Testing the difference for a categorical variable

ihtishamsultan — Tue, 25 Aug 2020 15:29:13 GMT

Hello, I have a binary/categorical variable - Location. I would like to see if there is a statistical difference between the presence of complaints between the two locations.

For complaints I have the following values-

"." missing values denotes no complaint in the dataset.

1 denotes the presence of complaint

I was tilting towards using proc freq to do a chi-square test but wasn't sure. Can anyone please guide.

Here is the sample data.

data test;
infile datalines;
input customernum complaint location$;
datalines;

100 1 ND
200 . D
300 1 D
400 1 D
500 . ND
600 1 ND
700 1 D
800 1 D
900 . ND
1000 1 ND
1100 1 D
1200 1 D
1300 1 D
;
RUN;

Re: Testing the difference for a categorical variable

ballardw — Tue, 25 Aug 2020 15:30:45 GMT

Proc Freq and Chi-sq would be one of my first choices. With the data you show you would need to include the MISSING option on the tables statement so that missing is treated as a category for the Complaint variable.

topic Re: Testing the difference for a categorical variable in SAS Programming

Testing the difference for a categorical variable

Re: Testing the difference for a categorical variable