Background There is considerable debate about the underlying factor structure of the Psychopathy Checklist: Screening Version (PCL: SV). An established view is that it reflects a unitary construct underpinned by two correlated factors. More recent research has, however, undermined this conceptualisation. Aims Our aim was to compare 10 competing models of the PCL: SV in a sample of civil psychiatric patients. Method Ten distinct factor models were specified and tested using conventional confirmatory factor analytic techniques, along with confirmatory bifactor modelling. Results A bifactor model, including two general factors (interpersonal–affective and antisocial–lifestyle), and four subordinate factors (interpersonal, affective, antisocial and lifestyle) provided the best fit to the data. The reliability of the conceptualisation was supported through the use of composite reliability, and the differential relationships exhibited between the general factors and measures of personality, impulsivity and mental health. Conclusions Our findings provide further weight to taking the two general psychopathy factors into account when interpreting the PCL: SV for clinical purposes. Copyright © 2015 John Wiley & Sons, Ltd.