Is there a way to run multiple Chi-Squared Test in SAS without separating files?

[sasworker16]

Hello,

This may sound like a very basic question... but,

I am working with a set of chemotherapy data that contains 3 types of chemotherapy and it would be really helpful if there is a way to run Chi-Squared Test in SAS without seperating the files.

My data and variables looks like the following:

Patient_ID=ID assigned to Patient

Chemo_Type=Type of Chemotherapy (1=type A, 2=type B, 3=type C)

Readmiited_YN=Readmission to Hospital after Chemotherapy (1=readmission, 0=no readmission),

Patient_ID    Chemo_Type   Readmitted_YN

10012                        1                       1

10013                        2                       0

10014                        3                       0

10015                        2                       1

10016                        2                       0

10017                        3                       1

10018                        1                       1

10019                        2                       1

.

.

.

(cont')

with this data set I have to do a run a chi-squared test for Chemo_Type=1 vs. Chemo_Type=2 & Chemo_Type=1 vs. Chemo_Type=3 (*I do NOT need to run a chi-squared test for Chemo_type=2 vs. Chemo_type=3)

Is there a way to run the chi-squared test for Chemo_Type=1 vs. Chemo_Type=2 & Chemo_Type=1 vs. Chemo_Type=3 with the current data set I have?

I am currently doing the following :

-------------------------------------------------------------------------------------------

data w.cancer_1vs2; set w.cancer; if Chemo_type=3 then delete;

proc freq data=w.cancer_1vs2;

tables Chemo_Type*Readmitted_YN/chisq;

run;

data w.cancer_1vs3; set w.cancer; if Chemo_type=2 then delete;

proc freq data=w.cancer_1vs3;

tables Chemo_Type*Readmitted_YN/chisq;

run;

---------------------------------------------------------------------------------------------

Is there a way to run this in a single file (w.cancer) without having to delete the undesired Chemo_Type for running Chi-Squared Test?

Thank you.

[Reeza · Accepted solution]

Yes. Most Procs support the WHERE statement. 

Proc freq data = have;
Where chemo_type ne 3;
Table .../ chisq;
Run;

Proc freq data = have;
Where chemo_type ne 2;
Table ... / chisq;
Run;

[sasworker16]

Thank you so much for your answer.

Your answer was really helpful.