Table 2.

Parameter estimates for the models estimating the effect of testing with short-answer questions and multiple-choice questions, time of test, and retrievability on short-answer questions and multiple-choice questions on learning performance in the criterial tests.

	Short-answer questions				Multiple-choice questions
Parameter	β	SE	z	p	β	SE	z	p
Intercept	-0.34	0.25	−1.36	0.173	0.07	0.29	0.25	0.803
Testing	0.44	0.24	1.84	0.033a	−0.42	0.24	−1.76	0.078
Criterial test 2	0.80	0.27	2.96	0.003	0.74	0.29	2.55	0.011
Criterial test 3	0.11	0.38	0.30	0.768	0.38	0.42	0.91	0.361
Testing × Criterial test 2	−0.14	0.22	−0.63	0.531	0.12	0.22	0.55	0.583
Testing × Criterial test 3	−0.23	0.32	−0.73	0.468	−0.11	0.34	−0.33	0.739
Low retrievability	0.03	0.25	0.10	0.917	−0.31	0.25	−1.23	0.219
Medium retrievability	0.09	0.23	0.40	0.692	−0.35	0.27	−1.33	0.184
Testing × Low retrievability	−0.60	0.28	−2.14	0.016a	0.17	0.27	0.62	0.534
Testing × Medium retrievability	−0.66	0.35	−1.88	0.030a	0.060	0.37	0.16	0.872
NParticipants	92				91
NItems	77				77

Testing (dummy-coded: testing = 1, restudy = 0). Criterial test 2 (dummy-coded: Criterial test 2 = 1, Criterial test 1 = 0). Criterial test 3 (dummy-coded: Criterial test 3 = 1, Criterial test 1 = 0). Low retrievability (dummy-coded: low retrievability = 1, high retrievability = 0). Medium retrievability (dummy-coded: medium retrievability = 1, high retrievability = 0). ^ap-values refer to one-tailed tests for β > 0. Other p-values refer to two-tailed tests.