Personal Values Strongly Predict Study Dropout

Abstract

Participant dropout poses significant selectivity problems in longitudinal studies. Although it is often assumed that participant’s value structures predict future participation and dropout, there is insufficient evidence supporting this proposition. This study aims to contribute to the literature by clarifying the relationship between participants’ personal values and study dropout. Data of the 2008 baseline sample of the German Aging Survey was used to predict future survey (non-)participation in subsequent follow-ups (N = 4442). Personal values were measured using the Portraits Value Questionnaire. It was found that different values had differential effects on survey dropout: Whereas higher achievement, security and power values predicted increased odds of dropout, higher benevolence, stimulation and self-direction values predicted decreased odds. Additionally, being older and having lower income were also associated with increased dropout. Therefore, participants’ personal values seem to have major influence on participant’s study participation behaviour. Given the importance of personal values in human behaviour, longitudinal research must account for these relationships for providing accurate scientific findings.