Table 4.

Effect sizes for the planned t-test comparisons between the different conditions (the post-testing effect, the effect of quiz position, and the pre-testing effect) to the trained and untrained questions (new, generalisation, and total untrained) and across questions

	Planned t-test comparisons
	Reading-quiz vs.	Reading-quiz vs.	Quiz-reading vs.
	Reading-reading	Quiz-reading	Reading-reading
Trained questions	d = 0.74 [0.44, 1.04]	d = 0.34 [0.05, 0.63]	d = 0.35 [0.07, 0.64]
	p < 0.0001	p = 0.019	p = 0.013
New questions	d = 0.27 [−0.02, 0.56]	d = 0.35 [0.06, 0.64]	d = −0.11 [−0.40, 0.17]
	p = 0.092	p = 0.017	p = 0.44
Generalisation questions	d = 0.34 [0.05, 0.63]	d = 0.22 [−0.07, 0.50]	d = 0.09 [−0.20, 0.37]
	p = 0.032	p = 0.14	p = 0.52
Untrained questions (total)	d = 0.36 [0.07, 0.66]	d = 0.32 [0.03, 0.61]	d = −0.01 [−0.29, 0.27]
	p = 0.028	p = 0.027	p = 0.95
All questions	d = 0.62 [0.32, 0.92]	d = 0.36 [0.07, 0.65]	d = 0.22 [−0.06, 0.51]
	p < 0.0001	p = 0.013	p = 0.12

Effect sizes (Cohen’s d) are given with 95% confidence intervals