本研究的主要目的在編製一份適用於臺灣大學的道德判斷評量工具，該評量表採用Kohlberg之理論架構以及Rest之評量形式，同時參考有關華人道德判斷的論述與研究，添加若干適於鑑別華人道德判斷殊相的指標。其編製過程如下。 一、 編制預試題本：參照Kohlberg的「道德評量晤談」編撰困境與在各困境中支持不同道德判斷的代表性理由，再加上道德判斷之普遍性與驅動性題項。 二、 項目分析：以分層叢集隨機取樣抽取516名大學生進行預試，根據各項目之平均數、變異數、與同分量表其他題項相關以及項目刪除後Chronbach α增加值等標準進行篩選。 三、 心理計量品質考驗：資料來自以分層叢集隨機取樣抽取的755名大學生。在信度考驗上採用Cronbach α係數進行內部一致性考驗。在效度考驗上，以皮爾森積差相關矩陣考驗分量表間的輻合與區域辨效度；以重複量數多變項變異數分析考驗道德結構與道德價值之跨情境差異以及道德結構間差異與道德價值間差異；以線性結構模式考驗觀察資料符合理論建構的程度。常模部份則考驗性別與年齡的差異。上述考驗結果大多相當正向。

　　　　　The purpose of this study was to develop the Moral Judgment Inventory for college students, which was based on Kohlberg's latest theoretical framework and took the form used by Rest. Literature on Chinese moral reasoning was also taken into account. The development process was as the follows. 1. The development of the pretest version. First, three dilemma, as well as correspondent reasons, from Kohlberg's Moral Judgment Interview were selected, translated, and revised Ratings on universalizability and driving potential of moral judgment were also included. 2. The pretest of the pretest version. It took place with 516 college students, sampled via stratified cluster random sampling. Item analyses were based on mean, standard deviation, within-structure consistency, within-value consistency, and after-deletion Chronbach α increase. 3. Examinations on psychometrical characteristics of the formal version. Stratified cluster radom sampling was used to collect data from 755 college students. As to reliability, internal consistency, indexed by Cronbach α coefficient, was examined. As to validity, the procedures used included (1)Pearson Product correlation for convergent and discriminant validity, (2)repeated measure multivariate ANOVA for cross-situation differences as well as cross-structure and crossvalue differences, and (3)linear structural equation model for the measurement model on moral structure and moral value, respectively. As to norm, examinations on sex and grade differences were conducted. Most of the above tests revealed positive evidences.