Mendenhall and Sincich (1993) present data that can be used to investigate allegations of gender discrimination in the hiring practices of a particular firm. These data are as follows:

In this table, y is a dummy variable that equals 1 if a potential employee was hired and 0 otherwise; x1 is the number of years of education of the potential employee; x2 is the number of years of experience of the potential employee; and x3 is a dummy variable that equals I if the potential employee was a male and 0 if the potential employee was a female. If we use the logistic regression model

to analyze these data, we find that the point estimates of the model parameters and their associated p-values (given in parentheses) are b0 = - 14.2483(.0191), b1 = 1.1549(.0552), b2 = .9098 (.0341). and b3 = 5.6037(.0313).
a. Consider a potential employee having 4 years of education and 5 years of experience. Find a point estimate of the probability that the potential employee will be hired if the potential employee is a male, and find a point estimate of the probability that the potential employee will be hired if the potential employee is a female.
b. Using b3 = 5.6037, find a point estimate of the odds ratio for x3. Interpret this odds ratio. Using the p-value describing the importance of x3, can we conclude that there is strong evidence that gender is related to the probability that a potential employee will be hired?

Transcribed Image Text:

1 0 1 1 0 1 0 1 1 0 0 1 1 0 rs ie e | 5 4 0 1 7 150591102 p궈 rs ca ye 46864446848646 y-10010000100011 n Xy 0 1 1 1 0 0 0 0 0 1 0 0 0 rs ie-2 0 6 3 1 3 2 4 102527 rs ye-64664844684846 Sy-00110100010000 p(xi, X2, X3) =