Table 3.
Treatment effects of left-behind experience in the whole sample.
| Matching Methods |
Nearest Neighbor Matching (k = 1) |
Nearest Neighbor Matching (k = 4) |
Nearest Neighbor Matching with Caliper (k = 4, caliper = 0.004) |
Caliper Matching (caliper = 0.004) |
Spline Matching |
|---|---|---|---|---|---|
| ATT | 4.419 *** (0.509) |
4.456 *** (0.498) |
4.458 *** (0.498) |
4.485 *** (0.489) |
4.517 *** (0.485) |
| ATU | 5.870 *** (0.831) |
5.913 *** (0.854) |
5.919 *** (0.855) |
5.537 *** (0.710) |
5.778 *** (0.723) |
| ATE | 5.834 *** (0.816) |
5.877 *** (0.839) |
5.883 *** (0.840) |
5.511 *** (0.700) |
5.747 *** (0.713) |
Note: *** p < 0.01. Standard errors in parentheses are estimated by bootstrap method with resampling 500 times. Each treatment observation is matched to control observations with replacement when performing nearest neighbor matching. ATT indicates the average treatment effect on the treated; ATU indicates the average treatment effect on the untreated; ATE indicates the average treatment effect.