健康与遗传心理学研究室知识库

目的：对攻击性问卷（AQ）在成年男性罪犯群体中进行初步的试用。方法：根据量表修订方法,随机选取北京某3所监狱的190名罪犯进行测验,分析中文版信效度。结果：①攻击性问卷得分及项目分析结果表明,成年男性罪犯在攻击性问卷上的得分比较均衡,有5道题目项目鉴别力系数低于0.3,剩余题目项目鉴别力良好（r〉0.348）;②探索性因素分析得到问卷5个因素（特征根大于1.5,累积方差解释量为50.744）,这5个因素分别为身体攻击、言语攻击、间接攻击、愤怒和敌意。项目的载荷范围在0.402-0.825之间;③问卷的5个分量表内容效度均在0.551-0.833之间（P〈0.01）,具有良好的内容效度。选取WFS量表作为问卷的效标,相关系数均在0.574以上（P〈0.01）,由此可见整个问卷具有较好的效标效度;④整个问卷的Cronbach＇sα系数为0.797,5个分量表的Cronbach＇sα系数在0.536-0.710范围内,整个量表的内部一致性良好,问卷分半信度均在0.770之上,重测信度为0.832,说明整个量表达到一定信度要求。结论：攻击性问卷在成年男性罪犯中试用具有较高信度和效度,可以作为测量罪犯攻击性的有效工具。

Objective：To examine the reliability and validity of Chinese version of the Aggression Questionnaire（AQ）.Methods：According to the revised scale method,we randomly selected a sample of 190 male criminals in Beijing prisons who were given the test,the reliabilities and validities of Chinese version of AQ were examined.Results：①AQ score and item analysis showed that the AQ scores of the adult male criminals were well balanced,the most（except 5items）of the items had good item discrimination（r〉0.348）;②The result of EFA had five factors（characteristic roots was greater than 1.5,the cumulative amount of variance explained 50.744%）,the five factors were physical aggression,verbal aggression,indirect aggression,anger and hostility.The range of factor loading was between 0.402 and 0.825;③ The content validity of five subscales were between 0.551 and 0.833（P〈0.01）.Selecting WFS as criterion,all correlation coefficients were above 0.574（P〈0.01）.The Chinese version of AQ had a satisfactory validity;④The Cronbach sαcoefficient of the total scale was 0.797,and five subscales was 0.536-0.710.The test retest reliability coefficient was 0.832 for the total scale.All split-half reliability were above 0.770.The whole scale and five subscales were all good on reliability requirements.Conclusion：The Chinese version of AQ has high reliability and validity as an effective instrument for measuring aggression of male criminals.