Table 4.

Regression analyses for diagnostic accuracy as outcome

Predictor	b	ß	ß 95% CI	p	Model test and fit
Model 3					F(3, 102) = 10.00, p < 0.001
Intercept	0.00			.981	R2 = 0.23
Hypothesis generation	0.48**	0.40	[0.22, 0.57]	< .001	Adj. R2 = 0.20
Evidence generation	0.40*	0.22	[0.05, 0.40]	0.012
Evidence evaluation	0.17	0.09	[− 0.09, 0.26]	0.331
Model 4a					F(2, 103) = 3.42, p = 0.037
Intercept	0.19*			0.040	R2 = 0.06
Conceptual knowledge	0.27	0.16	[− 0.06, 0.39]	0.157	Adj. R2 = .04
Strategic knowledge	0.19	0.12	[− 0.11, 0.35]	0.302
Model 4b					F(5, 100) = 6.99, p < 0.001
Intercept	− 0.12			0.335	R2 = 0.26
Conceptual knowledge	0.24	0.14	[− .006, 0.35]	0.172	Adj. R2 = 0.22
Strategic knowledge	0.12	0.07	[− 0.15, 0.30]	0.516
Hypothesis generation	0.49	0.40	[0.23, 0.58]	< 0.001
Evidence generation	0.26	0.15	[− 0.05, 0.34]	0.142
Evidence evaluation	0.16	0.08	[− 0.10, 0.26]	0.368

Model 3 is a multiple regression containing diagnostic activities variables. Model 4 is a hierarchical regression, consisting of knowledge variables in Model 4a and knowledge and diagnostic activities in Model 4b. b represents unstandardized regression weights. ß represents standardized regression weights. CI = confidence interval. *p < 0.05, **p < 0.01, ***p < 0.001