The Hare Psychopathy Checklist–Revised (PCL-R)

The manual standardized a 20-item clinician-rated instrument derived from Hare's earlier 22-item Psychopathy Checklist. Two items—previous diagnosis as a psychopath and drug or alcohol not being a direct cause of antisocial behavior—were dropped because they correlated poorly with the total score and proved difficult to score reliably. Scoring thresholds for some items were made more stringent, requiring more extreme behavioral evidence for a score of 2. The resulting instrument measures psychopathy across two empirically derived factors: an interpersonal-affective dimension (Factor 1) and an impulsive-antisocial lifestyle dimension (Factor 2). This two-factor structure, supported by factor-analytic studies Hare and colleagues had published in the late 1980s and early 1990s, became the manual's central organizing framework and the basis for its scoring architecture.

Cleckley's The Mask of Sanity described psychopathy through rich case narratives emphasizing superficial charm, absence of remorse, and a mask of sanity concealing profound affective deficits. The PCL-R manual translated those clinical descriptions into scored items but also incorporated explicitly criminal and antisocial behaviors that Cleckley had treated as incidental rather than definitional. This has generated a sustained scholarly dispute. Hare and colleagues later argued that some researchers' literal adherence to Cleckley's profile, drawn from an unrepresentative patient sample, had become problematical, and that the PCL-R's empirical grounding across structural, genetic, developmental, and neurobiological research was more informative than fidelity to early clinical formulations [citation:15]. Critics, by contrast, contend the instrument's inclusion of criminal history items means it measures criminality as much as psychopathy.

The manual's administration procedure reflects a deliberate methodological choice: psychopathic traits such as deceitfulness, manipulation, and lack of remorse are precisely the characteristics that make self-report unreliable. Requiring a semi-structured interview plus collateral records—prison files, psychiatric histories, police reports—allows the clinician to cross-check the individual's self-presentation against documented behavior. This design is a core feature of the instrument, not an incidental procedural detail. It also means PCL-R scores depend heavily on the quality and completeness of available file information, a practical limitation that becomes consequential in forensic settings where collateral records may be incomplete or contested. The interview-file convergence requirement distinguishes the PCL-R from questionnaire measures of psychopathy and partly explains its clinical authority.

Two lines of critique are best documented. First, the instrument's predictive validity is driven disproportionately by Factor 2, the impulsive-antisocial dimension, while the interpersonal-affective traits that define psychopathy conceptually contribute little to violence prediction [citation:3]. This creates an interpretive problem: the PCL-R may predict criminal behavior primarily because it measures prior criminality. Second, critics have documented substantial variability in cut-off scores used across studies, with some researchers employing thresholds as low as 9 out of 40—far below the conventional 30—which inflates sensitivity at the cost of producing high false-positive rates [citation:10]. The manual's own recommended cut-off of 30 has been subjected to limited validation, and the gap between research practice and clinical application remains a persistent controversy.

A documented pattern, acknowledged in the forensic literature, is that the Factor 1 personality items—remorselessness, superficial charm, lack of empathy—show lower inter-rater reliability than the more behavioral Factor 2 items such as juvenile delinquency or revocation of conditional release. A study by Blais, Forth, and Hare (2017) examining 280 participants in the developers' own training program found that inter-rater reliability for personality items was disappointingly low, leading the authors to conclude that raters' scores did not meet standards recommended for criminal cases [citation:16]. This is significant because the personality traits are often what jurors find most prejudicial, yet they are precisely the items clinicians struggle to score consistently. The finding complicates the instrument's authority in adversarial legal proceedings.

The Screening Version (PCL:SV), published by Hart, Cox, and Hare in 1995, was developed as a shorter, time-efficient alternative to the full PCL-R. Research comparing the two instruments in forensic and correctional samples has found inter-measure correlations for total scores of .95 in forensic settings and .94 in correctional settings, with area under the curve values of .98 when the PCL:SV was used to screen for a PCL-R total of 25 [citation:11]. The high correspondence indicates the screening version captures the same construct efficiently. However, the PCL:SV has received substantially less empirical attention than the full PCL-R, and some research suggests it may have minor practical advantages in specific populations—for instance, it omits the marital relationships item that proves difficult to score in offenders with intellectual disabilities [citation:17].