A Bayesian Approach for Item Response Theory in Assessing the Progress Test in Medical Students

The progress test is used to provide useful summative and formative judgments about medical students' knowledge without distorting learning. The test samples the complete knowledge domain expected of new graduates on completion of their course, regardless of the year level of the student. Item Response Theory (IRT) have been proposed for evaluating examinations in medical education. Although IRT relies upon some strong assumptions, it is useful and practical in many situations such as medical education testing. IRT assumptions are often violated and threaten the statistical inferences. To overcome these difficulties, Bayesian approach to IRT can be used. The objective of this paper is to provide an overview of both Item Response Theory (IRT) and Bayesian IRT to the practitioner involved in the development of medical education assessments. We analysed the results from the Progress test examination taken by 4th year undergraduate medical students in the academic year 2010 to 2011. Advantages and disadvantages of IRT are initially described. Secondly, Bayesian approach for IRT is given. Bayesian approach is applied to IRT data set from Ondokuz Mayıs University. Finally, general recommendations to the use of Bayesian approach in practice are outlined. We found support for the use of Bayesian approach for IRT in assessing of progress testing.