See also this earlier post.

An answer that keeps reality in mind and links to various definitions (typically in wikipedia) for terms used and concepts introduced is the very exemplar of . If it doesn’t have original calculations, graphs, etc. I would think that hard-science has a similar meaning to scientific publications, as different from “popular press”.

When does a hard-science tag simply not make sense for the subject (I don’t mean a counterfactual subject being disallowed, but asking for something that can’t be answered by calculation or literature search)? If there's no distinction between science-based and hard-science for non-math answers, should we drop it? Otherwise, how can we clearly state the difference, especially for those who are not familiar with scientific journals and might think that Popular Science is scientific when someone else is thinking of Nature as the relevant example?


3 Answers 3


I think hard-science should need to give references for every claim, but it should not need to require links to scientific articles (which, unless open access, the vast majority of readers won't have access to anyway), although linking to accessible scientific articles is definitely encouraged. Also the hard-science tag implies more rigorous argumentation.

It should be clear that a hard-science answer is not a scientific work in its own right. Probably most people who do the hard work to write an actual scientific article will publish it in a refereed journal or at least at a place like arXiv, instead of posting it on this site where you don't get scientific reputation for your hard work. Put your requirements for a tag too high, and you're almost guaranteed to get zero answers to questions tagged that way.

In the end, what people seek here is information. And the reason to put the hard-science tag on a question is to make sure that the information they get is actually backed by current science.

The requirements in the tag (note the "etc." and that this is clearly not an "and" list, as e.g. for some topics equations simply don't make sense) are to make sure this end is reached, but they are not ends themselves. So an answer should be judged on whether the real end is reached, not by mechanical checking items in a list.

A valid criticism for a hard-science answer would be "this claim is not sufficiently backed by the links/references you gave" or "the link/reference you gave was to a questionable web site", but IMHO not "this links to a reliable source for that information, but unfortunately that reliable source is not a scientific article".

  • 1
    $\begingroup$ Whomever downvoted this answer should provide an explanation as to why as a comment. If you have a specific grievance, it should be documented for the community. $\endgroup$
    – Ranger
    Commented Aug 24, 2016 at 19:28
  • 2
    $\begingroup$ (Disclaimer: I haven't voted on this answer; I'm 50-50) Regarding the statement Put your requirements for a tag too high, and you're almost guaranteed to get zero answers to questions tagged that way: I think this is true insofar as you'll get fewer answers, but as Serban Tanasa wrote here, I'd rather have one really good answer than a lot more lesser-quality ones. At the same time, though, if people are afraid of scaring off potential answerers, then they don't have to use the tag. But that doesn't prevent hard-science answers. $\endgroup$
    – HDE 226868 Mod
    Commented Aug 24, 2016 at 20:56
  • 2
    $\begingroup$ There are plenty of non hard-science questions for people to answer. What's wrong with having a small subset of questions on this site that have higher standards? And there are people who do enjoy answering these hard-science questions: it's not like they aren't getting any answers. $\endgroup$
    – user171
    Commented Aug 25, 2016 at 1:01
  • $\begingroup$ @NexTerren I haven't voted on that answer either. But on meta, votes tend to be "agree/disagree" things, instead of quality like on main. As for the answer itself, You don't have to do original research work to answer a hard-science question. But you should make sure that your answer is in accordance to existing one. Many publications are available to the public. And if it isn't, you might still cite parts of the paper, summaries are almost always freely available. $\endgroup$ Commented Aug 25, 2016 at 7:41
  • $\begingroup$ this links to a reliable source for that information, but unfortunately that reliable source is not a scientific article No one ever said that. The requirement isn't that you can only cite peer reviewed articles. The requirement is that you can't just link to a bunch of definitions from wikipedia and call it a day. And if people to make sure that the information they get is actually backed by current science, then the science-based tag is perfect for that. $\endgroup$
    – user171
    Commented Aug 25, 2016 at 13:21
  • $\begingroup$ @Hamlet: But when the Wikipedia article contains the actual information used, it's not just "linking to definitions", it's linking to the actual information. $\endgroup$
    – celtschk
    Commented Aug 25, 2016 at 15:20
  • 2
    $\begingroup$ Wikipedia is a bad reference source. By all means use wikipedia to get an overview and find the original sources but it should not be used as an reference itself. xkcd.com/978 $\endgroup$
    – Tim B
    Commented Aug 31, 2016 at 8:49
  • 1
    $\begingroup$ @NexTerren "Whomever downvoted this answer should provide an explanation..." but the question is: should that explanation include references? Let's discuss that on the meta meta! $\endgroup$
    – xDaizu
    Commented Sep 2, 2016 at 9:05

Science Based

For questions that require answers based in hard science, not magic or pseudo-science, but do not require scientific citations. Consider alternatively the hard-science and reality-check tags. Avoid using this tag as the only tag on a question.


For questions that require answers to be backed up by equations, empirical evidence, scientific papers, other citations, etc. Review the tag info before using this tag, and flag your question for moderator attention once posted to have the hard-science notice added. Consider alternatively the science-based and reality-check tags. Avoid using this tag as the only tag on a question.

The difference isn't exclusive to mathematical formulas. Hard-Science requires citations (of any sort) and implies a strict level of rigor on proving your claims, arguments, and conclusions. If you claim X or Y is the case, provide a link to a paper, or (as you mention) a mathematical formula to prove your point.

A Science-Based answer might read like an article in the magazine, Popular Science. A Hard-Science answer would not.

  • 3
    $\begingroup$ Agreed. I could see a Psychology based culture question with a hard science tag. Answer to that question would be required to have excerpts/links to literary journals to prove/support their conclusions $\endgroup$
    – James
    Commented Aug 24, 2016 at 20:47
  • $\begingroup$ So hard science questions, if they weren't about Worldbuilding, would be on Skeptics? $\endgroup$
    – user21719
    Commented Aug 30, 2016 at 15:47
  • $\begingroup$ @DonielF Skeptics is for "I heard a claim, is it true?" Instead you might consider the Biology Stack Exchange, Chemistry Stack Exchange, Physics Stack Exchange, or one of the others covering other topics. $\endgroup$
    – Ranger
    Commented Aug 30, 2016 at 16:37
  • $\begingroup$ @Nex I was referring to "backed up by equations, empirical evidence, etc." $\endgroup$
    – user21719
    Commented Aug 30, 2016 at 16:59
  • $\begingroup$ Sorry, no, hard-science answers can be sexy and racy. It's all a matter of style. Just because genuine scientific papers are written in constipated prose doesn't axiomatically imply detailed science answers will follow suit. Think NEW SCIENTIST or SCIENTIFIC AMERICAN (usually not as sexy as NS though). $\endgroup$
    – a4android
    Commented Sep 1, 2016 at 12:35
  • $\begingroup$ New Scientist: EM Drive, Gravity Behaving Badly, … not “science based” at all without editor who understand science. $\endgroup$
    – JDługosz
    Commented Sep 1, 2016 at 13:57

While I have distinct reservations about the term "hard-science", certainly there can be non-mathematical answers that are hard-science. May I suggest hard-science answers about biology, psychology (yes, it was mentioned earlier), chemistry, and geology. This will depend on the nature of the question.

Also, there can and will be answers where the tags of hard-science and science-based overlap. You may draw Venn diagrams to illustrate whatever notion of the overlap of these tags you may hold. The issue of how much mathematical explication is needed for answers will vary. It's easy to see that some answerers will provide equations and calculations for their answers to even science-based questions. After all, how can anybody stop an answer being more appropriate for the hard-science tag when the question was only science-based?

Frankly anyone familiar with classification theory (of which there are several) will know that any labels, categories, classifications, or taxonomies will inevitably be porous and any demarcation between related concepts will be permeable. This means that overlap and similarity between various subject matter are things that just have to be lived with. This doesn't diminish the value of tags, it's simply they don't have absolute value.

Earnest Rutherford may have said: "There is only physics and stamp collecting." This as about as absolutist as it gets (except he was probably joking and Rutherford was a New Zealander and New Zealanders have a funny sense of humour). While physics is awarded the hard-science, it's not the only game in town (I say this as someone who loves physics, but I'm not blind).

Simply expect hard-science answers to be more rigorous than science-based ones, but do not expect them to put in exclusively separate boxes. There will always be cross-overs. Tags are more like pointers showing which way the subject should be treated. There can't be absolutes.

  • $\begingroup$ I'd agree with everything said here, especially "Tags are more like pointers showing which way the subject should be treated." To that end while they may overlap tagging a question as both seems superfluous. The author should use the tag that best points in the general direction he wants the question treated as. $\endgroup$
    – Ranger
    Commented Sep 1, 2016 at 14:19

You must log in to answer this question.

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