table 3

The Impact of the Minimum School Starting Age on First Grade Enrollment

MalesFemales
(1)(2)(3)(4)(5)(6)(7)(8)
Age of youngest children (months)0.0340.1320.0590.1800.0340.1410.0460.175
(0.005)(0.023)(0.008)(0.013)(0.008)(0.021)(0.010)(0.011)
Youngest×(January–March)0.1420.1620.1360.145*
(0.021)(0.028)(0.021)(0.013)
Youngest×(April–June)0.1280.166*0.1380.184*
(0.024)(0.010)(0.020)(0.013)
Youngest×(July–September)0.113*0.160*0.124*0.160
(0.030)(0.015)(0.023)(0.030)
Includes state-specific linear trendNoNoYesYesNoNoYesYes
Sample size102,958102,958102,958102,95899,53499,53499,53499,534

Note: The dependent variable is one if the individual is enrolled in Grade 1 or higher and zero otherwise. All models are population weighted and clustered at the state of residence level. Heteroskedastic-consistent standard errors are in parentheses. Bold coefficients are significant at the 5 percent level. Columns 1, 3, 5, and 7 include indicators for the existence of publicly funded kindergarten, sex, race, state of residence, and census-division-specific year cohorts. Columns 2, 4, 6, and 8 further include birth quarter indicators and interactions between state of residence and birth quarter and year and birth quarter. The sample includes six-year-olds from the 1960, 1970, and 1980 U.S. Censuses. An asterisk in Column 2, 4, 6, or 8 indicates that youngest plus the specified interaction effect (i.e., β1 + β2, β1 + β3, or β1 + β4) is nonzero at the 10 percent level.