Testing Response Bias on Sensitive Topics in Self- Versus Enumerator-administered Surveys

Within a survey of adolescent males in Medellin schools (mainly 13-year old boys) the researchers are randomizing whether survey respondents answer sensitive questions themselves or are asked the questions by an enumerator. Questions are of three main varieties: (1) antisocial behaviors, (2) symptoms of depression, and (3) symptoms of anxiety.

Inclusion Criteria:All male students in seventh grade in a sample of public schools are eligible. The sample corresponds to a representative sample of public schools (we aim at 90 schools)Exclusion Criteria:Female students or students not in seventh grade