期刊名称：Journal of Systemics, Cybernetics and Informatics
印刷版ISSN：1690-4532
电子版ISSN：1690-4524
出版年度：2018
卷号：16
期号：1
页码：87-92
出版社：International Institute of Informatics and Cybernetics
摘要：The Rasch model is a practical framework for evaluating aconstruct validity of assessment instruments. It is capable ofdetermining how the measurement of person’s ability(endorsement) and item difficulty matches with each other. Thisstudy aimed at evaluating the psychometric properties(reliability, validity, and utility) of a basic science curriculumassessment instrument. Special emphasis was placed on findingthe strengths and challenges in the curriculum, and detecting theexistence of multidimensional structure. A total of 130 medicalstudents in academic year 2016/17 completed a 22-itemassessment instrument. Three major steps were involved in thisstudy. First, the parameters of person’s ability and item difficultywere separately estimated. Second, infit/outfit mean squareresiduals and standardized residual variance from principalcomponent analysis (PCA) were used to validate theunidimentionality assumption. Lastly, differential itemfunctioning (DIF) was assessed to determine the fairness of theassessment instrument. As a result, the baseline measures of thestrengths and challenges in medical curriculum were establishedfor continuous quality improvement. However, the unexplainedvariance for the first contrast value of 3.08 in PCA was greaterthan the criterion of 2.0, which shows some degree of violationof the unidimensionality assumption. Therefore, this instrumentmust be further revised for future application.
关键词：Construct Validity; Curriculum Assessment;Differential Item Functioning; Item Difficulty; Rasch Model
