敏感问题的调查与统计处理技术及其在学生考试作弊行为调查上的应用

Application of the Randomized Response Technique in Estimating the Proportion of cheating on Tests

目的：检验二个无关问题模型在学生作弊调查中的应用效率。方法：结合两个无关问题模型介绍了敏感问题的调查与统计处理技术（ＲＲＴ）的原理、实施步骤及计算方法。同时应用两个无关问题模型调查了某校两个班考试作弊情况，并作相应统计处理。结果：经分析发现，敏感问题在第二次调查时应答率下降（Ｐ＝０．０１），作弊率也有下降倾向，但Ｐ＞０．０５。不同性别作弊率则明显有差异，男生高于女生（Ｐ＝０．０２４７）。两个班作弊率没发现统计差异，总体作弊率较高。结论：调查结果说明用此法调查学生作弊情况是可行的。

Objective:To estimate the effect of two unrelated questions randomized response model in surveying proportion of cheating on test (PCOT).Methods:In this pape,two unrelated questions randomized response model of RRT was discussed and used in a certain college to estimate the PCOT among the students of two classes.Results:Response proportion of Sensitive questions was lower in the second survey(P=0.01).The PCOT was inclined to lower in the second,but no significant difference of statistics was found(P>0.05).The PCOT of male students was higher than femal(P=0.0247).The PCOT was different between two classes.The total PCOT was in high level.Conclusion:The results revealed that RRT can be used in survey of PCOT.