[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Statalist advice: a summary
Nick Cox <firstname.lastname@example.org>
st: Statalist advice: a summary
Fri, 21 Jun 2013 15:58:14 +0100
Old hands will know that I post something like this every few months as
a reminder of some basics. It is not to be thought of as the rules.
There is only one rule on Statalist, and it comes from the top, Marcello
Pagano. You are asked to use your full real name in posting. We are
deliberately and deliciously old-fashioned in that sense.
Everything else is _advice_ intended in everyone's best interests.
In a sentence: Help us to help you, and here's how.
Reminder: When you joined the list you were asked to read the FAQ
before posting, and we really did mean that. Please read the FAQ before
posting! Much of it can be skimmed or skipped on a first reading.
For another source of advice on Statalist, see
It is important to remember that Statalist is a discussion list, not a
help line. The distinction might seem a little obscure or subtle, so
let's spell it out:
* On a help line, someone is obliged to reply, even if the answer is
perfunctory. On a discussion list, people will happily delete your
posting if they don't understand it or it's too much like hard work even
to try to find out what you are seeking. There is no mechanism on
Statalist for ensuring that anybody answers, so a question that looks
too difficult for everyone will just lie there.
* A help line is aimed at helping individuals, and giving the client a
good answer is the key goal. You as a client will not know, and need not
care, about other clients. On a discussion list, it is fine if people
get individual help in public, but it is also important that such help
contributes to an archive of solved problems that people can search. A
lousy question that is too difficult to answer wastes the questioner's
time and also clogs the list and the archives with unhelpful posts.
Following all the advice below won't guarantee an answer that satisfies
you, but ignoring most or all of it will make such an answer much less
0. Use your full real name in postings. Incomplete or cute or cryptic
names may be standard elsewhere, but this is firm Statalist policy. If
you don't agree, please don't post. Other forums are likely to be more
to your taste.
1. Help yourself first. Use the Stata help, the Stata manuals, -search-,
the Stata FAQs, and the Statalist archives, in that order. (The idea
that you should search the web before you read the manual is a strange
belief shared by many new users.)
2. Explain your data structure clearly and with examples (variable types
etc.) We can only understand your dataset to the extent that you explain
3. Indeed, it is usually much clearer if you give an example which can
easily be replicated using data supplied with Stata or in your post.
4. Show the exact Stata syntax you used and show the exact Stata output
you got. (Never say just that something "doesn't work" or "didn't work",
but explain precisely in what sense you didn't get what you wanted.)
5. Specify the Stata version you used and the operating system you used
if it could possibly be relevant. Don't assume that the whole Stata
world uses MS Windows, let alone MS Excel.
6. Explain where user-written commands you refer to come from, for
example the Stata Journal, SSC, or someone's website. This makes clearer
what you are talking about, to everyone's benefit.
7. Give full literature references, not references of the form "Greene"
or the form "Sue, Grabbit, and Runne (2002)". That means references of a
standard that you would expect to find in a professional publication.
8. Ask a precise question that is easy to answer. Is this correct? or
what should I do with my data? usually don't qualify.
9. The best strategy is to ask a question that someone else will want to
answer, not to act clueless or desperate.
10. Do send plain text only. Don't send attachments or use formatting
such as HTML.
11. Post once and wait patiently for a reply. (To see if something "got
through", check the archives.)
12. Try to use Stata terminology wherever possible, not terminology that
may be more familiar to you because you are more used to other software.
Stata's the language we all share. For example, it is not a good idea to
assume that people likely to answer questions think or work using
13. The correct spelling is "Stata", not "STATA". Several of the most
active experts on the list can get irritated if you get that wrong, and
you are free to regard them as pedantic. More importantly, if you write
"STATA" you are making it all too obvious that you haven't studied the
14. Close threads with concise summaries sent to the list of what
worked. That is the best way to show appreciation and to contribute
further to the list.
* For searches and help try: