A Beta Mixture Model for Careless Respondent Detection in Visual Analogue Scale Data

Visual Analogue Scales (VASs) are increasingly popular in psychological, social, andmedical research. However, VASs can also be more demanding for respondents, potentiallyleading to quicker disengagement and a higher risk of careless responding. Existing mixturemodeling approaches for careless response detection have so far only been available forLikert-type and unbounded continuous data but have not been tailored to VAS data. Thisstudy introduces and evaluates a model-based approach specifically designed to detect andaccount for careless respondents in VAS data. To this end, we integrate existingmeasurement models for VASs (Noel and Dauvier, 2007) with mixture item response theorymodels for identifying and modeling careless responding. Simulation results show that the proposed model effectively detects careless responding and recovers key parameters, and highlights the unsuitability of the existing mixture factor analysis model for VAS data. Weillustrate the model’s potential for identifying and accounting for careless responding usingreal data from both VASs and Likert scales. First, we show how the model can be used tocompare careless responding across different scale types, revealing a slightly higherproportion of careless respondents in VAS compared to Likert scale data. Second, wedemonstrate that item parameters from the proposed model, accounting for carelessresponding, exhibit improved psychometric properties compared to those from a modelthat ignores it. These findings underscore the model’s potential to enhance data quality byidentifying and addressing careless responding.