Despite extensive research and efforts to boost survey response rates, nonresponse studies continue to show a declining trend in response. Survey climate and respondents’ attitudes towards surveys are often cited as key theoretical concepts for understanding nonresponse. To provide empirical data on survey attitude and its impact on nonresponse rates, de Leeuw et al. (2019; 2022) developed the Survey Attitude Scale (SAS). This scale reveals a replicable three-dimensional factor structure encompassing survey enjoyment, survey value, and survey burden. The reliability of the corresponding subscales was found to be satisfactory for all three dimensions. Moreover, (scalar) measurement equivalence was established across three panels that used two languages (German and Dutch) and three measurement modes (online, phone, and mail). In this study, we examine the extent to which respondents’ survey attitudes as measured by the SAS are stable across waves and evaluate their explanatory and predictive power regarding survey nonresponse and panel dropout. The results show that two-thirds of the variance captured by the SAS reflects enduring aspects of respondents’ survey attitudes, while one-third reflects situational factors. The SAS demonstrates strong explanatory power for both survey nonresponse and panel dropout beyond what can be accounted for by respondents’ psychographic and sociodemographic profiles. While psychographic and sociodemographic profiles are more effective at forecasting nonresponse, the SAS proves to be a better predictor of panel dropout. We recommend incorporating the SAS in the initial waves of a panel to identify respondents with a high likelihood of dropping out.
- Forthcoming at Survey Research Methods
- The project has been presented at the International Workshop on Household Survey Nonresponse 2016 and the General Online Research Conference 2017