Table 3.

Reduced-form RD estimates of educational attainment

	(1)	(2)	(3)
Dependent variable:	High school graduate
I(Grade 8 GPA < 2.0)	0.185**	0.175**	0.157**
	(0.073)	(0.073)	(0.073)
N	1,405	1,405	1,405
R2	0.101	0.107	0.151
Akaike's information criterion	323.5	321.8	259.5
Dependent variable:	Postsecondary enrollment, year 5
I(Grade 8 GPA < 2.0)	0.163	0.162	0.134
	(0.101)	(0.0988)	(0.0920)
N	1,405	1,405	1,405
R2	0.0906	0.110	0.146
Akaike's information criterion	1,689.3	1,667.0	1,616.9
Dependent variable:	Postsecondary enrollment, year 6
I(Grade 8 GPA < 2.0)	0.128	0.127	0.0976
	(0.100)	(0.098)	(0.092)
N	1,405	1,405	1,405
R2	0.0957	0.115	0.155
Akaike's information criterion	1,716.5	1,694.4	1,637.9
Demographic controls	—	Yes	Yes
Full controls	—	—	Yes

Each cell contains the results of a separate regression of the effect of I(Grade 8 GPA_i < 2.0) on graduation outcomes confirmed by the California Department of Education from any public school in the state. Our preferred measure of high school graduation includes imputed outcomes for district and state leavers (i.e., enrolled in private school, public school out-of-state or left country) who enrolled seamlessly in postsecondary or were on track to graduate based on internal district benchmarks for credits earned. All models contain linear splines. Demographic controls include student sex and race/ethnicity. The full controls include grade 8 (i.e., prior to treatment status) special education identification, English learner status, attendance and suspension history. Grade 8 GPA is centered at 2.0. Robust SEs are reported in parentheses. *P < 0.10, **P < 0.05, and ***P < 0.01.