Robustness checks
| HIV public stigma | Prejudice | Disclosure concerns | ||
| (1) | (2) | (3) | (4) | |
| Share of mobile phone ownership | −0.423* | −0.257* | −0.541*** | −0.113 |
| (0.205) | (0.119) | (0.143) | (0.198) | |
| Observations | 817 | 817 | 817 | 817 |
| Region and year FE | Y | Y | Y | Y |
| Population weights | Y | Y | Y | Y |
| Community-level confounders | Y | Y | Y | Y |
| Underidentification test (p value) | <0.01 | <0.01 | <0.01 | <0.01 |
| Kleibergen-Paap rk Wald F statistic | 8.69 | 18.95 | 21.74 | 21.74 |
Effects are reported as beta coefficients. SEs in parentheses are clustered at community level. Full-result tables that include all confounders are available in online supplemental appendix C.
Source: GDHS 2008 and 2014.
*p<0.05, **p<0.01, ***p<0.001.
FE, fixed effect; N, no; Y, yes.