Table 4

Placebo Effect: Estimates of Average Treatment Effect on the Treated for Lake Recreation

Participation in 2003 IDifference in Participation IITotal Trips in 2003 IIIDifference in Total Trips IV
Treatment Group 1: Unemployed, Part-time, and Retired
Nearest neighbor with replacement  0.019 (0.072)  0.007 (0.075)–0.069 (1.223)  0.226 (0.946)
Nearest five neighbors with replacement  0.035 (0.057)–0.024 (0.056)  0.748 (1.006)  0.045 (0.789)
Radius matching (caliper = 0.5 ×SD)  0.011 (0.047)–0.031 (0.052)–0.134 (0.713)  0.252 (0.676)
Kernel matching  0.019 (0.050)–0.035 (0.058)  0.121 (0.750)  0.224 (0.690)
Treatment Group 2: Unemployed and Part-time
Nearest neighbor with replacement  0.108 (0.094)–0.075 (0.100)–0.564 (1.545)  0.797 (1.225)
Nearest five neighbors with replacement  0.074 (0.073)–0.044 (0.073)  0.125(1.150)  0.361 (0.659)
Radius matching (caliper = 0.5 ×SD)  0.082 (0.065)–0.072 (0.071)  0.48 (0.741)–0.491 (0.647)
Kernel matching  0.077 (0.061)–0.074 (0.068)  0.295 (0.774)–0.532 (0.637)
Treatment Group 3: Retired
Nearest neighbor with replacement–0.039 (0.080)  0.005 (0.075)–0.228 (1.718)  0.745 (1.305)
Nearest five neighbors with replacement–0.011 (0.066)  0.017 (0.068)  0.222 (1.350)  0.277 (1.135)
Radius matching (caliper = 0.5 ×SD)–0.051 (0.060)  0.01 (0.064)–0.542 (1.072)  0.286 (1.043)
Kernel matching–0.047 (0.061)  0.007 (0.064)–0.398 (1.141)  0.158 (1.105)
  • 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. No statistical significance at the 10%, 5%, and 1% levels.