Statistical Methods for Psychology

(Michael S) #1
The calculation of chi-square for the data on sexual abuse follows.

The contingency table was a 4 3 2 table, so it has (4–1) 3 (2–1) 53 df. The critical
value for on 3 dfis 7.82, so we can reject the null hypothesis and conclude that the level
of adult sexual abuse is related to childhood sexual abuse. In fact adult abuse increases con-
sistently as the severity of childhood abuse increases. We will some back to this idea shortly.^4

Computer Analyses


We will use Unah and Boger’s data on criminal sentencing for this example because it
illustrates Fisher’s Exact Test as well as other tests. The first column of data (labeled Race)

x^2

=29.63


=


(512 2 494.19)^2


494.19


1


(54 2 71.51)^2


71.51


Á(18^2 26.21)


2
26.21

1


(12 2 3.79)^2


3.79


x^2 = a

(O 2 E)^2


E


Section 6.4 An Additional Example—A 4 3 2 Design 149

Table 6.5 Adult sexual abuse related to prior childhood sexual abuse
Abused as Adult
Number of Child
Abuse Categories No Yes Total
0 512 (494.49) 54 (71.51) 566
1 227 (230.65) 37 (33.35) 264
2 59 (64.65) 15 (9.35) 74
3–4 18 (26.21) 12 (3.79) 30
Total 816 118 934

(^4) The most disturbing thing about these data is that nearly 40% of the women reported some level of abuse.
Exhibit 6.1a SPSS data file and dialogue box
SOURCE: Courtesy of SSPS Inc.

Free download pdf