Applying generalized theory to optimize the quality of high-stakes objective structured clinical examinations for undergraduate medical students: experience from the French medical school

Table 3 Variance component estimates and generalizability coefficients for the 2-facet model (persons, stations) using a minimum norm quadratic unbiased estimation (MINQUE) approach

Variance component estimates *
Source of variation	Clinical skills score^**	% variance	Generic skills score^**	% variance	Global performance	Percentage of variance
Person (n = 73)	32.228	8.15	46.183	14.33	0.121	13.75
Station (n = 3)	95.582	24.17	11.505	3.57	0.125	14.20
Person x Station	108.898	27.54	137.725	42.71	0.312	35.45
Residual error	158.743	40.14	127.014	39.39	0.322	36.59
G-coefficient^**	0.503		0.923		0.742

^*Variance component estimates for persons (P = student or object of measurement); stations (S), examiners (E), interactions between facets and residual error (e) reflecting unidentified sources of error
^**expressed as a percentage
^***G-coefficients of students’ scores were calculated via the following equation:
Person/[Person + (Station/3) + (Person × Station/219)]

ISSN: 1472-6920