FIGURE 2.
HIV testing results (per 3-month interval). A, The share of participants that tested for HIV at a Bienestar site in a given 3-month interval (all 4 intervals pooled). Brackets are 95% CIs of the difference between the treatment arms and the comparison arm, estimated using a linear probability model with standard errors clusters by individual. B, The difference between each treatment arm and the comparison arm in each of the 3-month intervals with 95% CIs estimated in the same way as in A.
Panel A reports the share of participants that tested for HIV at a Bienestar site in a given three-month interval (all four intervals pooled). Brackets are 95% confidence intervals of the difference between the treatment arms and the comparison arm, estimated using a linear probability model with standard error clusters by individual. Panel B reposts the difference between each treatment arm and the comparison arm in each of the three-month intervals with 95% Confidence intervals estimated in the same way as in Panel A.