Assessing Differential Item Functioning Across Clinical and Community Samples in the Miller Forensic Assessment of Symptoms Test
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Assessment of malingering is a central component of forensic evaluations in criminal as well as civil litigation contexts. As such, there is a need for empirically sound measures of malingering to accurately identify individuals who may be feigning symptoms for personal gain. The detection of differential item functioning (DIF) using item response theory methods provides a powerful method of evaluating whether items on a malingering assessment measure function differently in civil versus criminal litigation contexts. The aim of the current study was to evaluate DIF based on litigation type in the Miller Forensic Assessment of Symptoms Test (Miller, 2011). The M-FAST was administered to a sample of criminal defendants at Arkansas State Hospital as well as an analogue sample of civil litigants comprised of undergraduates at a large public Southwestern university. Results indicated DIF for nine M-FAST items, with five of the items more easily endorsed by criminal litigants and four of the items more easily endorsed by civil litigants. Implications of these results for research and clinical assessment are discussed.