過去研究較少有由偵查持續追蹤至判決的研究成果,故本研究透過裁判書的閱讀與編碼,找出施用一級毒品行為人為檢察官具體求刑後,在進入審判程序後的運作現況,並藉此觀測比較具體求刑與法院宣告刑之差異與相關因素。本研究分析515筆案件起訴後的裁判結果,並將裁判書類中客觀可辨,且具有學理討論意味的訊息作為變項,進行判決書編碼,而後再利用類神經網路與邏輯斯迴歸探測變項與變項之間的關聯性。本研究發現,此次的研究對象中(N=515),77.9%使用了簡式審判程序,而絕大多數的被告都被判有罪,比例為99.8%。38.6%案件在宣判時定出應執行刑,但84.3%案件中,法院並沒有回應檢察官所提出的具體求刑。此外,邏輯斯迴歸分析顯示具體求刑的刑度與法院判決的宣告刑之間,不存在統計上的相關,但並不排除受到特定刑事訴訟實務運作干擾因子影響的可能性。本研究結果指出,被告如具有「犯後態度差」、「深陷毒癮難以自拔」、「不思悔改意志不堅」、「危害國民健康、破壞社會秩序」、「數罪併罰」等求刑因子時,為法院判處較重刑期機率較高。此點顯示,相較於求刑刑度,前述因子更能成為法官量刑的參考依據,因此研究建議檢察官或可於前述因子提供更詳盡的說明,來表達科刑意見,以彰顯舉證有罪下的職能影響力。
There are few studies that have been conducted in the past that have followed drug cases from the investigation of indictment to sentencing. In this study, we read and coded the court's judgments to find out the current process of the Schedule I drug offenders after indictments and trial, and to examine the difference between the indictments and the court's judgments and the related factors. In this study, we analyzed 515 indictments and coded the objectively identifiable and theoretically rational information from the paper of judgments as variables. We explored the correlation between the variables from indictments and the judgments by using neural networks and logistic regression. The results show that 77.9% of the samples (N=515) have been applied for summary trial proceeding, and the majority of defendants were found guilty (99.8%). 38.6% of the samples were sentenced with a specific length of prison terms, but in 84.3% of the cases, the court did not respond to the indictment applied by the prosecutor. Logistic regression analysis revealed no statistical correlation of sentence length between indictments and the sentences. One possible explanation could be the correlation was interfered with the litigation practice. In addition, the results indicated that defendants with the factors of "negative attitude towards the courts," "addicted to drugs," "unrepentant,'' ''low continuous retention in the addiction treatment," "harmful to the physical and mental health of nationals and disrupting the social order," and "committing several crimes at the same time" were more likely to be levied a heavier sentence by the court. Therefore, the study suggests that prosecutors may provide more detailed explanations of the above factors to express their opinions on sentencing in order to demonstrate their authority to present proposed charges.