Exploring sociodemographic subgroup differences in multiple mini-interview (MMI) performance based on MMI station type and the implications for the predictive fairness of the Hamburg MMI

Table 3 Hierarchical linear model predicting MMI station performance (N = 11,658 ratings within 1438 candidates over 8 years)

Fixed effects	b	95%-CI	p	Effect size
First language German	0.16	0.08; 0.24	< 0.001	− 0.000
Interaction terms (male vs female):
gender x station type			0.003	0.001
gender x age			0.003	− 0.000
age x station type			< 0.001	0.002
Younger than 21
Simulation	− 0.27	− 0.34; − 0.20	< 0.001
Interview	− 0.22	− 0.29; − 0.15	< 0.001
Group	− 0.01	− 0.16; 0.14	0.883
21 and older
Simulation	− 0.11	− 0.20; − 0.02	0.020
Interview	− 0.06	− 0.15; 0.03	0.174
Group	0.15	− 0.01; 0.31	0.072
Interaction medical family background x station type (at least one vs. no parent is physician)			0.004	0.001
Simulation	− 0.06	− 0.12; 0.01	0.087
Interview	0.07	− 0.00; 0.13	0.053
Group	− 0.05	− 0.21; 0.11	0.572
R²	.006
Between year variance	0.02	2.0% of total variance
Between candidate variance (within year)	0.12	15.4% of total variance
Within rating / unexplained variance	0.77	82.6% of total variance

ISSN: 1472-6920