Table 4.7: One-way Analysis of Variance for Mean MHCS Sub-scale Scores by Bias
Crime Indicator
Source df F rlp2 p
Between subjects
Indicator 2 68.65 .20 .001*
S within-group 547 (.90)
error
Note: 1) Values enclosed in parentheses represent mean square errors 2) rp2 reported is a partial eta
squared. An asterisk indicates significant results p <.01
A one way ANOVA was conducted to determine whether mean scores varied
across type of crime indicator and the results were significant F(2,547) = 68.65, p >.001,
rp2 = .20 (See Table 4.7). The Levene statistic for the model was significant indicating
the assumption of equal variances was violated. Post-hoc comparisons were conducted
with the Tamhane test to control for the violation of equal variance assumption. The
mean score for language (M= 6.19, SD = .76) was significantly different < .001 from
both bias symbols (M= 5.14, SD = .98) and the hate group membership of the perpetrator
(M= 5.24, SD = 1.08). The difference between the bias symbols and group membership
items was not significant (See Table 4.7).