性侵害收容人再犯危險評估工具效度之研究

Comparing the Validities of Sex Inmates Recidivism Risk Assessments

翁志忠(WENG CHIN CHUNG)

指導教授 ： 張淑慧

中原大學/理學院/心理學研究所/碩士(2017年)

https://doi.org/10.6840/cycu201700127

摘要

國內針對性侵害受刑人實施心理治療及外部監控預防模式歷經約20年，各類再犯評估工具仍持續發展。矯正機關目前使用的工具有靜態99量表、快速性侵害再犯因子評估表、明尼蘇達性罪犯篩選表、暴力危險性評估量表及再犯可能性評估量表，前三項工具之效度檢核已歷經ㄧ段時間，而後二項工具之效度還未經實證檢核，故有檢核上述五種工具效度之必要性，此乃本研究之主要目的。本研究蒐集自民國95年、96年於臺北、臺中、高雄監獄出監996人性罪犯，持續追蹤至102年5月底，資料處理係採差異性檢定及存活時間分析等方法，以檢驗各評估工具的效度。
研究結果顯示:(一)不同的出獄年、機關與再犯性犯罪之間無顯著關聯性。五年內再犯性犯罪與再犯非性犯罪比率分別為5.6% 和32.8%，皆以第1年再犯人數最多，再犯高峰期為出監後三年內；(二)再犯風險評估工具中，靜態99量表、快速性侵害再犯因子評估表及再犯可能性評估量表之總分及風險等級愈高，其再犯率也愈高、性犯罪存活時間則愈低，亦即對再犯性犯罪具有預測力；明尼蘇達性罪犯篩選表之風險等級對於再犯性犯罪預測力不佳，但量表總分對於再犯罪仍具有預測力；暴力危險性評估量表之總分與風險等級對於再犯性犯罪預測力皆欠佳。
依據研究結果作出下列建議，以提供各防治網絡機關及處遇人員在政策規劃或業務執行的參考:(一) 靜態99量表、快速性侵害再犯因子評估表、再犯可能性評估量表可持續使用，對於篩選出的高再犯可能性者規劃適當的監控、治療或處遇；(二)加強對性侵犯再犯高峰期的內在與外在監控； (三)定期追蹤再犯情形且檢視評估工具的效度。

關鍵字

再犯可能性評估量表；快速性侵害再犯因子評估表；明尼蘇達性罪犯篩選表性侵害；效度；暴力危險性評估量表；靜態99

並列摘要

The treatment of psychotherapy and external monitoring prevention mode for sex assault offenders has been implemented for about 20 years in Taiwan. The five scales currently adapted by the correctional institutions includes Static-99, RRASOR, MnSOST-R, Violence danger assessment scale, and Recidivism assessment scale. The former three scales have been tested their validities empirically. The last two scales need to be validated. The purpose of this study was to examine the validities of these five scales. The data was collected from 996 sex offenders released from Taipei Prison, Taichung Prison and Kaohsiung Prison in 2006 and 2007 and continued to track the samples until 2013. Data analysis was based on X2 and survival time analysis.
The results showed that: 1. There was no significant correlation between the time of release, the institution and the recidivism. The rate of repeated sex-offense and non-sex offense in five years was 5.62% and 32.83% respectively, and the most frequent recidivists occurred within the first year. The peak of the recidivism was the first three years after the inmates were released from the prison. 2. The Static-99, RRASOR and Recidivism assessment scales had fair to good predictive validities. Violence danger assessment scale and Recidivism assessment scale did not show enough validities.
The suggestions as follows: 1.The Static-99, RRASOR and Recidivism assessment scales could be used as routine assessment. 2. To strengthen the internal and external monitoring prevention within the peak recidivism period. 3. To follow up recidivism and validate the assessment routinely.

並列關鍵字

RRASOR；recidivism；risk evaluation；Static-99；sex assault；validity

參考文獻

黃健、溫瑞祥、黃永順、鄭安雄 (2005)。性犯罪再犯風險預測因子的本質與應用。亞洲家庭暴力與性侵害期刊，1(1)，137-162。

孫鳳卿 (2001)。性侵害加害人之犯罪型態與危險性探討。高雄醫學大學醫學研究所碩士論文。

許春金、陳玉書、蔡田木、洪千涵、白震福 (2014)。103年犯罪狀況及其分析-2014犯罪趨勢報告。法務部司法官學院104年委託研究計畫。

陳若璋、沈勝昂、林正修 (2005)。性侵害加害人評估使用手冊。內政部性侵害防治委員會委託中華民國團體心理治療學會。

Andrews, D. A., & Bonta, J. (2003). The psychology of criminal conduct(3th ed.). Cincinnati, Ohio: Anderson Pulishing.