Table 4.
Heterogeneity analysis results by 2SLS.
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| SRH | IADL | Physical activity ability | CES–D | SWB | |
| Panel A: Gender | |||||
| Father | 1.058*** | 0.241 | 1.257** | −3. 172** | 0.296 |
| Mother | 0.469* | 0.397 | 0.808 | −0.473 | −0.008 |
| Chow test | 23.49*** | 22.80*** | 16.80*** | 17.39*** | 2.78 |
| Panel B: Number of children | |||||
| Only one child | −3.811 | −1.899 | −7.147 | 7.698 | −1.848 |
| More than one child | 0.957*** | 0.553** | 1.456*** | 0.820 | −0.121 |
| Chow test | 10.06** | 15.44*** | 9.56** | 30.13*** | 11.66** |
| Panel D: Region | |||||
| East | 0.068 | −0.907 | 1.483 | 10.254 | −0.960* |
| Middle | 1.001*** | 0.969*** | 2.212*** | −0.514 | 0.057 |
| West | −3.586** | −0.794 | −0.779 | 8.523 | −3.391** |
| Chow test | 47.13*** | 23.25*** | 18.33** | 14.78** | 19.92*** |
***
p < 0.01,
**
p < 0.05, and
*
p < 0.1. Robust standard errors are in parentheses. Control variables included the three types of variables mentioned above: parental characteristics, children and family characteristics, and village–level characteristics, as well as year dummy variables and province dummy variables.