Summary of categorical variable

[Camilo Quntero]

Hi all,
I want know how calculate the average and standard deviation (sd) of a categorical variable. For example if I have a variable that tell me the education level in a any village in where 1=none and primary level, 2= secondary, 3= technical and university and 4= postgraduate, and I only want know the average and sd for the people in secondary level... How do I do in Stata?
I know that is a simple question but I don´t remember, please point me the way.

Thanks...

[Joseph Coveney]

summarize <your variable of interest here> if education_level == 2

[Camilo Quntero]

Joseph thanks for reply my post.

You know when I have a binary variable the command <sum> show me the average and sd the group with attribute. but with categorical variable is difficult. I run the command like you point me, but the result in average is the sum of the all one - one is attribute- and the value in sd is rule -"." in state-. I don´t if you understand me joseph??

thanks again

[Carlo Lazzaro]

Camilo:
the best way to get understood is to post what you typed and what Stata gave you back (as per FAQ).
As far as I can get your query, Stata reported a missing standard deviation for your variable. I'd bet that mean, miminum and maximum are the same, too.
This occurs whenever there's no variation in the values of a given variabl:

. set obs 1
number of observations (_N) was 0, now 1

. g var=1

. su

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
         var |          1           1           .          1          1

That said, it's meanigless to calculate descriptive statistics of a categorical variabe such as education level for two reasons, at least:
- each level that variable is composed of is created at research convenience (ie, does not mesure anything);
- as discussed above, you cannot have any variance for a constant value (such as education level 1, 2 or else).

Probably, your goal was different, that is calculating descriptive statistics for a given variable under the condition:

if education level==2

It's easy to do it with Stata:

. use "C:\Program Files (x86)\Stata14\ado\base\a\auto.dta", clear
(1978 Automobile Data)

. sum price if foreign==0

Variable | Obs Mean Std. Dev. Min Max
-------------+---------------------------------------------------------
price | 52 6072.423 3097.104 3291 15906

. bysort foreign: sum price

-----------------------------------------------------------------------------------------------------------------
-> foreign = Domestic

Variable | Obs Mean Std. Dev. Min Max
-------------+---------------------------------------------------------
price | 52 6072.423 3097.104 3291 15906

-----------------------------------------------------------------------------------------------------------------
-> foreign = Foreign

Variable | Obs Mean Std. Dev. Min Max
-------------+---------------------------------------------------------
price | 22 6384.682 2621.915 3748 12990