Does respondent motivation affect item-nonresponse for split-ballot designed survey data? Comparative evidence from the European Social Survey

Abstract

AbstractSurvey quality would be enhanced if respondents willingly and accurately provided their responses. This seems feasible in ideal interview settings where respondents fully understand questions and provide truthful responses. Therefore, survey outcomes may be impacted by respondents’ feelings during the interview. The purpose of this study is to determine whether there is any relationship between respondent motivation, as operationalized by willingness and ability, and the item-nonresponse level for split-ballot designed questions. The 9th round of the European Social Survey (ESS9), a cross-national survey conducted across Europe, is the source of the data. This study assesses item-nonresponse using “don’t know” and “no answer” response choices for a set of questions designed with the split-ballot technique, which has not been dealt with much. In addition to the percentage distribution of item-nonresponse levels, the correlation analysis is used to understand the relationship between respondent motivation and item-nonresponse descriptively. Multivariable analyses use multiple linear regression modeling to explore the impact of respondent motivation on item-nonresponse, controlling for respondent and interviewer characteristics. Furthermore, bivariate relationships between outcome variable and covariates in the models were assessed using design-adjusted Wald-F tests. The findings pointed out that respondent motivation and item-nonresponse have a significant, negative, and moderate association. According to the complex sample design adjusted estimates of the statistical models, it may be able to reduce item-nonresponse with the increased motivation of respondents. This study offers several recommendations for questionnaire design, interviewing techniques, and interviewer evaluations for each respondent at the end.