Table 4 Multilevel Logistic Regression Models Predicting Educational Outcomes
Attended CollegeScience/Engineering (Attended College)
Level 3—Villages/Urban BlocksN = 1763N = 1474
Distance to College km (ln)−0.08---
(0.06)
Urban (Rural)0.030.26
(0.18)(0.15)
Level 2—HouseholdsN = 4205N = 2875
Social Capital: Formal Sector Contacts−0.000.15**
(0.04)(0.05)
# Adults0.04−0.00
(0.03)(0.04)
# Children−0.020.01
(0.03)(0.05)
# Teens−0.05−0.13*
(0.04)(0.06)
Consumption (ln)0.34***0.32**
(0.08)(0.12)
Income (ln)−0.020.03
(0.03)(0.05)
Nonagriculture Income (Agriculture)−0.000.12
(0.10)(0.17)
Caste/Religion (Forward Castes)
Other Backward Class−0.090.29
(0.10)(0.16)
Scheduled Caste−0.03−0.07
(0.12)(0.20)
Scheduled Tribe0.48*0.18
(0.22)(0.30)
Muslim0.00−0.19
(0.15)(0.21)
Christian, Sikh, Jain0.000.41
(0.15)(0.23)
Level 1—IndividualsN = 5128N = 3353
On Pace in Education------
Female0.47***−0.48***
(0.08)(0.12)
Age−0.08***−0.02
(0.02)(0.03)
Father Alive and Present0.030.21
(0.16)(0.24)
Mother Alive and Present0.46−0.09
(0.26)(0.48)
Father Education 0–Std. 11 (Std. 12+)−0.46***−0.37*
(0.12)(0.18)
Mother Education 0–Std. 11 (Std. 12+)0.27−0.29
(0.16)(0.18)

Note: *p < 0.05, **p < 0.01, ***p < 0.001. Population averaged results with robust estimates are presented. State fixed effects are not presented but are included in each model. All variables are centered on the grand mean. 2005 weights are used.