Table 1.
Orthogonality tests and summary statistics: respondent time-invariant characteristics, as reported in endline
| Means and standard errors | P value from t test or f test | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mixed groups with education | Female groups with education | Mixed groups without education | Female groups without education | (1) vs. (4) | (2) vs. (4) | (3) vs. (4) | Joint orthogonality test of all | ||||||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | ||||||||
| Panel A: orthogonality tests | |||||||||||||||
| Housing characteristics | |||||||||||||||
| Household water source score (0–2, higher is cleaner) | 1.053 | 0.894 | 0.970 | 0.806 | 0.139 | 0.576 | 0.317 | 0.472 | |||||||
| (0.116) | (0.102) | (0.112) | (0.119) | – | – | – | – | ||||||||
| Has higher quality floor materials | 0.366 | 0.347 | 0.357 | 0.320 | 0.476 | 0.679 | 0.547 | 0.889 | |||||||
| (0.051) | (0.051) | (0.047) | (0.042) | – | – | – | – | ||||||||
| Has higher quality roof materials | 0.655 | 0.630 | 0.656 | 0.642 | 0.870 | 0.888 | 0.865 | 0.990 | |||||||
| (0.063) | (0.063) | (0.064) | (0.054) | – | – | – | – | ||||||||
| Education level | |||||||||||||||
| No schooling | 0.838 | 0.860 | 0.858 | 0.853 | 0.639 | 0.828 | 0.895 | 0.906 | |||||||
| (0.023) | (0.023) | (0.025) | (0.022) | – | – | – | – | ||||||||
| Up to primary school | 0.125 | 0.109 | 0.107 | 0.124 | 0.972 | 0.603 | 0.551 | 0.871 | |||||||
| (0.019) | (0.018) | (0.020) | (0.020) | – | – | – | – | ||||||||
| More than primary school | 0.037 | 0.030 | 0.036 | 0.023 | 0.165 | 0.423 | 0.197 | 0.445 | |||||||
| (0.008) | (0.007) | (0.008) | (0.006) | – | – | – | – | ||||||||
| Number of observations | 807 | 959 | 899 | 960 | – | – | – | – | |||||||
| Means and standard errors | All groups | P value from t test or f test | |||||||||||||
| Nonparticipant | Participant | Nonparticipant | Participant | Nonparticipant | Participant | Nonparticipant | Participant | Nonparticipant | Participant | (1) vs. (2) | (3) vs. (4) | (5) vs. (6) | (7) vs. (8) | (9) vs. (10) | |
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | (13) | (14) | (15) | |
| Panel B: selection analysis: mean (participant) − mean (nonparticipant) | |||||||||||||||
| Participation rate | 31.23% | 29.09% | 27.03% | 33.23% | 30.15% | ||||||||||
| Housing characteristics | |||||||||||||||
| Household water source score (0–2, higher is cleaner) | 1.051 | 1.060 | 0.859 | 0.978 | 1.058 | 0.728 | 0.905 | 0.606 | 0.964 | 0.834 | 0.937 | 0.361 | 0.008 | 0.007 | 0.039 |
| (0.122) | (0.138) | (0.101) | (0.148) | (0.113) | (0.132) | (0.123) | (0.125) | (0.057) | (0.073) | – | – | – | – | – | |
| Has higher quality floor materials | 0.369 | 0.361 | 0.358 | 0.320 | 0.354 | 0.366 | 0.323 | 0.313 | 0.350 | 0.338 | 0.872 | 0.613 | 0.829 | 0.854 | 0.666 |
| (0.052) | (0.062) | (0.052) | (0.079) | (0.054) | (0.051) | (0.047) | (0.050) | (0.025) | (0.030) | – | – | – | – | – | |
| Has higher quality roof materials | 0.695 | 0.567 | 0.657 | 0.565 | 0.650 | 0.671 | 0.630 | 0.665 | 0.657 | 0.618 | 0.077 | 0.243 | 0.781 | 0.580 | 0.280 |
| (0.065) | (0.079) | (0.064) | (0.086) | (0.072) | (0.071) | (0.062) | (0.059) | (0.033) | (0.037) | – | – | – | – | – | |
| Education level | |||||||||||||||
| No schooling | 0.819 | 0.881 | 0.856 | 0.870 | 0.856 | 0.863 | 0.854 | 0.852 | 0.847 | 0.866 | 0.044 | 0.715 | 0.830 | 0.946 | 0.281 |
| (0.028) | (0.023) | (0.027) | (0.032) | (0.028) | (0.031) | (0.023) | (0.035) | (0.013) | (0.016) | – | – | – | – | – | |
| Up to primary school | 0.140 | 0.091 | 0.117 | 0.090 | 0.108 | 0.104 | 0.124 | 0.123 | 0.121 | 0.103 | 0.049 | 0.355 | 0.902 | 0.980 | 0.204 |
| (0.024) | (0.017) | (0.021) | (0.024) | (0.024) | (0.024) | (0.024) | (0.027) | (0.012) | (0.012) | – | – | – | – | – | |
| More than primary school | 0.042 | 0.028 | 0.027 | 0.040 | 0.037 | 0.033 | 0.022 | 0.025 | 0.032 | 0.031 | 0.283 | 0.398 | 0.802 | 0.808 | 0.972 |
| (0.009) | (0.011) | (0.007) | (0.014) | (0.010) | (0.011) | (0.006) | (0.012) | (0.004) | (0.006) | – | – | – | – | – | |
| Number of observations | 555 | 252 | 680 | 279 | 656 | 243 | 641 | 319 | 2,532 | 1,093 | – | – | – | – | – |
Panel A standard errors clustered at the village level and reported in parentheses. Above variables chosen due to high likelihood of time invariance, thus used for orthogonality tests, given lack of baseline data. Column 8 reports P value from the f test from the regression of each row's variable on all three treatments (columns 1, 2, and 3).