Robustness Check of Estimates of Average Treatment Effect on the Treated for Lake Recreation (with Different Group of Covariates)
| Participation in 2009 I | Difference in Participation II | Total Trip in 2009 III | Difference in Total Trip IV | |
|---|---|---|---|---|
| Treatment Group 1: Unemployed, Part-time, and Retired | ||||
| Nearest neighbor with replacement | 0.046 (0.054) | 0.084 (0.058) | –0.378 (1.173) | –0.147 (0.979) |
| Nearest five neighbors with replacement | 0.073 (0.051) | 0.114** (0.056) | 0.226 (1.040) | 0.173 (0.945) |
| Radius matching (caliper = 0.5 × SD) | 0.083 (0.052) | 0.122** (0.054) | 0.264 (0.973) | 0.398 (0.915) |
| Kernel matching | 0.085 (0.052) | 0.123** (0.053) | 0.397 (1.031) | 0.43 (0.891) |
| Treatment Group 2: Unemployed and Part-time | ||||
| Nearest neighbor with replacement | 0.082 (0.078) | 0.174* (0.091) | –1.69 (1.247) | 1.622 (1.204) |
| Nearest five neighbors with replacement | 0.097 (0.066) | 0.191** (0.086) | –1.294 (1.083) | 1.902* (1.106) |
| Radius matching (caliper = 0.5 × SD) | 0.118* (0.065) | 0.185** (0.082) | –1.059 (1.124) | 1.336 (1.068) |
| Kernel matching | 0.105 (0.065) | 0.185** (0.082) | –1.152 (1.100) | 1.352 (1.070) |
| Treatment Group 3: Retired | ||||
| Nearest neighbor with replacement | 0.064 (0.070) | 0.051 (0.070) | 1.875 (1.450) | –0.445 (1.311) |
| Nearest five neighbors with replacement | 0.054 (0.068) | 0.043 (0.067) | 1.739 (1.448) | –0.312 (1.280) |
| Radius matching (caliper = 0.5 × SD) | 0.058 (0.065) | 0.058 (0.062) | 1.266 (1.434) | –0.246 (1.241) |
| Kernel matching | 0.064 (0.069) | 0.055 (0.061) | 1.517 (1.433) | –0.357 (1.259) |
Note: Caliper is chosen as one-half of standard deviation of propensity score. Standard errors reported for nearest neighbor estimators are Abadie-Imbens (AI) robust standard errors. Standard errors reported for radius matching and kernel matching estimators are obtained from bootstrapping with 1,000 replications.
↵*, ** Statistical significance at the 10% and 5% levels, respectively.