Appendix A: Standardization, Reliability, and Validity Tables

Table A.59. Interaction Effects between Rater Race/Ethnicity and Ratee Race/Ethnicity: EQ 360 2.0 Normative Sample

The following table summarizes the results from the analysis of covariance (ANCOVA) and multivariate analyses of covariance (MANCOVAs) conducted on the composite scales and subscales to examine potential effects of ratee race, rater race, and the interaction between the two on EQ 360 2.0 scores in the normative sample. Wilks’ lambda ranges from 0.00 to 1.00 and depicts the amount of variance not explained by the demographic variable. F and p values convey the statistical significance of the demographic variables. Partial η2 values provide an effect size for describing the effects as small, medium, or large. The effect sizes found (all less than .06) suggest negligible differences among participants based on rater and ratee race.

Analysis	Effect	Wilks’ Lambda	F (df)	p	partial η2
Total EI	Rater Race (White vs. non-White)	n/a	0.10 (1, 2948)	.757	.00
	Ratee Race (Black vs. Hispanic/Latino vs. White)	n/a	1.95 (2, 2948)	.143	.00
	Rater Race x Ratee Race	n/a	2.41 (2, 2948)	.090	.00
Composite Scales	Rater Race (White vs. non-White)	.994	3.46 (5, 2944)	.004	.01
	Ratee Race (Black vs. Hispanic/Latino vs. White)	.995	1.43 (10, 5888)	.160	.00
	Rater Race x Ratee Race	.995	1.48 (10, 5888)	.141	.00
Subscales	Rater Race (White vs. non-White)	.980	3.72 (16, 2931)	< .001	.02
	Ratee Race (Black vs. Hispanic/Latino vs. White)	.980	1.89 (32, 5862)	.002	.01
	Rater Race x Ratee Race	.982	1.68 (32, 5862)	.010	.01

Note. N = 2,956. F-test values represent univariate effects for the Total EI score and multivariate effects for the composite scales and subscales. Guidelines for evaluating partial η2 are .01 = small; .06 = medium; .14 = large. n/a = not applicable.