Table 2.
Estimates the Effects of the Pill and Griswold on Weight at Birtha
| Independent variableb | Dependent variable = logarithm of the share of births at low birthweight
|
|||
|---|---|---|---|---|
| 2-1 | 2-2 | 2–3 | 2–4 | |
| PillSalesLegal × 1958–65c | −0.007 (0.007) | −0.005 (0.007) | 0.025 (0.017) | −0.009 (0.038) |
| PillSalesLegal × 1966–76d | 0.004 (0.016) | −0.004 (0.012) | 0.051 (0.031) | −0.001 (0.059) |
| Mean of dependent variable (not in logarithms) | 0.0776 | 0.0684 | 0.127 | 0.123 |
| No. of observations | 1,104 | 1,102 | 1,095 | 368 |
| R2 | 0.964 | 0.936 | 0.914 | 0.899 |
| Sample | All births | White births | Nonwhite births | Nonwhite births, South |
Source: Author’s regressions using data from the 1954–67 volumes of Vital Statistics and NCHS (2003).
a
Coefficients are least-squares estimates of τ using a restricted specification of equation 1 as described in the text. Heteroskedasticity-robust standard errors corrected for an arbitrary within-state covariance are in parentheses.
b
Each dummy variables equal to 1 when a sales of contraceptives were legal in the state where the mother resided and the birth occurred in the indicated period, and zero otherwise. The period 1950–57 is omitted.
c
Period after introduction of the Pill.
d
Period after Griswold when states with sales bans lifted these restrictions.