Tabulation/Descriptive Statistics

[Sattar Khan]

Greetings Everyone!

Hopefully, you will be doing well. I have variable gender diversity in number. I have sub-divided this variable in the following way,

gen DV1 = diversityinnumber == 1 if !mi(diversityinnumber)
gen DV2 = diversityinnumber == 2 if !mi(diversityinnumber)
gen DV3 = diversityinnumber == 3 if !mi(diversityinnumber)
gen DV4 = diversityinnumber > 3 if !mi(diversityinnumber)

After that now i wanted to do tabulation or summary statistics that howmany companies have 1 female in the board or two females in the board or three female in the board etc. Kindly help me in this regard. Warms Regards Sattar Khan

[Ken Chui]

Hello!

I'd suggest making all the four DV# as just one variable that takes values of 1, 2, 3, and 4 based on the coding scheme. And then using -bysort- or -statsby- to make multiple summaries. See below for what I mean. (Also, is there no company with 0 female in the board?)

clear

input diversityinnumber
1
2
3
4
5
6
.
end

recode diversityinnumber (1 = 1 "1 female") ///
                         (2 = 2 "2 females") ///
                         (3 = 3 "3 females") ///
                         (nonmissing = 4 "4 or more females") ///
                         (missing = .), into(dv)

* Subgroup summary
gen somevariable = rnormal()
bysort dv: sum somevariable

[Sattar Khan]

Hello!

I'd suggest making all the four DV# as just one variable that takes values of 1, 2, 3, and 4 based on the coding scheme. And then using -bysort- or -statsby- to make multiple summaries. See below for what I mean. (Also, is there no company with 0 female in the board?)

clear

input diversityinnumber
1
2
3
4
5
6
.
end

recode diversityinnumber (1 = 1 "1 female") ///
(2 = 2 "2 females") ///
(3 = 3 "3 females") ///
(nonmissing = 4 "4 or more females") ///
(missing = .), into(dv)

* Subgroup summary
gen somevariable = rnormal()
bysort dv: sum somevariable

recode diversityinnumber (1 = 1 "1 female") ///
Rules defining value labels not allowed when overwriting a variable
r(198);

Dear Chui
following error appear when applying the code

[Ken Chui]

It's likely caused by the generated variable already created. If you use a -drop xxxxx- command to drop that variable first, then the recode should work.