Alireza Mehridehnavi - ۱.Medical Physics and Engineering Department, Medical School, Isfahan University of Medical Sciences, Isfahan, IRAN ۲.School of Optometry and Visual Science, University of Waterloo, Waterloo, Canada.
Hossein Mehri-Dehnavi - ۳.Department of Physics, Babol University of Technology, Babol, Iran.

Background: Score worksheets have been extracted for PhD candidates in Epidemiology, Microbiology and Parasitology for the academic year of 1386-87. The main objective in this study is the impact and importance of different subjects on the discrimination results of admission. Methods: Extracted scores from worksheets are analyzed by MINITAB15 statistical toolbox. The regression values of different subjects were extracted and sorted for each branch of study. All subject correlation coefficients were calculated in comparison to overall candidate performance. A redundancy measure was suggested on the basis of correlation of different subject performances. The distribution box plots were shown for different subject to visualize comparison. Results: The repeatability and redundancy between different subjects were high in bacteriology. The different subjects in this branch of study had efficient and equal impact on discrimination. The bacteriology as a subject had shown lowest significance in epidemiology, therefore it has the lowest taxonomy. The bacteriology and fungicide subject despite its low weight  showed the highest impact of 6.44 in parasitology . Conclusion: In this study we realized that in branches such as epidemiology different subjects have equal impact. In subject cases which the significance of impacts are very low, there should be further consideration and analysis on the subject matter to whether wave the subject. In parasitology branch bacteriology and fungicide subject has higher impact on discrimination, it is worth to increase its weight and the number of questions on the subject.

Score Coefficients, Written Exam, Basic Science, Subject Correlation, Density Distribution