| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
| Psychiatric Bulletin | Advances in Psychiatric Treatment | All RCPsych Journals |
|
|
|
||||||||||
|
||||||||||||
Electronic Letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Psychopathy and Antisocial Tendencies : A reply to Cooke, Michie, and Skeem
Erroneous conclusions about the PCL-R based on faulty modeling
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Michael J. Vitacco, Associate Director of Research Mendota Mental Health Institute
Send letter to journal:
vitacmj{at}dhfs.state.wi.us Michael J. Vitacco
|
British Journal of Psychiatry (BJP) published an article by Cooke and colleagues titled “Understanding the structure of the Psychopathy Checklist-Revised…,” 2007, 190, s39-s50. The Cooke article involves serious omissions of relevant literature, which directly challenge their thesis that antisocial tendencies are not part of the psychopathy construct. To quote the article, “In our view there is no compelling empirical evidence to support the conclusion that antisocial behavior is a central feature of psychopathy” (p. s48). However, in the very same issue of BJP as the Cooke article is a study by Viding and colleagues that directly contradicts the Cooke thesis—i.e., Viding et al. find a common genetic component to callous-unemotional traits and antisocial tendencies. Other studies cited in the Viding et al. article find similar results. More recently, Larsson et al., (2007) reported that the same general four factors as present in our four-factor model of psychopathy (Vitacco et al., 2005) all loaded onto a single genetic factor. Longitudinal research (also not cited) contradicts the assertions of the Cooke article as well—e.g., antisocial tendencies are significantly linked to the longitudinal stability of psychopathic traits (Frick et al., 2003). The Cooke group refers to the work of Cleckley to support their position. However, from Cleckley’s accounts of psychopathy, antisocial behaviours play an important role. As Patrick (2006, p. 608) noted, “There is no question that Cleckley considered persistent antisocial deviance to be characteristic of psychopaths. Without exception, all the individuals represented in his case histories engage in repeated violations of the law–including truancy, vandalism, theft, fraud, forgery, fire-setting, drunkenness and disorderly conduct, assault, reckless driving, drug offenses, prostitution, and escape.” As Blackburn (2007, p. 145) recently put it, “Contra Cooke,…antisocial behavior, conceived broadly, is a characteristic feature of psychopathy.” Although not cited, Cooke and colleagues are well aware of our 2005 paper (Vitacco, et al., Journal of Consulting and Clinical Psychology), given that they cite it in their companion article in Psychological Assessment, and which provides several lines of evidence that contradict their thesis. In our 2005 paper, which is based on a very large sample, we demonstrate the conceptual errors and modeling flaws that went into the development of Cooke’s model, and provide evidence for the four-factor model. Interestingly, Cooke and colleagues chose instead to cite our small preliminary studies to criticize our research, despite the fact that they are in-line with the larger study not cited in their article. In the Skeem and Cooke companion piece to their BJP article, these authors admit that antisocial tendencies are meaningfully linked to other psychopathic features. Specifically, Skeem and Cooke wrote: “Some antisocial behavior seems inherent to the interpersonal and affective core of psychopathy.” Their claim that we view criminality as central to psychopathy is wrong. Indeed, the Psychopathy Checklist: Screening Version (PCL: SV) contains two items that refer to antisocial behavior and that can be scored without evidence of criminality; the PCL-R and PCL: SV are virtually identical psychometrically, as noted in Cooke’s own previous research. References Blackburn, R. (2007). Personality disorder and antisocial deviance: Comments on the debate on the structure of the Psychopathy Checklist- Revised. Journal of Personality Disorders, 21, 142-159. Frick, P. J., Kimonis, E. R., Dandreaux, D. M., & Farell, J. M. (2003). The 4 years stability of psychopathic traits in non-referred youth. Behavioral Sciences and the Law, 21, 1-24. Larsson, H., Tuvblad, C., Rijsdijk, F. V., Andershed, H., Grann, M., & Lichtenstein, P. (2007). A common genetic factor explains the association between psychopathic personality and antisocial behavior. Psychological Medicine, 37, 15-26. Patrick, C. J. (2006). Back to the future: Cleckley as a guide to the next generation of psychopathy research (p. 605-618). In C. J. Patrick (Ed.), Handbook of Psychopathy. New York: Guilford. Vitacco, M., Neumann, C. S., & Jackson, R. L. (2005). Testing of a four-factor model of psychopathy: Associations with gender, ethnicity, intelligence and violence. Journal of Consulting and Clinical Psychology, 73, 466-476. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Craig S. Neumann, Psychologist University of North Texas, Robert D. Hare
Send letter to journal:
csn0001{at}unt.edu Craig S. Neumann, et al.
|
British Journal of Psychiatry published an article by Cooke and colleagues titled “Understanding the structure of the Psychopathy Checklist- Revised…,” 2007, 190, s39-s50 (hereafter referred to as the ‘Cooke article’). The Cooke article contains a number of fundamental modeling errors. First, the authors continue to present an over-factored model (i.e., hierarchical ‘three’ factor model with testlets), which results in negative variances. This 13-item model actually contains 10 factors: 6 first-order factors/testlets, 3 second- order factors, and 1 third-order factor (simply count the number of circles/factors in their Figure 1). Any model can achieve good fit when it is as complex as the data it attempts to summarize. We have shown that this ‘testlet’ model results in untenable parameters in four separate studies (cf. Neumann et al., 2006). One of Cooke’s authors have also suggested that the testlet model is over-factored (Skeem et al., 2003). Cooke does not acknowledge this problem of an over-factored model, even though it is evident in his published work (see Cooke & Michie, 2001, Figures 2 and 3, which contain zero variance terms that EQS sets to zero-value when estimating negative variances). The Cooke article mentions that we have criticized their use of testlets, but does not dispute our point that it creates a misspecified model with untenable parameters. Our analysis of the testlet model, illustrating impossible parameters, is available upon request, though readers can see this for themselves by analyzing the data matrix provided by Cooke. The Cooke article provided a polychoric correlation matrix, ostensibly to give investigators the opportunity to replicate their findings. However, as noted in the EQS program manual cited in the Cooke article, robust procedures can only be conducted with the raw items. Thus, the Cooke article appears to be ‘transparent’ when in reality no one will be able to unambiguously verify their analyses! When one analyzes their published correlation matrix, using a non- robust procedure, very different findings from those published in the Cooke article result. Also, Cooke relied upon a maximum likelihood (ML) procedure for estimating model parameters, despite the fact that it is well known that this procedure under-estimates model parameters and model fit when used with ordinal data (Everitt & Dunn, 2001), such as the PCL-R items. There was no serious discussion on why robust ML with polychoric correlations was employed, except that it is recommended in the EQS version 6 manual. Nonetheless, the verisimilitude of this new approach is currently unknown. A program such as Mplus, which employs a robust weighted least squares procedure for ordinal data is an accepted approach (Neumann et al., 2006). The Cooke article involved limited use of Mplus. Our Mplus analyses of the UK data are presented in Table 1 (along with our previously published findings). Contrary to the Cooke article, the four-factor model clearly fits as well or better than a viable three-factor model. Moreover, our recent research, which Cooke pointedly ignores, indicates that the four first-order factors are explained by a cohesive superordinate factor (Neumann et al., 2006, 2007). Citations- Cooke, D.J., & Michie, C. (2001). Refining the construct of psychopathy: Towards a hierarchical model. Psychological Assessment, 13, 171-188. Everitt, B. & Dunn, G. (2001). Applied multivariate data analysis. New York: Oxford University Press. Neumann, C. S., Kosson, D. S., Forth, A. E., & Hare, R. D. (2006). Factor structure of the Hare Psychopathy Checklist: Youth Version (PCL:YV) in incarcerated adolescents. Psychological Assessment, 18, 142-154. Neumann, C. S., Hare, R. D., & Newman, J. P. (2007). The super- ordinate nature of the Psychopathy Checklist-Revised. Special Section on Psychopathy: Journal of Personality Disorders, 21, 102-117. Skeem, J. L., Mulvey, E. P., & Grisso, T. (2003). Applicability of traditional and revised models of psychopathy to the Psychopathy Checklist: Screening Version. Psychological Assessment, 15, 41-55.
Table 1 Model Fit Results for United Kingdom and North American Samples: Hare et al. Four Factor Model and Cooke et al. (degraded) Three Factor Model*
Note. All UK (Her Majestys Prison Service) data are from the PCL-R Manual (Hare, 2003) and specific sample characteristics reported therein. All analyses conducted with Mplus (Muthen & Muthen, 2001) using a robust weighted least squares (WLSMV) procedure, as done in the Cooke article. TLI (aka non-normed fit index) = Tucker Lewis Index, RMSEA = Root Mean Square Error of Approximation, SRMR = Standardized Root Mean Square. * The 13-item degraded three-factor model is the only statistically tenable latent variable model since the 13-item three-factor testlet model actually contains ten factors (i.e., 6 first-order factors [testlets], 3 second-order factors, and 1 third-order factor). See Neumann, Vitacco, Hare, & Wupperman (2005) for criticism of the Cooke testlet model. Indeed, when we tested the testlet model using the HMP data it resulted in a mis-specified model, having two negative variances (i.e., untenable parameters). 1 The mega sample is composed of all UK listwise data, plus Swedish offender and psychiatric described in the PCL-R manual. 2 Neumann, C. S., Hare, R. D., & Newman, J. P. (2007). The superordinate nature of the Psychopathy Checklist-Revised. Journal of Personality Disorders, 21, 102-117. Pooled sample: 4865 male offenders, 1099 female offenders, 965 male forensic patients. 3 Neumann, C. S. & Vitacco, M. J., Hare, R. D., & Wupperman, P. (2005) Reconstruing the "Reconstruction" of psychopathy: A comment on Cooke, Michie, Hart & Clarke. Journal of Personality Disorders, 19 ,624-640. 4 Neumann, C. S., Kosson, D. S., Forth, A. E., & Hare, R. D. (2006). Factor structure of the Hare Psychopathy Checklist: Youth Version (PCL:YV) in incarcerated adolescents. Psychological Assessment, 18, 142-154. 5 Vitacco, M., Neumann, C. S., & Jackson, R. L. (2005). Testing of a four-factor model of psychopathy: Associations with gender, ethnicity, intelligence and violence. Journal of Consulting and Clinical Psychology, 73, 466-476.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
| Psychiatric Bulletin | Advances in Psychiatric Treatment | All RCPsych Journals |
eLetters: