Calculate a percentage for each group of respondent to questionnaire

[Fimi Karimi]

Hi everybody,

I have a questionnaire which is answered by respondent in different country. I need to calculate the percentage ( the number of people who have given specific answer to each question to total number of respondents to that question) and I need to exclude missing values in my calculation.

. dataex S002 S003 S024 A001 A035 A075

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. For more info, type help dataex
clear
input byte S002 int(S003 S024) byte(A001 A035 A075)
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 1
7 20 207 2 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 1
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 0 1
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 2 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 1
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 2 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 1 1
7 20 207 1 1 1
7 20 207 1 1 0
7 20 207 2 1 1
7 20 207 1 1 0
7 20 207 1 1 1
7 20 207 2 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 0 0
7 20 207 1 0 0
7 20 207 1 1 1
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 3 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
7 20 207 1 1 0
7 20 207 2 0 0
7 20 207 1 1 0
7 20 207 2 1 0
7 20 207 1 1 0
7 20 207 2 0 0
7 20 207 1 1 0
7 20 207 1 1 0
7 20 207 1 0 0
end
label values S002 S002
label def S002 7 "2017-2022", modify
label values S003 S003
label def S003 20 "Andorra", modify
label values S024 S024
label def S024 207 "Andorra (7)", modify
label values A001 A001
label def A001 1 "Very important", modify
label def A001 2 "Rather important", modify
label def A001 3 "Not very important", modify
label values A035 A035
label def A035 0 "Not mentioned", modify
label def A035 1 "Important", modify
label values A075 A075
label def A075 0 "Not mentioned", modify
label def A075 1 "Belong", modify

------------------ copy up to and including the previous line ------------------

Listed 100 out of 439531 observations
Use the count() option to list more

1- Questions are A001 A035 A075

2- Answers:
For A001 : "Very important", and "Rather important"
For A035: "Not mentioned"
For A075: "Belong"

3-I want to calculate this ratio for each question, in terms of each country ( S003) and the time range (S002)

For example:

Country Time range Question1 Question 2 Question3
Argentine 2017-2022 0.28 0.56 0.96
Argentine 2010-2014 0.35 0.78 0.94
Argentine 2005-2009 0.64 0.77 0.86
Argentine 1999-2004 0.82 0.96 0.96
Brazil 2017-2022 0.98 0.56 0.96
Brazil 2010-2014 0.65 0.34 0.28
Brazil 2005-2009 0.34 0.71 0.94
Brazil 1999-2004 0.72 0.96 0.37


Could I ask you help me with the command?
Thanks.

[Carlo Lazzaro]

Fimi:
an idea may be:

. egen wanted_1=pc( A001)
. tab wanted_1

   wanted_1 |      Freq.     Percent        Cum.
------------+-----------------------------------
   .9090909 |         91       91.00       91.00
   1.818182 |          8        8.00       99.00
   2.727273 |          1        1.00      100.00
------------+-----------------------------------
      Total |        100      100.00

.