498 Chapter 11:Goodness of Fit Tests and Categorical Data Analysis
as Republicans, and 32 women who classified themselves as Independents; that is,N 11 =
68,N 12 =56, andN 13 =32. Similarly,N 21 =52,N 22 =72, andN 23 =20.
Use these data to test the hypothesis that a randomly chosen individual’s gender and
political affiliation are independent.
SOLUTION From the above data, we obtain that the six values ofnpˆiqˆj=NiMj/nare as
follows:
N 1 M 1
n
=
156 × 120
300
=62.40
N 1 M 2
n
=
156 × 128
300
=66.56
N 1 M 3
n
=
156 × 52
300
=27.04
N 2 M 1
n
=
144 × 120
300
=57.60
N 2 M 2
n
=
144 × 128
300
=61.44
N 2 M 3
n
=
144 × 52
300
=24.96
The value of the test statistic is thus
TS=
(68−62.40)^2
62.40
+
(56−66.56)^2
66.56
+
(32−27.04)^2
27.04
+
(52−57.60)^2
57.60
+
(72−61.44)^2
61.44
+
(20−24.96)^2
24.96
=6.433
Since (r−1)(s−1)=2, we must compare the value ofTSwith the critical valueχ.05,2^2.
From Table A2
χ.05,2^2 =5.991
SinceTS≥5. 991, the null hypothesis is rejected at the 5 percent level of significance.
That is, the hypothesis that gender and political affiliation of members of the population
are independent is rejected at the 5 percent level of significance. ■
The results of the test of independence of the characteristics of a randomly chosen
member of the population can also be obtained by computing the resultingp-value. If
the observed value of the test statistic isT=t, then the significance levelαtest would