Table 4.
Robustness: The importance of open defecation is robust to different model specifications.
| (1)a | (2) | (3) | |
|---|---|---|---|
| control variables same as in Table 2, | Column 5 | Column 5 | Column 5b |
| open defecation in child’s locality | −0.350*** | −0.292** | −0.335*** |
| (0.120) | (0.117) | (0.126) | |
| region by time FE | yes | ||
| region FE | yes | yes | |
| 2000 data | yes | ||
| 2005 data | yes | yes | yes |
| 2010 data | yes | yes | yes |
| n (children under 5) | 6389 | 6389 | 9698 |
Standard errors clustered by survey primary sampling unit in parentheses. Two-sided p-values: *** p < 0.01, ** p < 0.05, * p < 0.1. All specifications include 120 age-in-month dummies, separately for boys and girls. In each column, the difference from Table 2, Column 5 are in bold.
a
This is the same model as Table 2, Panel (b), Column 5. Coefficient is repeated for comparison purposes.
b
Province-part characteristics (consumption and density) have been excluded from this model as they are not available for the year 2000. Water source has also been excluded because of inconsistencies in data collection from 2005 to 2010.