11
$\begingroup$

I am just wondering why it takes more time to get a response in Cross Validated than in math.SE.

Is there any specific reason?

$\endgroup$
  • 15
    $\begingroup$ Statistics is harder. ;P $\endgroup$ – Alexis Apr 16 at 16:38
  • 1
    $\begingroup$ not having the time to write a full answer, I don't think math.SE has as many "How do I analyze images in Python?" like question as stackoverflow. $\endgroup$ – Cliff AB Apr 25 at 15:22
27
$\begingroup$

Math.se has more site traffic than Stats.SE.

https://stackexchange.com/sites?view=list#traffic

Math.SE has much more site traffic. Moderator @whuber notes that depending on the statistic, the ratio of the sites' traffic statistics can vary between 3:1 and 11:1.

The sites' topics are very different

Math.SE tends to attract questions which are more-or-less help with routine coursework. As long as the notation is clear, answers are well-understood. The more open-ended questions tend toward questions about intuition, or how certain mathematical concepts fit in with others (Borel measurability versus Lebesgue measurability), or why some object ($\sigma$-fields) or requirement ("measurability") is necessary for a result.

Stats.SE covers two domains. One domain is answering questions about probability and statistics as a matter of mathematical study, such as showing that OLS coefficients are BLUE or properties of matrix arithmetic or proving results from probability theory. These questions have the most similarity to Math.SE questions.

The other domain is applied statistics. Applications are messy. Subjects of study are not always self-consistent, data collection can be incomplete, and experimental design can leave room for endogenous effects or confounding. It's challenging to write a question which concisely explains the task and asks an answerable question about how to carry it out. It's also hard to write an answer that understands the problem completely and presents the correct answer, especially when the question is ambiguous or its explanation is incomplete.

Some questions don't have clear-cut answers, so a good answer will contrast the deficits of several alternative approaches.

As someone who writes more answers than questions, I've found that I am susceptible to a particular trap: writing answers as if OP was asking a routine or textbook question. Sometimes OP is asking a routine question, but sometimes they're not. Asking for clarification (which may not ever arrive) is an important part of answering questions, as is imagining the myriad of different circumstances in which the question could arise and attempting to pin down which one matches OP's needs.

We have some other threads that are related to this

I'm not aware of an exact duplicate of this specific question, comparing Math.SE and Stats.SE, but we have discussed the Stats.SE answer rate before.

$\endgroup$
  • 12
    $\begingroup$ +1. The ratio actually is between 3:1 and 11:1, depending on the statistic. It may be worth remarking that the great majority of questions on the math site are routine and require little time to answer, while here it generally takes much more work to get most questions even framed in an answerable form. $\endgroup$ – whuber Apr 13 at 17:27
  • 2
    $\begingroup$ @whuber I hope that my revision is not fatuous. $\endgroup$ – Reinstate Monica Apr 13 at 17:43
  • 3
    $\begingroup$ Not at all: it's informative and revealing. Thank you. $\endgroup$ – whuber Apr 13 at 22:38

You must log in to answer this question.

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