Modelling Forced‐Choice Response Formats

Validity Evidence of the Abstract and Spatial Reasoning Test

“…Although most of the items in the first version presented appropriate discrimination indices, the exclusion of some items improved the balance of the difficulty of the AR and SR subtests. This balance is in line with the discussions proposed by the psychometrics literature regarding the distribution of the difficulty of the items: mostly median items and few easy or difficult items (Irwing and Hughes 2018). The second version of the TRAE presented evidence of internal structure.…”

Wetzel, Frick, & Brown (2020)_Does multidimensional forced-choice prevent faking?

“…In the Thurstonian item response model, ranks are coded into binary outcome variables from all pairwise comparisons between the items in the block. The model parameters can be estimated from the binary outcome variables using limited information methods (for details see Brown & Maydeu-Olivares, 2011;2018a). The FC format has been shown to have similar or even better criterion-related validity than the rating scale (RS) format (Bartram, 2007;Lee, Lee, & Stark, 2018;Salgado & Táuriz, 2014;Wetzel, Roberts, Fraley, & Brown, 1 For more information on how to estimate the Thurstonian item response model, see Brown and Maydeu-Olivares (2012).…”

Effects of Response Format on Psychometric Properties and Fairness of a Matrices Test: Multiple Choice vs. Free Response

“…The question regarding the most appropriate response format to gain valid information concerning test takers' characteristics is currently becoming a "hot topic" within the assessment of personality in the job context with reference to the newly emerged forced choice format (see Usami et al, 2016;Lin and Brown, 2017;O'Neill et al, 2017;Brown and Maydeu-Olivares, 2018;Wetzel and Greiff, 2018). However, studies addressing response formats within the domain of the assessment of knowledge, skills, and abilities with stem-equivalent items are rare.…”