高雄市藥物濫用者再犯風險與保護評估量表之建立研究

Research on the establishment of recidivism risk and protection assessment scale for drug addicts in Kaohsiung City

楊士隆(Shu-Lung Yang) ； 林瑩蓉(Ying-Jung Lin) ； 李志恒(Jih-Heng Li) ； 郭昭宏(Chao-Hung Kuo) ； 黃光甫(Kuang-Fu Huang) ； 許俊龍(Chung-Lung Hsu) ； 劉堯民(Yao-Min Liu)

《藥物濫用防治》 7卷2期 (2022/12) Pp. 1-32

https://doi.org/10.6645/JSAR.202212_7(2).1

摘要

毒品犯罪問題於我國一直為首要解決難題之一，如何減少毒品的供給與降低毒品的需求勢必是兩大重要著重面向。然而，作為減少需求面向中的毒品施用人數卻不易下降，顯見為減少藥物濫用仍需投注更多的資源與精進政策。且為使處遇資源得以有效運用於再次施用藥物的預防，勢必需要評量工具以測量其施用風險，以達準確評估個案之狀況。基此，本研究透過國內外研究文獻探討，編製問卷並進行專家焦點座談審視及預試後，針對高雄市211名因犯施用毒品罪而列管之個案進行施測檢視，期能透過實證性之研究建立適合高雄市之藥物濫用者再犯風險評估量表。基於研究理論及實務專家學者之意見，本研究建構風險與保護六大面向因子，並據以編製評估量表。研究發現本量表研究變項之Cronbach's α皆大於0.7，各構面的因素負荷量皆大於0.5，總解釋變異量也大於50%，顯示本問卷的收斂效度良好。此外，所有構面的積差相關係數皆小於構面的信度，表示本問卷的具有良好的區別效度。而風險量表部分前27%與後27%的受測者在回答上有顯著性的差異；保護因子量表部分，同樣的顯示出前27%與後27%的受測者在回答有顯著的差異性，表示量表具有有鑑別度。量表整體模式適配度顯著，且風險動態因子整體預測率達76.6%，而保護因子整體預測率達75.6%。本研究建議後續可持續追蹤本研究樣本，據以確保本量表之信效度與再犯之預測率。而在實際量表施測操作上，建議持續強化辦理相關研習，以精進操作人員對於量表之操作與運用解釋。

關鍵字

保護因子 ； 風險因子 ； 再犯 ； 評估量表 ； 藥物濫用

並列摘要

The drug problem has always been one of the primary problems in Taiwan. Reducing the supply of drugs and reducing the demand for drugs are two critical aspects. However, the number of drug users is not easy to decline. Therefore, it is evident that more resources and comprehensive policies are still needed to reduce drug abuse. Furthermore, to effectively utilize treatment resources to prevent drug recidivism, assessment tools are necessary to accurately measure the risk of recidivism and assess the case's status. Therefore, based on domestic and international research literature, this study researched 211 drug use cases in Kaohsiung City via compiling questionnaires, holding focus groups, and conducting pre-tests to establish a recidivism risk assessment scale for drug addicts in Kaohsiung City. Based on the research theory and the opinions of experienced practitioners and scholars, this study constructs six major factors related to risk and protection and compiles the assessment scale accordingly. The study found that Cronbach's α of the research variables of this scale was all greater than 0.7, the factor loadings of each dimension were all greater than 0.5, and the total explained variance was greater than 50%, indicating that the convergent validity of the questionnaire was good. In addition, the product-moment correlation coefficients of all dimensions are less than the reliability of the dimensions, indicating that the questionnaire has good discriminant validity. In terms of the risk factor scale, there is a significant difference in the responses of the first 27% and the last 27% of the respondents; in terms of the protection factor scale, similarly, the first 27% and the last 27% of the respondents have a significant difference in the answers. The difference means that the scale has a degree of discrimination. The overall model adaptation degree of the scale was significant, the overall prediction rate of dynamic risk factors was 76.6%, and the overall prediction rate of protection factors was 75.6%. This study suggests that the research sample of this study should be continuously tracked in the follow-up to ensure the reliability and validity of this scale and the prediction rate of recidivism. In addition, this study recommended constantly holding relevant studies or workshops to improve the related officials' operation and interpretation of the assessment scale.

並列關鍵字

protective factor ； risk factor ； recidivism ； assessment scale ； substance abuse