This data set is a subset of the data from the randomized response technique survey experiment conducted in Nigeria to study civilian contact with armed groups. The survey was implemented by Blair (2014).



A data frame containing 2457 observations. The variables are:

  • Quesid: Survey ID of civilian respondent.

  • rr.q1: Randomized response survey item using the Forced Response Design asking the respondent whether they hold direct social connections with members of armed groups. 0 if no connection; 1 if connection.

  • cov.age: Age of the respondent.

  • cov.asset.index: The number of assets owned by the respondent from an index of nine assets including radio, T.V., motorbike, car, mobile phone, refrigerator, goat, chicken, and cow.

  • cov.married: Marital status. 0 if single; 1 if married.

  • Education level of the respondent. 1 if no school; 2 if started primary school; 3 if finished primary school; 4 if started secondary school; 5 if finished secondary school; 6 if started polytechnic or college; 7 if finished polytechnic or college; 8 if started university; 9 if finished university; 10 if received graduate (masters or Ph.D) education.

  • cov.female: Gender. 0 if male; 1 if female.

  • civic: Whether or not the respondent is a member of a civic group in their communities, such as youth groups , women's groups, or community development committees.


Blair, Graeme, Kosuke Imai and Yang-Yang Zhou. (2014) Replication data for: Design and Analysis of the Randomized Response Technique.


Blair, G. (2014). "Why do civilians hold bargaining power in state revenue conflicts? Evidence from Nigeria."

Blair, Graeme, Kosuke Imai and Yang-Yang Zhou. (2014) "Design and Analysis of the Randomized Response Technique." Working Paper. Available at