Table A1

Covariate Balance (Full Sample with Imputed Values)

VariableSampleaMean Value PSAMean Value Non-PSAbDiff. Mean ValueAvg. Raw eQQ Diff.cMean eCDF Diff.dp-Values for t-Test/K-S Test
1992 forest cover (ha)Unmatched86.1338.2147.9220.090.140.03/0.04
 Matched86.1368.9917.134.960.040.16/0.84
Farm size (ha)Unmatched165.1171.4393.6840.000.190.06/ < 0.01
 Matched165.11108.4656.6517.000.080.17/0.22
Participation in forestry programsUnmatched0.320.150.170.170.140.02
 Matched0.320.280.030.030.020.22
Distance to extension office (km)Unmatched30.1024.605.505.000.100.02/0.08
 Matched30.1027.782.331.000.030.23/0.71
% Farm on steep slopesUnmatched38.4025.1413.2610.000.13< 0.01/0.06
  38.4038.110.290.000.040.83/0.87
Forest cover change 1986–1992Unmatched− 11.35− 10.83− 3.025.890.140.94/ < 0.01
 Matched− 11.35− 8.33− 0.522.000.080.46/0.26
  • Note: PSA, Costa Rica’s Programa de Pagos por Servicios Ambientales.

  • a N = 202; 50 PSA farms.

  • b Weighted means for matched controls.

  • c Mean (for categorical covariate) or median (for continuous covariate) difference in the empirical quantile-quantile plot of treatment and control groups on the scale in which the covariate is measured.

  • d Mean eCDF = mean differences in empirical cumulative distribution function.