بررسی مقیاس اختلالات فکری و روانی خودگزارشی لونسون با استفاده از یک مدل واکنش مدرج

The current study examined gender differences on the Levenson Self-Report Psychopathy Scale (LSRP; Levenson, Kiehl, & Fitzpatrick, 1995), a self-report measure of psychopathic personality traits, in an undergraduate sample using a Graded Response Model. Results demonstrated that some items on the LSRP are not contributing to the total score of the LSRP and could be removed. Additionally, differential item functioning demonstrated that items are endorsed differently by gender. Specifically, males endorsed items suggesting proneness to boredom and impulsivity, as well as “active” characteristics of psychopathy. This suggests that males with psychopathic characteristics actively seek out individuals to harm. However, females endorsed more “passive” items, suggesting a propensity to manipulate individuals with whom they come into contact, rather than seeking out individuals to harm.

Psychopathic individuals possess a dearth of basic prosocial personality traits such as empathy, guilt, and perspective-taking and exhibit glibness, superficial charm, grandiosity, deception, and the tendency to manipulate others (e.g., Hare, 2003, Lykken, 1995, Marcus et al., 2004 and Poythress and Skeem, 2006). They reportedly have little depth of emotion (Hare, 2003), deficient emotional processing (Lorenz & Newman, 2002), and deficits in fear-potentiated startle response (Patrick, 1994). Guiltlessness, dishonesty, antisocial behavior, and failure to form close emotional bonds are often included in definitions of psychopathy (Hare, 2003). Psychopathic traits have primarily been studied in male samples; therefore, it is often assumed that the male expression of psychopathy can be superimposed on females. Because of this, potential gender differences have largely been ignored (Forouzan & Cooke, 2005). The current study attempts to address this deficit in the literature by examining gender differences in a popular self-report measure of psychopathic traits, the Levenson Self-Report Psychopathy Scale (LSRP; Levenson et al., 1995). Understanding gender differences in psychopathic characteristics could have several implications. If males and females demonstrate psychopathic characteristics differently, this could suggest that there is a need for different psychopathy measures and diagnostic criteria by gender. 1.1. Gender differences in psychopathic traits Several studies have reported clear gender differences in the prevalence of psychopathic traits, such that psychopathy is more common among males than females across populations (e.g., undergraduates, substance abusers, civil psychiatric patients, forensic psychiatric patients, and offenders/inmates; Nicholls, Ogloff, Brink, Psych, & Spidel, 2005) and measures (Cale and Lilienfeld, 2002, Forth et al., 1996, Levenson et al., 1995, Wilson et al., 1999 and Zagon and Jackson, 1994), including the LSRP, Psychopathy Checklist-Revised (PCL-R; Hare, 2003), Psychopathy Checklist: Short Version (PCL:SV; Hart, Cox, & Hare, 1995), and Psychopathic Personality Inventory (PPI; Lilienfeld & Andrews, 1996). The base rate of psychopathy ranges from 9% to 23% in females and 15–30% for males in offender samples (Vitale, Smith, Brinkley, & Newman, 2002). Based on these differences, some authors have suggested that it would be useful to refine the cut scores on psychopathy measures to apply specifically to females (Cale and Lilienfeld, 2002, Forth et al., 1996, Levenson et al., 1995, Vitale et al., 2002, Wilson et al., 1999 and Zagon and Jackson, 1994). Gender differences in prevalence rates could be the result of several different factors including sampling bias, gender bias in diagnostic criteria, and differential presentation (Dolan & Vollm, 2009). It is unknown whether these differences are qualitative or quantitative; females may possess the same underlying characteristics of psychopathy and express them differently or psychopathic characteristics may differ by gender. The interpersonal symptoms are influenced by culture and gender, such that they are commonly seen in males but are only present in females when they possess a severe number of symptoms (Cooke et al., 2005 and Forouzan and Cooke, 2005). Psychopathic characteristics demonstrated by females are related to extraversion, agreeableness, guilt, lying, deceitfulness, and lack of control, whereas psychopathic characteristics demonstrated by males are related to externalizing behaviors, risky driving, antisocial and aggressive behaviors, and criminality (Blonigen et al., 2008, Lee and Salekin, 2010 and Strand and Belfrage, 2005). 1.2. Levenson Self Report Psychopathy Scale The PCL-R is generally considered to be the gold standard for assessing psychopathy (Skeem & Cooke, 2010); however, its development on a criminal population and reliance on corroborating file data make it inappropriate for use in non-incarcerated samples (Hare, 1991). Other measures have been developed that assess psychopathic personality traits without relying on criminal behavior. One such measure, the LSRP (Levenson et al., 1995), is based on the PCL-R and validated on a college population. It measures information about antisocial behaviors more typical of community life which makes it more appropriate for use in college populations. A recent study using an undergraduate sample found intercept bias, indicating that similar scores by gender on the LSRP could have different implications for males and females (Marion & Sellbom, 2011). Initially, a two-factor model of the PCL-R was validated, in which Factor 1 was related to affective/interpersonal deficits, and Factor 2 was related to an antisocial, impulsive lifestyle (Hare, 1991 and Hare, 2003). The two factors of the LSRP were designed to measure the PCL-R factors (Levenson et al., 1995). It is comprised of primary psychopathy (LSRP1: 10 items) and secondary psychopathy (LSRP2: 16 items). Research findings have provided some support for the two-factor structure of the LSRP (Brinkley et al., 2001, Levenson et al., 1995 and Lynam et al., 1999). However, recent findings have shown that a two-factor model did not fit the data well (Brinkley, Diamond, Magaletta, & Heigel, 2008) and have provided support for a three-factor model of psychopathy, as measured by the PCL-R and LSRP (Brinkley et al., 2008, Patrick et al., 2009 and Sellbom, 2010). The three factors that were identified were Egocentric (manipulative interpersonal style), Antisocial (impulsive and antisocial lifestyle), and Callous (callous predatory approach to life). These findings are similar to Cooke and Michie’s (2001) proposed three-factor structure for the PCL-R that was conducted using a sample of female inmates. 1.3. Current Study The current study sought to examine gender difference on the LSRP using a Graded Response Model. Psychopathy appears to be a dimensional construct (e.g., Marcus et al., 2004), and it has been argued that studying individuals lower on the continuum of psychopathy is desirable as the effects of confounding variables typical in samples higher on the continuum can be ruled out (e.g., drug use, the acute effects of incarceration; Lilienfeld, 1994). Thus, the following hypotheses were investigated in a college sample: 1. Previous research has demonstrated that females demonstrate psychopathic characteristics related to extraversion, agreeableness, guilt, lying, deceitfulness, and lack of control. Conversely, males demonstrate psychopathic characteristics related to externalizing behaviors, risky driving, antisocial and aggressive behaviors, and criminality. As such, we hypothesize that males will endorse items related to externalizing behaviors and aggression. 2. Additionally, we hypothesize that females will endorse items related to manipulation, deceitfulness, and lying.