Surveys are increasingly offering web as part of a mixed mode design to elicit survey data from respondents. One concern with this approach is that
respondents, in the absence of an interviewer, must be sufficiently motivated to provide accurate, complete, and honest data. Respondents that do not put forth the effort to ensure the accuracy of their responses engage in what survey researchers have dubbed satisficing behavior. Satisficing behavior can lead to data quality issues. Past studies have shown the utility of incorporating both proactive and reactive interventions into survey designs to prevent or discourage this behavior (Conrad et al., 2011; Zhang, 2013). In this study, we use the 2013 National Survey of College Graduates (NSCG) web paradata to classify respondents based on their tendency to satisfice. Specifically, we use factors such as response time, non-differentiation among answers, omissions, and early termination to identify degrees of satisficing. We also look for correlations between satisficing behaviors and demographic characteristics to help characterize these respondents. Successful classification is instrumental in targeting interventions planned for future studies based on the type of satisficing. Conrad, F. G., Tourangeau, R., Couper,M. P., & Zhang, C. (2011). Interactive Interventions in Web Surveys Can Increase Response Accuracy.