Objective: This theoretical paper refines the use of chaining likelihood ratios (LR) for performance validity tests (PVTs; Larrabee, 2008), a Bayesian approach, to more accurately classify credible effort (CE) versus non-credible effort (NCE). Larrabee (2008) determined a 93.6% probability of NCE with ≥2 failed PVTs. Using this method, Boone (2013) and Meyers et al. (2014) argued for ≥2 PVTs for determining NCE. We demonstrate by only chaining LRs for failed PVTs, this method significantly overestimates the true probability of NCE. Method: Data from Meyers et al. (2014), the most recently published study of chained LRs, was re-analyzed using all 11 PVTs administered (rounded average sensitivity = 0.80, specificity = 0.90). They determined ≥2 failed PVTs equals an NCE probability of 97.8%. Results: Re-calculated posterior probabilities of NCE for all PVTs administered (passed and failed), assuming a 40% base rate of NCE: 2 failed, 9 passed = 0.0056%, 3 failed, 8...

Article PDF first page preview

Article PDF first page preview
Article PDF first page preview
You do not currently have access to this article.