Table 3

Estimates of Average Treatment Effect on the Treated for Lake Recreation

Participation in 2009 IDifference in Participation IITotal Trips in 2009 IIIDifference in Total Trips IV
Treatment Group 1: Unemployed, Part-time, and Retired
Nearest neighbor within caliper  0.113* (0.066)  0.113 (0.073)  1.019 (1.373)  0.725 (1.268)
Nearest five neighbors  0.103* (0.056)  0.103* (0.060)  1.123 (1.104)  0.186 (1.007)
Radius matching within caliper  0.083* (0.048)  0.123** (0.052)  0.719 (0.823)  0.433 (0.772)
Kernel matching  0.084* (0.051)  0.125** (0.056)  0.755 (0.903)  0.411 (0.794)
Treatment Group 2: Unemployed and Part-time
Nearest neighbor within caliper  0.222*** (0.081)  0.189* (0.115)  0.444 (1.564)–0.456 (1.324)
Nearest five neighbors  0.137** (0.066)  0.117 (0.087)  0.732 (1.119)  0.302 (1.043)
Radius matching within caliper  0.139** (0.066)  0.138* (0.083)  0.427 (0.958)  0.484 (0.916)
Kernel matching  0.132** (0.063)  0.138* (0.083)  0.2 (0.936)  0.544 (0.933)
Treatment Group 3: Retired
Nearest neighbor within caliper  0.00 (0.077)  0.034 (0.077)–1.121 (1.927)–0.357 (1.482)
Nearest five neighbors  0.034 (0.063)  0.041 (0.072)–0.238 (1.464)  0.266 (1.268)
Radius matching within caliper  0.008 (0.060)  0.066 (0.065)–0.492 (1.227)–0.213 (1.097)
Kernel matching  0.02 (0.063)  0.076 (0.066)–0.329 (1.269)–0.008 (1.115)
  • Note: The caliper is chosen as one-half of standard deviation of propensity score. Standard errors reported for nearest neighbor estimators are Abadie-Imbens 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%, 5%, and 1% levels, respectively.