Table 5.

Tests of differences between intervention components

Test	Coefficient difference (1)	F statistic (2)	df (3)	p value (4)
Coaching of personnel = CAI	0.263	2.75	27.86	0.109
Coaching of personnel = incentives	0.136	1.92	15.94	0.185
Coaching of personnel = peer‐assisted	0.093	0.36	17.23	0.554
Coaching of personnel = progress monitoring	−0.090	0.61	17.42	0.445
Coaching of personnel = small group	−0.121	1.26	19.10	0.275
Coaching of personnel = other domain	0.088	0.36	26.49	0.553
CAI = incentives	−0.127	0.60	15.49	0.449
CAI = peer‐assisted	−0.170	2.10	16.79	0.165
CAI = progress monitoring	−0.353	4.84	9.56	0.054
CAI = small group	−0.384	7.71	20.80	0.011
CAI = other domain	−0.175	2.55	9.59	0.142
Incentives = peer‐assisted	−0.043	0.08	21.79	0.784
Incentives = progress monitoring	−0.226	2.26	17.11	0.151
Incentives = small group	−0.257	4.99	15.99	0.040
Incentives = other domain	−0.048	0.09	22.81	0.771
Peer‐assisted = progress monitoring	−0.183	2.04	16.92	0.172
Peer‐assisted = small group	−0.214	4.13	21.72	0.055
Peer‐assisted = other domain	−0.005	0.00	19.16	0.959
Progress monitoring = small group	−0.031	0.06	15.41	0.803
Progress monitoring = other domain	0.178	2.77	21.01	0.111
Small group = other domain	0.209	3.53	22.66	0.073

Note: Column 1 reports the difference between the coefficients of the two components mentioned in the Test‐column. To calculate the F statistic, the degrees of freedom, and the (two‐sided) p value in columns 2–4, we used the test described in Tipton and Pustejovsky (2015) and implemented it using an extension to the function Wald_test and the “HTZ” small‐sample correction procedure contained in the R package Wald_test (Pustojevsky, 2019). The coefficients and variance estimates are from the model reported in Table 4, column 5.