
The procedure is also used for performing Chi-Square test, which helps establishing the independence of two or more samples on the basis of frequency.
Let's understand it by an examples.
Download the data using following link, we will use the same for almost all tests:
Data for tests
Download the data, keep it at a location and assign the location to library "a" in SAS.
Let's try to answer following questions :
1. How Many Males and Female are there in Data ?

Proc freq data = a.sample_1;
tables female /
out = gender_freq ;
run;
2. How Many Males and Female are there in each program type ?

Proc freq data = a.sample_1;
tables female * prog /list
out = prog_gender_freq ;
run;

Proc freq data = a.sample_1;
tables female * prog /
out = prog_gender_freq ;
run;
tables female * prog /
out = prog_gender_freq ;
run;
The matrix format shows the important stats : Row Pct and Col Pct which are nothing but row percentage and column percentage.
Out of total population under Pct of prog = 1 when female = 0 category we have 21 people.
So Row Pct of prog = 1 when female = 0 i.e. for male population is 23.08 has been calculated as 21/( 21+47+23). It means that In male population , out of all programs, frequency of Prog 1 is 23.08%.
If we don’t put an asterisk (*) in between the variables, it will consider them individually and in the output data, frequency as pr last variable would come.
Try :
Proc freq data = a.sample_1;
tables female prog /list
out = prog_gender_freq ;
run;
tables female prog /list
out = prog_gender_freq ;
run;
Another usage of Proc Freq is in performing a Chi Square test :
No comments:
Post a Comment
Do provide us your feedback, it would help us serve your better.