Darke S, Kaye S, Finlay-Jones R, Hall W
National Drug and Alcohol Research Center, Department of Psychiatry, University of New South Wales, Australia.
J Pers Disord. 1998 Summer;12(2):162-71. doi: 10.1521/pedi.1998.12.2.162.
The proposed two-factor structure of the Psychopathy Checklist (Revised) was assessed by confirmatory factor analyses on a sample of 376 community and prison based methadones maintenance patients. Confirmatory factor analyses indicated that the two-factor model proposed by previous researchers did not fit well with the data from this population. Exploratory analyses extracted five interpretable factors, three psychological factors (glibness/manipulative, callousness, irresponsibility), and two behavioral factors (criminal behaviors, promiscuity), which accounted for 61% of the variance. The data provided support for the psychometric reliability of the PCL-R, as indicated by a Chronbach's alpha of 0.83, weighted kappas for concordance between raters ranging from 0.51 to 1.00, and 100% agreement on diagnoses of psychopathy between raters. The data indicate that a two-factor structure does not adequately capture the dimensions of psychopathy in this population. While the factors do fall into distinct psychological and behavioral dimensions, more than two general factors are required to describe the dimensions of psychopathy.
在376名社区及监狱美沙酮维持治疗患者样本中,通过验证性因素分析对修订版精神病态量表(Psychopathy Checklist,PCL-R)所提议的双因素结构进行了评估。验证性因素分析表明,先前研究者提出的双因素模型与该人群的数据拟合度不佳。探索性分析提取出五个可解释的因素,三个心理因素(油嘴滑舌/善于操纵、冷酷无情、不负责任)和两个行为因素(犯罪行为、滥交),它们解释了61%的方差。PCL-R的心理测量学可靠性得到了数据支持,克朗巴哈系数(Chronbach's alpha)为0.83,评分者间一致性的加权卡帕值在0.51至1.00之间,评分者对精神病态诊断的一致性为100%。数据表明,双因素结构无法充分捕捉该人群中精神病态的维度。虽然这些因素确实分为不同的心理和行为维度,但描述精神病态维度需要两个以上的一般因素。