The Implications of Questionnaire Redesign on Trend Estimates in the 2011 Police Public Contact Survey

• Activity Number: 649
• Type: Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Glynis Ewing* and Marcus Berzofsky and Lynn Langton and Michael Planty
• Companies: RTI International and RTI International and Bureau of Justice Statistics and Bureau of Justice Statistics
• Keywords: questionnaire design ; ratio adjustment ; split-sample experiment ; trend estimation ; measurement error ; redesign

Abstract

Each time an established periodic survey introduces a newly redesigned field instrument, there is a risk of introducing between-year measurement error that can interrupt the continuation of statistical trend. Regardless of improvements to the measures of interest, a redesigned questionnaire introduces potential instability to the validity of trend estimates for outcomes of interest between survey years. This paper will discuss sources of measurement error specific to redesigned survey instruments, describe methods for evaluating estimate comparability between survey years in the event of a redesign, and provide a solution for adjustment to correct for measurement error, illustrated by select examples from the 2011 Police-Public Contact Survey (PPCS).