Math.SE. E.g., 77 upvotes on I registered to math.stackexchange just to vote for this wonderful answer! (which is a deserving answer of course). $\endgroup$ – user53153 Feb 5 '13 at 16:27
  • $\begingroup$ @AsafKaragila I suppose it's mostly a personal issue for myself, since I'm somewhere in my career where most questions on mathotherflow are above me and most of the questions here that I'm interested in don't garner a lot of attention. Perhaps when I get out my early on adolescent-graduate student phase things will even out. $\endgroup$ – JSchlather Feb 5 '13 at 22:44

Variants of this idea have been used on several SE sites, the most often used is the tag, while not exactly a difficulty tag generally marks questions that are easier.

There are several fundamental problems with the idea of using tags to mark the difficulty of a question on an SE site, and the use of such tags is discouraged across the SE network.

The tag system isn't meant for this kind of tag. Meta tags, that are about the type of question, and do not describe the subject itself don't work well within the tag system that currently exists.

The biggest problem is that classifying question according to difficulty is hard and subjective. If you ask different users, with often very different educational backgrounds, they will often disagree on the exact classification. Or imagine that you retag a question by another user as , and that user considered his problem to be much harder and is offended by this classification. This adds a significant potential for unnecessary conflict.

Another aspect that makes this kind of classification less practical is that the higher you go with the difficulty, the more it depends on your specific background and not only your general level of education. If you are a professor, you still likely won't be able to answer research-level questions outside your field of expertise, so a purely level-based categorization would also be not enough.

I don't think such a classification can work within the existing framework, there are just no appropriate tools available. This is a hard problem, and while it would be great to be able to select question that fit the own expertise, I don't know how this problem could be solved in a really satisfying way.

  • 2
    $\begingroup$ "The biggest problem" would disappear if the tags secondary-ed, undergraduate, postgraduate described the level of studies at which the question was encountered. There would be nothing to argue about, or to get offended by. (Questions not related to curricular studies would simply not have any such tags.) // The paragraph "this kind of classification... would also be not enough" looks like a statement that tagging by level is not a panacea. Of course it is not; nothing is. $\endgroup$ – user53153 Feb 4 '13 at 20:27
  • 5
    $\begingroup$ @5PM: That asssumes that which level a given topic must be treated at is a God-given truth that is inherent in the topic and which everybody has access to. In practice, this assumption is quickly falsified simply by observing that different educational systems arrange different subjects in different ways. $\endgroup$ – Henning Makholm Feb 4 '13 at 20:57
  • 3
    $\begingroup$ It would be easy (and very useful) to classify as: grade-school, high-school, undergraduate, graduate, or research level, possibly with modifiers such as beginner, intermediate, advanced. $\endgroup$ – Math Gems Feb 4 '13 at 22:27
  • 2
    $\begingroup$ @HenningMakholm I don't assume that, and I did not suggest that we try to infer the level from a topic. What I wrote was "the level of studies at which the question was encountered [by the OP]". If the OP does not tag by the level, that's fine. If they do, they increase the chances of the question being noticed by a person with particular interest in questions at that level. Level tags don't have to be in perfect correlation to the difficulty in order to be useful. $\endgroup$ – user53153 Feb 4 '13 at 22:33

You must log in to answer this question.

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