Table of Contents Table of Contents
Previous Page  262 / 762 Next Page
Basic version Information
Show Menu
Previous Page 262 / 762 Next Page
Page Background

262

Chapter 6: Categorical Data

The next table,

Table 6-15

, shows the conditional distributions for the

majors

given the

gender

of the applicant.

Table 6-15:

Conditional Distributions for Majors

given Gender

From these conditional distributions for

major

given the

gender

of the

applicants, one can observe that for the males, the percentage decreases from

major to major whereas it stayed relatively the same for the female applicants

except for major 2. For the male applicants, more are accepted in major 1

and major 2, while more females than males are accepted in the other majors.

Again, one can make counterarguments for bias. Here again, the variable of

major

was

confounding

the

gender

variable in the computation of the 70%

of males being accepted compared to the 30% of females.

Figure 6-12

shows the bar graph displaying these conditional distributions for

Table 6-15

. From the graph of the conditional distributions for major given

the gender, one can observe again that for the different majors, the

percentages for the males and females are generally going in opposite

directions.

Click here for the Contingency Table Workbook