Update: Based on the lively discussion here, there is now a featured meta post where you can propose specific changes.

When users ask their first question, they now are shown the following information (to see this, log out or open a private browser window and go to https://chemistry.stackexchange.com/questions/ask):

enter image description here

To give an example of what another site did, below is the modified information from Code Golf, which they requested after discussing it on their meta.

enter image description here

Physics mentions homework-like questions not on the "first-question" prompt but always as step 1 of asking a question:

enter image description here

In contrast, our step 1 directions are much shorter and without links (maybe some of the physics language could go into the "first-question" box, or into the step 1 side-bar):

enter image description here

How could we customize our prompt (or anything in the help center) to address some of the typical shortcomings of first questions (and how could we test whether it had an impact)?

  • $\begingroup$ A customized prompt, if effective, could reduce the need for guiding comments after the question was asked, chemistry.meta.stackexchange.com/q/5205 $\endgroup$
    – Karsten Mod
    Aug 20, 2022 at 11:24
  • $\begingroup$ "Typical shortcomings" in first questions might include asking questions already answered on the site, lack of chemistry formatting, lack of cited sources, not asking a question when posting a textbook/exam problem, screenshots with text instead of writing the text, asking a trivial question the OP could answer themselves. $\endgroup$
    – Karsten Mod
    Aug 20, 2022 at 11:28
  • 3
    $\begingroup$ I guess we could use similar bold part, but about homework. $\endgroup$
    – Mithoron
    Aug 20, 2022 at 15:32
  • $\begingroup$ Being very honest, people whose questions are affected by the change won't change much. 3. When appropriate, describe what you've tried $\endgroup$ Aug 21, 2022 at 9:28
  • $\begingroup$ @SafdarFaisal I agree it's not likely to be a huge difference, but it's a pretty low-effort change, so even if the benefit is small I don't see why not to do it. $\endgroup$ Aug 21, 2022 at 11:09

2 Answers 2


I guess, given that we're starting from totally generic text, simply adding links to the appropriate page(s) in the help centre would already be a net positive.

Beyond that, it's a question of how efficient we can be with words.

I still really, really dislike the homework policy, but bearing in mind that this screen is only shown to first time posters---if that's what's most likely to be relevant to them, then adding a note about that would make sense.

MathJax formatting would be the other one. FAQ: How can I format math/chemistry expressions on Chemistry Stack Exchange? That's my personal #1 annoyance, because it requires quite a bit of work to fix, unlike homework which you just vote to close and move on.

As for how to monitor whether it's working: I'm going to argue that there isn't a way to do that. But at least, you will have access to site analytics. In my opinion, the close rate is one of the more telling statistics.

  • $\begingroup$ Some pointer to formatting chemical formulas/equations would be great. The question editing window has a link to chemistry.stackexchange.com/help/formatting, but for our site, chemistry formatting (\ce{}) should be at the top of that help page. Do you know who can edit that page? $\endgroup$
    – Karsten Mod
    Aug 21, 2022 at 0:44
  • $\begingroup$ I added what physics has on the side bar for step 1 (i.e. every time anyone asks a question). They use the term homework-like instead of homework. $\endgroup$
    – Karsten Mod
    Aug 21, 2022 at 0:52
  • $\begingroup$ Off the top of my head, I'm not sure which of the help centre pages are editable. Some can be edited by you, some can't. It should be in the moderator help somewhere, or maybe on meta... $\endgroup$ Aug 21, 2022 at 11:08

First-time askers

To see what kind of impact better guide for first-time askers might have, I ran a SEDE query concerning the number of questions users ask on this site. Here is a summary:

  • 14,581 users asked a single question (so 14,581 questions)
  • 2419 users asked 2-10 questions (16,565 questions)
  • 431 users asked more than 10 questions (10,533 questions)

So as expected, most of our users ask no questions at all, and of those who ask questions, most ask a single one. More than a third of the questions on this site are from one-question-only askers, and more than 40% of questions on our site are the first question the OP ever asked. So seeing the "Asking good questions" pane is a common experience, and it is worth thinking about customizing it, in my opinion.

Here is the SEDE query (I entered a cut-off of top 3000 askers):

SELECT TOP ##numberOfRecords:int?100## OwnerUserId AS [User Link],
  COUNT(*) AS 'Questions',
  COUNT(*) - COUNT(AcceptedAnswerId) AS 'Without accepted answer',
  COUNT(*) - SUM(SIGN(AnswerCount)) AS 'Unanswered'
  FROM Posts
  WHERE PostTypeId = 1 -- Questions
    AND OwnerUserId IS NOT NULL
    AND OwnerUserId != -1 -- Community user
  GROUP BY OwnerUserId

And here is a histogram, excluding the zero and one question users and those asking more than 20 questions:

enter image description here

  • $\begingroup$ Vote up just for using SEDE. I've been meaning to use the tool more and it's nice to see you jump right in and do it. $\endgroup$
    – Buck Thorn Mod
    Aug 25, 2022 at 5:20
  • $\begingroup$ BTW this should strictly be an edit to your question, since it is not an answer to the original post :) $\endgroup$
    – Buck Thorn Mod
    Aug 25, 2022 at 5:25
  • $\begingroup$ There are a lot of additional questions which this begins to answer. For instance: what do we expect from first time users in terms of investment and quality of posts? Who are the people who ask second and more questions, what motivates them to continue using the site? Are their posts of better quality than posts of people who ask a single question? Does the site lose good questions because first time users shy away after the initial interaction? $\endgroup$
    – Buck Thorn Mod
    Aug 25, 2022 at 5:32

You must log in to answer this question.

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