Table 8 Alternative IV estimates: Lewbel's Heteroskedasticity-Based IV approach
From: The effect of Internet use on adolescent nutritional outcomes: evidence from China
ZBMI | Overweight | |
|---|---|---|
(1) | (2) | |
Internet | 0.469*** | 0.067** |
(0.122) | (0.033) | |
Age | − 0.074*** | − 0.025*** |
(0.007) | (0.002) | |
Gender | 0.209*** | 0.097*** |
(0.020) | (0.005) | |
Education | − 0.015** | − 0.005*** |
(0.007) | (0.002) | |
Exercise | − 0.008** | − 0.002*** |
(0.003) | (0.001) | |
Household size | − 0.007 | − 0.001 |
(0.006) | (0.002) | |
Household income | 0.013 | − 0.002 |
(0.011) | (0.003) | |
Urban | 0.074*** | 0.018** |
(0.027) | (0.007) | |
Constant | 0.616*** | 0.495*** |
(0.181) | (0.051) | |
Year | Yes | Yes |
Province | Yes | Yes |
Observations | 19, 317 | 19, 317 |
R-squared | 0.054 | 0.079 |
Kleibergen-Paap rk LM statistic | 287.765 | 287.765 |
Kleibergen-Paap rk Wald F statistic | 18.142 | 18.142 |
Hansen J statistic | 0.155 | 0.155 |