Table 3.

Estimation of number of live births (2SLS model)

	1965 (N = 3662)	1967 (N = 4868)	1976 (N = 4678)	1980 (N = 3852)	1985 (N = 3817)
First stage: estimation of number of contraceptives known
Listen to radio regularly	0.36***	0.82***	0.51***	0.21*	0.30***
Read newspapers regularly	1.88***	1.59***	1.18***	0.83***	1.07***
Watch TV regularly	N/A	0.72***	0.83***	0.29**	−0.03
Read magazines regularly	N/A	N/A	0.89***	0.84***	0.54***
Join organizations	N/A	N/A	0.50***	0.41**	0.52***
F statistics	39.91	86.74	51.75	17.23	35.42
Second stage: estimation of number of live births
Number of contraceptives known	−0.15***	−0.10**	−0.13***	−0.18***	−0.20***
No son	−1.68***	−1.76***	−1.24***	−0.96***	−0.92***
Age rage 18–22 (base: 38–44)	−4.09***	−3.86***	N/A	−2.07***	−1.76***
Age range 23–27	−3.31***	−3.16***	−2.25***	−1.46***	−1.20***
Age rage 28–32	−1.98***	−2.04***	−1.45***	−0.79***	−0.72***
Age range 33–37	−0.76***	−0.96***	−0.58***	−0.22***	−0.33***
Working outside of family	−0.24**	−0.13*	−0.13**	−0.11**	−0.18***
Education years 12 or above (base: 7–11)	−0.19	−0.54***	−0.33***	−0.25***	−0.20**
Education years 0–6	−0.02	0.15+	0.20***	0.06	−0.23*
Living in a city	−0.67**	−0.11	−0.06	−0.17*	−0.06
Live with parents-in-law	−0.12+	−0.06	0.01	0.05	0.04
Husband’s income	N/A	0.07	0.03*	0.05	0.04
R-squared	0.50	0.50	0.43	0.40	0.36
Over-identification test (P-value)	0.4707	0.0826	0.1415	0.0307	0.0099

+ P < 0.10, * P < 0.05, ** P < 0.01, *** P < 0.001

Standard errors are robust and clustered in township level. N/A indicates no information in that category. Also control for ethnicity, husband’s education, and township fixed effects. The first stage equation also controls for other individual characteristics [see Eq. (2)]