How to determine confidence intervals around percentages

[Nila Saffari]

Hi all,
I would appreciate it if you could help with my question.

I used the following command to get percentages for my main independent variable (sex):
tabulate sex variable2, col chi2 --> got a 2x2 table (each variable has 2 levels)

Now, i need to get confidence intervals around the percentages in each cell (four cells). I used the following command but doesn't work:
ci proportions sex variable2

And this command (ci proportions sex) gives me confidence intervals for sex not for men and women separately
Thank you very much,
Nila

[Carole J. Wilson]

Please provide the exact command you gave in Stata and the exact output (use code delimiters--explained in FAQ #12).

When I run the command you give, I get the following error:

. ci proportions sex variable2
variable proportions not found

So this is obviously not what you typed.

The command:

proportion sex, over(variable2)

will give you confidence intervals for the same proportions of cases given in the command:

tab sex variable2, col

[Nila Saffari]

Thank you, thank you very much; I really appreciate it. I spent the whole day to figure it out but you saved me.
I used the command "proportion sex, over (variable2)" and as you mentioned, it gave me CIs for the proportions I needed.
Thank you very much, Nila

[Dirk Enzmann]

Note that there are different approaches with different results for calculating confidence intervals of proportions. You will find examples and links to the literature in http://www.statalist.org/forums/foru...eneral/1328961

[Nick Cox]

Note further that ci proportions is legal syntax as of Stata 14.1

http://www.stata.com/help.cgi?whatsnew

-------- update 29oct2015 (Stata version 14.1)

2. ci and cii have the following improvements and changes:

a. With new commands ci variances and cii variances, you can now compute confidence intervals for variances or standard deviations. Classical confidence intervals for normal data and Bonett confidence intervals for nonnormal data are provided.

b. ci means and cii means are now used to compute confidence intervals for means of normally and Poisson distributed variates. The previous commands were ci and cii. The old syntax is preserved under version control.

c. ci proportions and cii proportions are now used to compute confidence intervals for proportions. The previous commands were ci with option binomial and cii with option binomial. The old syntax is preserved under version control.

[Nila Saffari]

Thank you very much Dirk; I really appreciate the link to your detailed explanations.
Nila

[Nila Saffari]

Thank you very much Nick; I was wondering how come the code didn't work with my stata; thank you.
Nila