At the request of the Bureau of Justice Statistics (BJS), the Center for Behavioral Science Methods (CBSM) conducted cognitive testing of the redesign of the 2018 Police-Public Contact Survey (PPCS), a supplement to the National Crime Victimization Survey (NCVS). The PPCS examines the types of contacts respondents have had with police in the previous 12 months, whether these contacts were self-initiated or police-initiated, the number of contacts that occurred, the outcome of the encounter, and how respondents perceived the behaviors of the officers involved. The questionnaire is comprised of two parts – a screener to determine whether the respondent had any contact with police, and if so, a second component of the questionnaire capturing the details about the most recent encounter.

Due to findings from the previous field administration of the survey, testing focused on the PPCS screener questions. The findings from our testing indicate that respondents struggled to correctly include some of their contacts in response to the screener question intended to measure that type of contact. Respondents were not always able to make clear distinctions in how they classified crimes, disturbances, suspicious activity, non-crime emergencies, and non-emergency assistance. As a result, they reported the same incident in response to multiple questions because they were not sure which question was a more appropriate fit or whether a question was going to be asked that would better fit their situation.