Skip to main content
. Author manuscript; available in PMC: 2013 Jan 1.
Published in final edited form as: J Dev Econ. 2012 Jan 1;97(1):118–129. doi: 10.1016/j.jdeveco.2010.12.003

Table 1.

Summary sample sizes and characteristics.

Panel A: sample sizes Obs
Offered HIV test in 2004 3185
Accepted HIV test in 2004 2894
Main samplea 2767
Panel B: baseline characteristics Obs (1) Mean (2) SD (3)

Demographics Male 2767 0.455 0.498
Age 2767 33.373 13.622
Married 2762 0.709 0.454
HIV status 2767 0.062 0.241
Had sex in the last 12 months 2761 0.760 0.427
Distance from VCT 2767 2.013 1.263
Educationb 2767 3.325 3.706
Assetsb 2767 4.899 2.582
Tribe Chewa 2767 0.248 0.432
Tumbuka 2767 0.292 0.455
Yao 2767 0.248 0.432
Other 2767 0.021 0.143
Religion Catholic 2767 0.145 0.352
CCAP 2767 0.162 0.369
Other Christian 2767 0.220 0.415
Muslim 2767 0.222 0.416
Other 2767 0.033 0.178
Incentives and neighbors Received a non-zero incentive 2767 0.782 0.413
Amount of incentive (Kwacha) 2767 106.773 95.082
Respondent got results 2767 0.696 0.460
Number of neighbors in band 0–0.2 km 2767 8.323 8.195
Number of neighbors in band 0.2–0.4 km 2767 10.252 10.286
Number of neighbors in band 0.4–0.6 km 2767 11.674 10.703
Number of neighbors in band 0–0.5 km 2767 24.161 19.415
a

The Main sample consists of those who accepted an HIV test, did not test as indeterminant, and who had non-missing information for age, the amount of the incentive received and have non-missing gps coordinates.

b

There are 284 individuals for which education (as measured by years of schooling) and/or the number of assets owned are missing. To include these individuals in the analysis we have imputed their education and assets as the mean (separately by gender) and include a dummy for whether they had missing asset information and missing education information in all specifications.