概化理论在《医学统计学》期末考试成绩评估中的应用研究

Application of Generalizability Theory in Evaluation of Academic Record in Final Examination of Medical Statistics

目的应用概化理论的方法来评价《医学统计学》期末考试学生成绩的信度,以提出试卷及阅卷方式的修改建议.方法应用SPSS统计软件包,采用概化理论原理计算各种变异的方差分量估计值,进而计算出相对误差、绝对误差、概化系数、可靠性指数等指标.结果采用随机单面交叉设计时,概化系数为0.64,可靠性指数为0.226.随机双面嵌套设计时,概化系数为0.613,可靠性指数为0.492,均大于0.4.结论概化理论可用于评价医学统计学期末考试成绩的信度,其分析结果对改进试卷结构和阅卷方法有很大的帮助.

Objective To evaluate the reliability of of academic record in final examination of Medical Statistics by using the methods of generalizability theory and to give advice on how to modify examination paper and the way of checking examination papers.Methods Estimation value of variance component from different variations were calculated by applying the principle of generalizability theory and using statistical software SPSS.Furthermore,relative error,absolute error,generalizability coefficient,index of dependability,etc were calculated.Results When it was the design plan that every student answered every item and all items were checking by one teacher,generalizability coefficient was 0.64 and index of dependability was 0.226.When it was the design plan that every student answered every item and all items were divided to four parts without repetitions and items of the four parts were randomly assigned to four teachers to check them,generalizability coefficient was 0.613 and index of dependability was 0.492.Conclusions Generalizability theory can be used in evaluating the reliability of academic record in final examination of Medical Statistic.Its analytic result is very beneficial to the improvement of framework of examination papers and the way of checking examination papers.