Displaying Counts of Multiple Variables by Gender

[Demetriusz Rominski]

Hi,
I’m having trouble getting the counts of variables in rows by gender.
I want the rows to be: N (total number of women/men), then var1, var2 — listed one after another.
In the columns, at the top, there should be gender (grouping variable), with categories 0 and 1 displayed below the gender label. After that, some statistics.
My biggest problem is showing the counts of each variable in the rows broken down by gender.
Is it possible to do this using table or collect?
I would like to get table like this:

	sex
	male	female
	frequency	frequency
N	50	60
var1	20	10
var2	30	50

clear
input byte(Sex var1 var2)
1 1 1 
2 1 0 
1 0 1
2 0 0 
2 0 1 
2 . 0 
1 . 1
1 . 1
1 1 1
2 0 0
2 1 0
2 . 1 
1 1 0 
1 . 1
2 . 1
2 1 1
end

tabulate Sex 
tabulate var1 
tabulate var2

table (Sex), statistic(frequency)

collect layout (var1 var2) (result)

[Rich Goldstein]

not sure I completely understand, but I think what you want is the -dtable- command; see

h dtable

[Demetriusz Rominski]

My variables var1 and var2 are binary variables (0,1) — representing whether a person answered the question or not.
In the table, I would like to get the number of people who answered the question, meaning I want to take from this variable, for example var1, only the option ‘1’.

[Hemanshu Kumar]

If you have Stata 18 or higher, you can simply do

dtable, by(Sex, nototal) cont(var1 var2, stat(total)) nformat(%2.0f total) sample(, stat(freq))

which produces:

--------
     Sex
     1 2
--------
N    7 9
var1 3 3
var2 6 4
--------

Note: I am assuming you are looking not for the counts of the variables (as your thread title says), but the sum of the binary variables (as #3 seems to suggest).

[Demetriusz Rominski]

Thank you. I didn’t realize that summing 0 and 1 is basically just counting the 1s.

Is there any way to combine dtable outputs, for example merging separate categories like gender and education for the same variables (var1, var2)?
I don’t want to use the by option, because I’d like to add more variables like gender or education later.

/*    edu_lvl
1.No education
2.Primary education
3.Secondary education
4.Higher education */

clear
input byte(Sex edu_lvl var1 var2)
1 2 1 1 
2 1 1 0 
1 3 0 1
2 4 0 0 
2 1 0 1 
2 2 . 0 
1 4 . 1
1 3 . 1
1 2 1 1
2 2 0 0
2 1 1 0
2 3 . 1 
1 4 1 0 
1 2 . 1
2 1 . 1
2 4 1 1
end

[Andrew Musau]

See https://www.stata.com/support/faqs/r...ltiple-tables/.

[Demetriusz Rominski]

Is there any function in Stata, in dtable, that allows calculating row percentages of weighted N?
For example: var1: female 65.1%, male 34.9%, total 100%.

input byte(Sex edu_lvl var1 var2)
1 2 1 1 
2 1 1 0 
1 3 0 1
2 4 0 0 
2 1 0 1 
2 2 . 0 
1 4 . 1
1 3 . 1
1 2 1 1
2 2 0 0
2 1 1 0
2 3 . 1 
1 4 1 0 
1 2 . 1
2 1 . 1
2 4 1 1
end


dtable, by(Sex) ///
    cont(var1 var2, stat(total mean )) ///
    nosample ///
    nformat(%2.0f total) ///
    name(a1)

collect composite define column1 = total
collect composite define column2 = mean
collect label levels result column1  "N" column2 "Row % of weighted N", modify

collect layout (var) (Sex#result)