ACSPRI Conferences, RC33 Eighth International Conference on Social Science Methodology

Font Size:  Small  Medium  Large

Interviewer Effects on Measurement Error in Panel Surveys

Ulrich Krieger

Building: Law Building
Room: Breakout 1 - Law Building, Room 024
Date: 2012-07-12 01:30 PM – 03:00 PM
Last modified: 2011-12-12


Assigning the same interviewers to panel respondents over panel waves is considered to be the ideal strategy when fielding panel surveys. Respondents and interviewers are said to build a trusted relationship over time and thus prevent panel attrition.
Having the same interviewer recording the answers of the same respondent multiple times may introduce a bias in the measurements. Perceptions and attitudes of the interviewer might influence the recorded answers more strongly in this cases than when a change of interviewer occured. An interviewer interviewing the same case again might be less aware of change in the respondents lifes or repeat the same errors while recodring the answers.
Using wave four data of the Survey of Health, Aeging and Retirement in Europe (SHARE) collected in Germany effects on measurement of interviewers is examined. Rates of change and stabiliry in lfe domains such as health, employment and family are compared between respondents interviewed by their respective wave three interviewers and those that had been assigned new interviewers. The trade-off between a reduced attrition and measurement error is beeing discussed.