Skip to main content
. Author manuscript; available in PMC: 2011 Jun 15.
Published in final edited form as: Am Econ Rev. 2008 Dec 1;98(5):1829–1863. doi: 10.1257/aer.98.5.1829

Table 8.

Probit Estimates: Reactions to Learning HIV Results

Dependent variable: Bought any condoms
OLS (1) IV (2) Probit (marginal effects) (3) Bivariate probit (marginal effects) (4)
Panel A: HIV positive
Got results 0.427*** (0.149) 0.273 (0.200) 0.437*** (0.138) 0.370*** (0.141)
Male 0.038 (0.115) 0.024 (0.117) 0.043 (0.129) 0.026 (0.131)
Age 0.001 (0.040) −0.018 (0.043) 0.004 (0.039) −0.009 (0.038)
Age2 0.000 (0.001) 0.000 (0.001) 0.000 (0.000) 0.000 (0.000)
Simulated average distance 0.025 (0.074) 0.026 (0.072) 0.026 (0.074) 0.025 (0.074)
Rumphi −0.084 (0.179) −0.08 (0.177) −0.126 (0.195) −0.117 (0.193)
Constant 0.055 (0.823) 0.511 (0.974)
Sample size 52 52 52 52
R2 0.21 0.18 0.18
Mean 0.38 0.38 0.38 0.38

Panel B: HIV negative
Got results −0.023 (0.025) −0.072 (0.061) −0.033 (0.029) −0.128 (0.088)
Male 0.113*** (0.028) 0.111*** (0.027) 0.118*** (0.029) 0.114*** (0.029)
Age −0.007 (0.005) −0.006 (0.005) −0.005 (0.005) −0.004 (0.005)
Age2 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000)
Simulated average distance −0.052** (0.022) −0.053** (0.022) −0.043** (0.018) −0.044** (0.018)
Rumphi −0.362*** (0.040) −0.368*** (0.041) −0.356*** (0.036) −0.365*** (0.037)
Constant 0.754*** (0.109) 0.784*** (0.116)
Sample size 956 956 956 956
R2 0.19 0.19 0.18
Mean 0.25 0.25 0.25 0.25

Notes: Sample includes HIV-positive and negative respondents in Balaka and Rumphi who were tested for HIV, were reinterviewed in 2005, and who had sex in 2004. Robust standard errors clustered by village (for 57 villages) with district fixed effects are in parentheses. Coefficients in column 1 are linear OLS estimates. Coefficients in column 2 are linear IV estimates where knowing HIV status is instrumented by having any nonzero incentive, the total amount of the incentive, total amount squared, distance from the HIV results center, distance-squared, and all terms interacted with gender. Column 3 presents marginal probit estimates, which do not control for endogenous selection into learning HIV results. Column 4 presents the marginal effects of a seemingly unrelated bivariate probit, controlling for endogenous selection into learning HIV results (see text).

***

Significantly different from zero at 99 percent confidence level.

**

Significantly different from zero at 95 percent confidence level.

*

Significantly different from zero at 90 percent confidence level.