Psychopathy NOT Predictive of Violence in Latino Americans

Psychopathy NOT Predictive of Violence in Latino Americans

Law and Human BehaviorThe power of psychopathy to predict violence may vary by ethnic group. This is the bottom line of a recently published article in Law and Human Behavior. Below is a summary of the research and findings as well as a translation of this research into practice.

Featured Article | Law and Human Behavior | 2013, Vol. 37, No. 5, 303-311

Psychopathy and Criminal Violence: The Moderating Effect of Ethnicity


Zach Walsh, University of British Columbia


This study aimed to determine the cross-ethnic stability of the predictive relationship of psychopathy for violence. Participants were 424 adult male jail inmates. Psychopathy was assessed using the Psychopathy Checklist—Revised and criminal violence was assessed using a comprehensive database of arrests for violent crimes. Ethnic categories included the groups that make up the vast majority of U.S. inmates: European American (EA, n = 166), African American (AA, n = 174), and Latino American (LA, n = 84). Ethnically aggregated Cox regression survival analyses identified predictive effects for psychopathy. Disaggregated analyses identified ethnic differences: Psychopathy was more strongly predictive of violence among EA (R2 = .13, 95% CI [.04, .22], p < .01) relative to AA inmates (R2 = .05, 95% CI [.00, .11], p < .01) and was not related to violence among LA participants (R2 = .02, 95% CI [.00, .08], p = .22). Receiver operating characteristic curve analyses yielded an equivalent pattern of results. These findings add to a growing literature suggesting cross-ethnic variability in the predictive power of psychopathy for violence.


psychopathy, violence, ethnicity, personality disorder

Summary of the Research

The relationship between psychopathy and violent behavior has been widely demonstrated: Higher levels of psychopathy (typically as measured by the PCL-R) are predictive of violence. Does this relationship, however, exist when we start to examine it within different ethnic groups? The featured research article uses a prospective design (as opposed to the more typical –-and easier-to-implement—retrospective design) to investigate the predictive power of psychopathy for criminal violence among jail inmates self-identifying as belonging to one of three different ethnic groups—those most common within the US population: African American, European American, and Latino American.

Participants were 424 inmates in a northeastern Illinois county jail. Inmates’ self-reported ethnic classification on the jail roster was used to divide the participants into three groups:  174 African Americans, 166 European Americans, and 84 Latino Americans. Each participant was administered the PCL-R (Psychopathy Checklist – Revised; Hare, 2003) while in jail and then, upon release, was monitored for violent recidivism by pulling data from the Law Enforcement Agencies Data System, the most comprehensive index of criminal arrests in the USA.

The group, as a whole, survived an average of 65.85 months (SE = 2.66) and 57% were arrested for a violent offense. Being African American or Latino American was associated with higher risk for violent recidivism than was being European American; no differences were found between African American and Latino American ethnic group status and violent recidivism.

When the three ethnic groups were aggregated, psychopathy predicted violence. However, when the groups were split apart by ethnicity, psychopathy was predictive of violent recidivism in the European American and African American groups but not in the Latino American group. When the entire group was split into two—European American and non-European American (aggregating the African American and Latino American groups into one), psychopathy was a stronger predictor of violent recidivism within the European American group.

“This study assessed ethnic differences in the predictive power of psychopathy for violence among male offenders in the United States. At the aggregate level, these findings are generally consistent with prior research: Psychopathy predicted criminal violence with a medium-size effect (Cohen, 1988). However, ethnic aggregation masked variability in the predictive power of psychopathy for violence. Specifically, psychopathy was more strongly predictive among EA relative to AA men and was not predictive among LA men. The present findings add to a growing literature that suggests limitations on the cross-ethnic generalizability of the relationship between psychopathy and violence (Edens et al., 2007; Leistico et al., 2008; Singh et al., 2011)” (p. 306).

Translating Research into Practice

Given the results of this research, as well as other work in this general area (cited in the featured article), the author concludes, “the current state of empirical knowledge does not appear to warrant equivalent levels of confidence in the predictive utility of psychopathy for violence among adult male AA offenders relative to EA offenders” (p. 308). So, the bottom line is that we need to be cautious about using psychopathy as a predictor of violence behavior in non-European American individuals and samples.

More research in this area is warranted and researchers are called upon to consider ethnic group differences as well as within-group variation in attempting to further elucidate the relations amongst psychopathy and violence.

Clinicians need to use caution when attempting to determine the relative weight to give to variables such as psychopathy in risk for future violence amongst individuals belonging to ethnic minorities. Although the relationship between psychopathy and violence appears well established in the United States, the predictive power of psychopathy may be lower or non-existent with various ethnic minority groups.

Other Interesting Tidbits for Researchers and Clinicians

This article also provides some interesting information regarding the factors and facets of psychopathy. The data are analyzed in a couple different ways in an attempt to elucidate the debate regarding whether PCL-R scores best conform to a two-factor/four-facet model that includes antisociality or a three-factor model that omits antisociality.

There are also some interesting analyses and discussion regarding the role of ethnic differences and individual differences in sociodemographic risk factors, such as socioeconomic status, unemployment, and residing in a criminogenic neighborhood. Along with these are some analyses by high- and low-psychopathy groups.

Join the Discussion

As always, please join the discussion below if you have thoughts or comments to add!

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.