Deriving Psychopathy Subtypes Using Model-Based Cluster Analysis

Identifying Psychopathy Subtypes Using a Broader Model of Personality: An Investigation of the Five Factor Model Using Model-Based Cluster Analysis

“…The current study used MBCA of the five NEO-PI-R domain scores (i.e., Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness). This type of analysis breaks down the sample into distinct subtypes (Falkenbach, Beltrani & Reinhard, 2018; Rapkin & Luke, 1993) based on patterns of scores across multiple measures. This analysis uses a goodness-of-fit index (see Fraley, 1998, for further details concerning the assumptions underlying the goodness-of-fit index) and the Bayesian Information Criterion (BIC) to objectively determine the number and parameters of each cluster (see Falkenbach, Beltrani, & Reinhard, 2018, for a review).…”

“…This type of analysis breaks down the sample into distinct subtypes (Falkenbach, Beltrani & Reinhard, 2018; Rapkin & Luke, 1993) based on patterns of scores across multiple measures. This analysis uses a goodness-of-fit index (see Fraley, 1998, for further details concerning the assumptions underlying the goodness-of-fit index) and the Bayesian Information Criterion (BIC) to objectively determine the number and parameters of each cluster (see Falkenbach, Beltrani, & Reinhard, 2018, for a review). MBCA tests different models that vary by assumptions (amount of variability [size], strength of relationship [shape], and direction of relationship [orientation]).…”

“…MBCA tests different models that vary by assumptions (amount of variability [size], strength of relationship [shape], and direction of relationship [orientation]). The relationship between each of the components dictates the subtypes yielded and the assignment of individuals to respective clusters (see Falkenbach, Beltrani, & Reinhard, 2018; Fraley, 1998). The Mclust program used in the current study tests 14 models for up to nine subpopulations for a potential total of 126 different models, an generates a corresponding BIC score.…”

“…Correlational studies in nonforensic samples support psychopathy subtyping theory, revealing that Factor 1 is associated with overt/healthy narcissism (Falkenbach, Howe, & Falki, 2013), positive affect (Delgaizo & Falkenbach, 2008), and low levels of anxiety (Benning, Patrick, Salekin, & Leistico, 2005; Marcus, Fulton, & Edens, 2013), whereas Factor 2 correlates with BPD traits (Miller et al, 2010), anxiety, and covert/pathological narcissism. However, correlations alone are insufficient for analyzing the multiple traits of subtypes (Falkenbach, Beltrani, & Reinhard, 2018; Lykken, 1995).…”

“…Model-based cluster analysis (MBCA) objectively reveals predictable and distinct groupings based on multiple variables (Falkenbach, Beltrani, & Reinhard, 2018; Skeem et al, 2003). Model-based cluster analysis research with forensic (Poythress et al, 2010; Skeem, Johansson, Andershed, Kerr, & Louden, 2007), undergraduate (Falkenbach, Barese, Balash, Reinhard, & Hughs, 2015; Falkenbach, Poythress, & Creevy, 2008; Falkenbach, Reinhard, & Larson, 2017; Falkenbach, Stern, & Creevy, 2014; Lee & Salekin, 2010), and even police (Falkenbach, Glackin, & McKinley, 2018) samples has confirmed the existence of theoretically coherent psychopathy subtypes.…”