目的：探討接受強制診療之性侵害加害人否認型態之集群分析及其社會人口變項在強制治療中之意義。方法：本研究採自編之「否認量表」，以接受強制診療之性侵害加害人 47 位為研究樣本。再將社會人口、犯罪史、犯行特徵之資料統整，以單因子變異數分析、集群分析等統計方法進行研究。結果：本研究透過集群分析分成三群，分別是A 型、B 型及C 型，每群與人口統計做交叉分析後，再與否認量尺四因素：強暴事件的迷思、加害者的迷思、受害者的迷思、治療中的否認，做單因子變異數分析，結果達統計意義。結論： 結果顯示接受強制診療之性侵害加害人其否認型態可分類為：(1) A 型：否認型態為對於強暴事件的迷思及加害者的迷思；(2) B 型：否認型態為對於受害者的迷思及治療中的否認；(3) C 型：在各否認型態的傾向特徵為最不明顯者。

Objective: The aim of this study was to explore sexual offender denial types under compulsory treatment using cluster analysis and demographic data. Methods: The subjects included 47 sexual offense prisoners who accepted compulsory treatment. A Denial Scale questionnaire was administrated to all subjects to collect demographic data, criminal history information and of offense characteristics for further data analysis. Analysis of Variance (ANOVA) was used to analyze the data.We also used cluster analysis to analyze three groups: type A, type B and type C. Results: No significant differences were found when each group was analyzed by statistical analysis of population. However, one-way ANOVA produced statistical significance for the four factors on the Denial Scale (rape facts myth, sexual offender's myth, victim myth and denial in treatment). Conclusions: The results indicate that sexual offenders under compulsory treatment can be divided into three denial classifications: Type A: denial condition causing rape facts myth and sexual offender's myth; Type B: denial condition causing victim myth and denial in treatment; Type C: denial condition not significant.