4.4. Relationships Between Sets of Tables
Suppose you transposed the rows and columns of Table 4.7. You would obtain the following:
Diagnosis | Adverse Effects | Placebo | Dosage1 | Dosage2 | Dosage3 | Dosage4 |
---|---|---|---|---|---|---|
I | No | 26 | 26 | 23 | 18 | 9 |
I | Yes | 6 | 7 | 9 | 14 | 23 |
II | No | 26 | 12 | 13 | 1 | 1 |
II | Yes | 6 | 20 | 20 | 31 | 31 |
Furthermore, suppose you analyzed these tables as two 2 × r tables, making the response variable the row variable and the grouping variable the column variable.
proc freq order=data; weight count; tables diagnosis*response*treatment / cmh; tables diagnosis*treatment*response / cmh; run;
Look at the resulting table of Mantel-Haenszel statistics for DIAGNOSIS by RESPONSE by TREATMENT and compare it to the reprinted table of DIAGNOSIS by TREATMENT by RESPONSE.
Get Categorical Data Analysis Using The SAS® System, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.