Table 3.

Generalizability study for p × i × r and nested r:(p × i) designs using peers as raters among 95 preclinical medical students, ten items, and three raters

Source of variation p × i × r design	% of the total variance	Source of variation r:(p × i) design	% of the total variance
Student (P)	20.10	Student (P)	24.41
Item (I)	1.26	Item (I)	1.20
Rater (R)	0.03
PI	11.24	PI	7.03
PR	12.81
IR	0.00	Rater (R):PI	67.35
Residual (PIR, e)	54.57
Total	100.00		100.00