TABLE III.

Effect of Public Treatment on Sign-up Decision

	(1)	(2)	(3)	(4)	(5)	(6)
Dependent variable	Dummy: student signed up for the SAT prep course
Public treatment	−0.1083*** [0.040]	−0.1195*** [0.040]	−0.1231*** [0.040]	0.0157 [0.033]	0.0095 [0.032]	0.0092 [0.031]
Inference robustness (p-values)
Robust standard   errors	.007	.003	.002	.631	.766	.766
Wild bootstrap	.012	.006	.002	.596	.706	.720
Permutation test	.006	.001	.002	.480	.748	.749
Mean of private  take-up		0.717			0.917
Includes individual  covariates	No	Yes	Yes	No	Yes	Yes
Includes classroom  and surveyor  fixed effects	No	No	Yes	No	No	Yes
Observations	560	531	531	265	258	258
R-squared	0.013	0.042	0.104	0.001	0.035	0.139
Sample	Nonhonors classes			Honors classes

Notes. Columns (1) to (3) restrict the sample to nonhonors classes, and columns (4) to (6) restrict to honors classes. Columns (1) and (4) present OLS regressions of a dummy variable for whether the student signed up for the SAT prep course on a public sign-up dummy. columns (2) and (5) replicate and add individual covariates (age and dummies for male and Hispanic). columns (3) and (6) further add surveyor and classroom fixed effects. Robust standard errors in brackets. *** p < .01, ** p < .05, * p < .1. See Section III.D for a description of the wild bootstrap and permutation test procedures.