Beschreibung

"Combining survey data with administrative data can make both sources more valuable to researchers, but often respondent consent is needed before such linkage can be done. Asking the linkage consent question at the beginning of the questionnaire yields higher consent rates than asking at the end. However, placing this question at the beginning of the survey may impact respondents' response behavior throughout the rest of the survey. In this study, we use techniques of causal analysis to investigate the effects of granting consent at the beginning of the survey on later response behavior. We find that responses to three of the four tested variables are not affected, but one variable did show increased measurement error." (Author's abstract, IAB-Doku) ((en))