Sinharay Sandip, Johnson Matthew S
Educational Testing Service, NJ, USA.
Appl Psychol Meas. 2025 Jan 27:01466216251316559. doi: 10.1177/01466216251316559.
This article suggests a new approach based on Bayesian decision theory (e.g., Cronbach & Gleser, 1965; Ferguson, 1967) for detection of test fraud. The approach leads to a simple decision rule that involves the computation of the posterior probability that an examinee committed test fraud given the data. The suggested approach was applied to a real data set that involved actual test fraud.
本文提出了一种基于贝叶斯决策理论（例如，克龙巴赫和格莱泽，1965年；弗格森，1967年）的新方法来检测考试作弊行为。该方法引出了一个简单的决策规则，该规则涉及计算在给定数据的情况下考生进行考试作弊的后验概率。所建议的方法被应用于一个涉及实际考试作弊行为的真实数据集。
