Notwithstanding that both statsSE and mathSE are both stackexchange sites, and share a similar subject matter, often attended by the same or similar groups of people, I have noticed what I would describe as different 'cultural norms' between life on statsSE and life on mathSE, just as New York is different to Washington. I find this somewhat intriguing and I thought I might enquire if others have noticed the same?


  • New users: The first issue that brought this difference of culture to light to me was the way questions from new users are handled. Over at mathSE, questions from new users are often treated with little respect or encouragement, and even if the question is novel or of research value, such questions are sometimes CLOSED DOWN (inappropriately in my view). By contrast, here at statsSE, my impression is that a new user is encouraged to make changes (if they are needed), that mods are familiar with the subject matter they are modding, and that experienced users try generally to be helpful.

  • Moderators: over here on statsSE, one feels that if one has the need to call on moderators, they are there to protect the interest of users or the pursuit of knowledge, virtue and veritas.

As an example: a few days ago, a new user posed a nice problem to mathSE, which reduced to the following after some clarification:

Let $Z \sim N(0,1)$ and $X \sim N(\mu,\sigma^2)$ be independent random variables. Is it possible to find a closed-form solution for the pdf of $Y=Z X^2$ ?

This question was closed down on mathSE. It might look easy, but it is actually a very difficult problem (I don't even know if a solution exists) and a great little question, even in a simpler form. I raised my concern over the treatment of the new user and the closing of the question ... see, for instance, this thread I posed yesterday:


... and in return, instead of the question being opened up again:

a) the question got deleted (i.e. hide the evidence)

b) the same mods then tried to close the meta question about the question!

This reminds me a little of a quip about the difference between calling in the police (i.e. moderators) in Canada vs the USA. In Canada, if you call the police, they come to keep the public safe. In the US, if you call the police, YOU risk getting shot.

And maybe this is the advantage of statsSE - of staying focused and small - avoiding the pitfalls of growing so large in the number and breadth of questions that the site become de-personalised, with moderators forced to deal with subject matter that they are not familiar with. By contrast, statsSE is like a cool little European town pub.

  • Voting: statsSE users seem much more friendly and convivial in their warmth and friendliness, and willingness to up vote both questions and answers

I am curious if others have noted such differences, or indeed any other cultural norms / differences that have emerged between the sites?

| |
  • 7
    $\begingroup$ What kind of results would you expect to achieve with this post? I mean, what possible outcomes of it could benefit the CV SE site? I understand your frustation with the other site, but fail to see how this post is relevant here. Could you clarify? Thanks. $\endgroup$ – Andre Silva Jul 23 '17 at 14:08
  • $\begingroup$ Hi Andre. The 'discussion' tag is designed to solicit opinions or best-practices on a particular topic: there are different ways of dealing with new users & moderation both here and on other SE sites ... which have given arise to the differences in 'cultural norms' that I refer to. I am interested to see if others agree that there are differences of experience here compared to mathSE, and if they have encountered other differences. It is not my role to seek change per se - but I am interested in understanding the differences between sites, and what causes them, within the scope of the tag $\endgroup$ – wolfies Jul 23 '17 at 14:23
  • 11
    $\begingroup$ All the same, it's something of a loaded question, your presentation of the example could be considered rather partial, & it feels like you're inviting us to start chipping in with our two-pennorth regarding an ongoing dispute on another site. $\endgroup$ – Scortchi - Reinstate Monica Jul 23 '17 at 15:07
  • $\begingroup$ @Scortchi In what way do you feel the presentation of the given example is 'partial'? Can you think of any reason why such a question would be closed down on statsSE? And if you can't, then that is exactly the type of cultural differences I am referring to. In any event, please do constrain your comments to factors relevant to statsSE. $\endgroup$ – wolfies Jul 23 '17 at 15:16
  • 9
    $\begingroup$ I doubt whoever deleted the question would describe their motivation as "to hide the evidence", for a start. For the reasons I gave in my comment above, I don't want to answer this question in its current form. $\endgroup$ – Scortchi - Reinstate Monica Jul 23 '17 at 15:42
  • $\begingroup$ @Scortchi I must agree that they would not describe their motivation as to "hide the evidence". As a moderator, could you perhaps comment what the motivation might be then to delete a newly coined question, of words as stated above, where the question has been nominated for re-opening, and is also the subject of a separate meta question bringing to light colleagues who closed it? What could the motivation be, as a moderator, conceptually, to delete such a question? $\endgroup$ – wolfies Jul 23 '17 at 18:11
  • 7
    $\begingroup$ The question on cultural differences between stats.SE and maths.SE could be interesting, but I agree with @Scortchi: this Q is phrased like a rant and that's not helpful. I am not surprised to see it downvoted already. (By the way, I got interested by your US/Canada remark on police and did some googling. The number of people shot by police in Canada is estimated to be ~20 per year whereas in US the unofficial estimates go up to ~1000, but the number of police officers killed in US per year is ~150 whereas in Canada last year one officer died on duty, and it was an automobile accident.) $\endgroup$ – amoeba Jul 23 '17 at 19:18
  • 4
    $\begingroup$ @wolfies: As a moderator, I assume other users are trying to do the best for the site by their own lights; & if I disagree with their actions try to engage with them constructively, without casting aspersions. $\endgroup$ – Scortchi - Reinstate Monica Jul 23 '17 at 19:41

The OP's post does sound like an invitation to admire our CV-selves in the mirror -but it is totally boring to admire your self, it is preferable, say, to let others admire you. So I downvoted it, and I feel compelled to write an answer here so that people have the opportunity to downvote it -for balance.

It is not difficult to conclude that the atmosphere over at math.se is much more combative than in CV.

My proposed explanation: consider the scientific world a statistician lives in, and the one that a mathematician lives in.

The statistician starts by acknowledging uncertainty, incomplete information, loose ends, you name it, as well as the imperfection and the inherent limits of the tools at his disposal in order to contain it, and therefore that, at the end of his efforts, they will ever remain.

No such thing in the mathematician's world: at the conceptual end of the scientific endeavor, things are either correct and exact, or not (that bits and pieces of failed attempts may be useful and re-used is certain - I am talking about the symbolic final stage). This, psychologically, is close to a "quest for the truth" (an abstract truth perhaps, but truth nevertheless), and every dispassionate observer of history, society and everyday life knows how heated the situation can become when people are "on a quest for the truth" (perhaps rightfully so, but I don't care much for value judgements here).

So philosophically, after all said and done mathematics is more of a binary "yes/no" situation, while statistics, no matter how rigorous and sophisticated mathematical machinery is used, is an "almost/maybe/to a certain degree" situation.

Now compare war and peace: war is much closer to a "yes/no" situation while peace is much closer to an "almost/maybe/to a certain degree" situation. Hence CV is more peaceful, and math.se is more combative.

| |
  • 1
    $\begingroup$ +1 Notwithstanding all that downvoting (aha!), what an elegant and interesting posting. Your observation that mathSE is more combative is spot on (in my experience), and I thoroughly enjoyed your adventure in philos as explanation. $\endgroup$ – wolfies Jul 24 '17 at 20:50
  • $\begingroup$ @wolfies Thank you for your kind words... but there is no downvoting for a comment, neither can I doubly downvote you post, in order to tangibly express my appreciation for the opportunity you gave me to wax philosophical here... I apologize :) $\endgroup$ – Alecos Papadopoulos Jul 24 '17 at 20:57
  • 3
    $\begingroup$ No need - with the help of others, I am hoping to make it to -10 in the near future. $\endgroup$ – wolfies Jul 24 '17 at 21:10
  • 3
    $\begingroup$ Another possibility is that I think it is far more common for statisticians to be collaborators with other researchers, & need to work constructively w/ others & nudge them gently towards a better way. I think that is more rare for mathematicians. $\endgroup$ – gung - Reinstate Monica Jul 25 '17 at 12:13

You must log in to answer this question.

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