Yon V, Loas G, Monestes J-L, Verrier A, Deligne H
Service Universitaire de Psychiatrie de l'Adulte et de Psychologie Médicale, Hôpital Pinel, 80044 Amiens cedex 01, France.
Encephale. 2007 May-Jun;33(3 Pt 1):249-55. doi: 10.1016/s0013-7006(07)92036-9.
The aim of this article is firstly to present the French translations of the Cognitive Slippage (Miers and Raulin, 1987) and the Schizotypal Ambivalence Scales (Raulin, 1986), and secondly to determine their psychometric properties in different samples of non-clinical and clinical subjects.
Chapman et al. have developed trait-oriented scales based on Meehl's manual of schizotypy, such as the Physical and Social Anhedonia Scales, the Perceptual Aberration Scale, the Magical Ideation Scale, the Impulsive Nonconformity Scale, the Cognitive Slippage Scale (CSS), and the Schizotypal Ambivalence Scale (SAS). The Chapman scales of psychosis proneness are the most internationally used instruments for the assessment of schizotypy and some of them such as the Cognitive Slippage and Schizotypal Ambivalence Scales were still not available in French.
The study was conducted in three different samples: the non-clinical sample (n=158) comprised firstly 128 university students (113 females, 15 males) with a mean age of 24.67 years (sd=5.18) and secondly 30 healthy subjects (12 males, 18 females) with a mean age of 33.26 years (sd=7.35); the clinical sample (n=167) comprised firstly 106 psychiatric patients hospitalized in a general hospital (73 males, 33 females) with a mean age of 38.35 years (sd=11.60) and 61 (35 males, 26 females) in or outpatients of a psychiatric department with a mean age of 37.75 years (sd=10.72); 15 schizotypal university students presenting high score of the Revised Social Anhedonia Scale (1 male, 14 females) with a mean age of 21.26 years (sd=1.1). Using the ICD-10, the diagnoses for the general hospital sample were neurotic or anxious disorders (47%) and depressive disorders (24%). For the psychiatric department sample, the diagnoses were schizophrenic disorders (29%), mood disorders (16%), neurotic disorders (18%) and personality disorders (15%). The internal consistency was determined by the Kuder-Richardson coefficient (KR 20) (CSS: KR 20=0.85 in the university sample; 0.89 in the general hospital sample; 0.87 in the psychiatric department sample; SAS: KR 20=0.79 in the university sample; 0.82 in the general hospital sample; 0.85 in the psychiatric department sample).
The mean of the correlations between each item and the total score was calculated (CSS: 0.41 in the university sample; 0.44 in the general hospital sample; SAS: 0.46 in the university sample; 0.49 in the general hospital sample). The discriminant validity was studied by a Student's t test showing that the schizotypal university students had significant higher scores on the CSS and SAS than the healthy university students. The reliability was explored using a test-retest in the 30 healthy subjects. The subjects filled out the scales three months after the first testing. The intraclass coefficients were 0.81. Finally, the cutoff scores were calculated following the procedure described by Chapman et al. (score higher than two standard deviations above the mean of the university sample). The values for the CSS and SAS were respectively 15 and 12 for males and females.
本文的目的一是呈现认知失误量表（米尔斯和劳林，1987年）以及分裂型矛盾情感量表（劳林，1986年）的法语翻译版本，二是在不同的非临床和临床样本中确定它们的心理测量特性。
查普曼等人基于米希尔的分裂型人格手册开发了特质导向量表，如躯体和社交快感缺乏量表、感知畸变量表、神奇观念量表、冲动性不遵奉量表、认知失误量表（CSS）以及分裂型矛盾情感量表（SAS）。查普曼的精神病易感性量表是国际上用于评估分裂型人格最常用的工具，其中一些量表，如认知失误量表和分裂型矛盾情感量表，尚无法语版本。
该研究在三个不同样本中进行：非临床样本（n = 158）首先包括128名大学生（113名女性，15名男性），平均年龄24.67岁（标准差 = 5.18），其次包括30名健康受试者（12名男性，18名女性），平均年龄33.26岁（标准差 = 7.35）；临床样本（n = 167）首先包括106名在综合医院住院的精神科患者（73名男性，33名女性），平均年龄38.35岁（标准差 = 11.60），以及61名精神科门诊或住院患者（35名男性，26名女性），平均年龄37.75岁（标准差 = 10.72）；15名在修订社交快感缺乏量表上得分较高的分裂型大学生（1名男性，14名女性），平均年龄21.26岁（标准差 = 1.1）。使用国际疾病分类第10版（ICD - 10），综合医院样本的诊断为神经症或焦虑症（4�）和抑郁症（24�）。对于精神科样本，诊断为精神分裂症（29�）、心境障碍（16�）、神经症（18�）和人格障碍（15�）。内部一致性通过库德 - 理查森系数（KR 20）确定（CSS：大学样本中KR 20 = 0.85；综合医院样本中为0.89；精神科样本中为0.87；SAS：大学样本中KR 20 = 0.79；综合医院样本中为0.82；精神科样本中为0.85）。
计算了每个项目与总分之间的平均相关性（CSS：大学样本中为0.41；综合医院样本中为0.44；SAS：大学样本中为0.46；综合医院样本中为0.49）。通过学生t检验研究了区分效度，结果表明分裂型大学生在CSS和SAS上的得分显著高于健康大学生。在30名健康受试者中通过重测法探索了信度。受试者在首次测试三个月后填写量表。组内相关系数为0.81。最后，按照查普曼等人描述的程序计算临界分数（得分高于大学样本均值两个标准差以上）。CSS和SAS的男性和女性临界分数分别为15和12。
