7.1 Estimation of domain proportions

If we are interested in understanding how people comment on the sentence, “People would be afraid to walk alone after dark in this area” (\(\texttt{ENV_AFRDWLK_MCQ}\)) among the groups of people with or without a university degree, we may use the following codes.

R

svymean(~ENV_AFRDWLK_MCQ, 
         design = subset(CLSA.design, ED_HIGH_TRM ==                                                         "Non_university"))
svymean(~ENV_AFRDWLK_MCQ, 
         design = subset(CLSA.design, ED_HIGH_TRM == 
                                           "University"))

SAS

PROC SURVEYMEANS data = CLSAData mean;
VAR    ENV_AFRDWLK_MCQ;
DOMAIN ED_HIGH_TRM; 
STRATA GEOSTRAT_TRM;   
WEIGHT WGHTS_INFLATION_TRM; 
RUN;    

SPSS

Analyze \(\rightarrow\) Complex Samples \(\rightarrow\) Frequencies… \(\rightarrow\) Select the file “\(\texttt{CLSADesign.csaplan}\)” in the Plan panel \(\rightarrow\) select \(\texttt{ENV_AFRDWLK_MCQ}\) to the Frequency Table panel \(\rightarrow\) select \(\texttt{ED_HIGH_TRM}\) to the “Subpopulations:” panel \(\rightarrow\) click “Statistics…” \(\rightarrow\) select “Table percent” and “Standard error” \(\rightarrow\) click “Continue”\(\rightarrow\) click “OK”.

Stata

*We first change the survey design to that with inflation survey weight
svyset entity_id, strata(StraVar) weight(WGHTS_INFLATION_TRM) 
      vce(linearized) singleunit(certainty)
* For people who never attained any university 
svy linearized, subpop(if ED_HIGH_TRM == "Non_university") : tabulate
   ENV_AFRDWLK_MCQ ,  cell se ci stubwidth(20) format(%10.0g)
* For people who have attained an university 
svy linearized, subpop(if ED_HIGH_TRM == "University") : tabulate 
   ENV_AFRDWLK_MCQ , cell se ci stubwidth(20) format (%10.0g)

Result comparison

	R		SAS		SPSS		Stata
Population Estimates	Mean	SE	Mean	SE	Mean	SE	Mean	SE
Non university
Strongly Agree	0.0857	0.0141	0.0857	0.0141	0.0857	0.0141	0.0857	0.0141
Agree	0.1080	0.0177	0.1080	0.0177	0.1080	0.0177	0.1080	0.0177
Disagree	0.2548	0.0312	0.2548	0.0312	0.2548	0.0312	0.2548	0.0312
Strongly Disagree	0.5515	0.0357	0.5515	0.0357	0.5515	0.0357	0.5515	0.0357
University
Strongly Agree	0.1122	0.0275	0.1122	0.0275	0.1122	0.0275	0.1122	0.0275
Agree	0.1388	0.0287	0.1388	0.0287	0.1388	0.0287	0.1388	0.0287
Disagree	0.3006	0.0392	0.3006	0.0392	0.3006	0.0392	0.3006	0.0392
Strongly Disagree	0.4484	0.0448	0.4484	0.0448	0.4484	0.0448	0.4484	0.0448