Logistic Regression Using SAS®: Theory and Application

4.2. A Logit Model for a 2 × 2 Table

Let’s begin with the simplest case—a dichotomous dependent variable and a single dichotomous independent variable. That leads to a 2 × 2 table like the one in Table 2.2, reproduced here as Table 4.1.

Table 4.1. Death Sentence by Race of Defendant

	Blacks	Nonblacks	Total
Death	28	22	50
Life	45	52	97
Total	73	74	147

If we have access to the individual-level data, we can simply estimate a logit model directly, as with the following GENMOD program:

PROC GENMOD DATA=my.penalty;
  MODEL death = blackd / D=B;
RUN;

Results are shown in Output 4.1. Exponentiating the BLACKD coefficient yields 1.47, which is the odds ratio we calculated earlier, directly from the table. It is not statistically significant.

Output 4.1. GENMOD Output ...