Bayesian Factor Mixture Modeling with Response Time for Detecting Careless Respondents

Careless respondents inject noise into data which can distort research findings and compromise model fit. To address this, factor mixture modeling (FMM) has been widely used to identify careless respondents. Traditionally, researchers have relied on reverse-worded questions in FMM to facilitate the detection of careless responding. With the rise of online data collection platforms, response time has appeal as a means for detecting careless respondents. We introduce a Bayesian factor mixture model that utilizes response time to identify careless respondents. By jointly modeling responses and response time, this approach effectively identifies careless individuals exhibiting rapid and random response styles. Through simulation studies, we found that: (1) the proposed model achieves high estimation accuracy of key model parameters (i.e., loadings and intercepts); (2) it demonstrates high accuracy and sensitivity in correctly classifying respondents as either attentive or careless; and (3) it maintains classification error rates at an acceptable level. An additional benefit is that incorporating response time into the model enhances model convergence as well as accuracy of classification and estimation. An empirical study tests the applicability of the proposed model in real-world scenarios, comparing its performance to the traditional method based on reverse-worded questions. The results underscore the practical advantages of enriching FMM with collateral response time information in excluding careless responses and improving data quality.