Construction of Item Difficulty Model of Senior High School Academic Achievement Test
LIU Fang1**, WANG Wei-Qun2, WU Xing3
1. Jiangsu Province Education Examination Authority, Nanjing 210024, China;
2. College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China;
3. College of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, China
Abstract This paper clarifies the relationship among the factors influencing item difficulty through the main-factor analysis method. On this basis, the main factors influencing item difficulty of senior high school achievement test are found from the three dimensions of problem representation, problem solving and result output. Next, through the writing and testing of items of alternate forms, the difficulty sequence of factors of different types and corresponding grading rules are established. After assigning values to the 7 factors of all the items of alternate forms, the linear regression method of machine learning is used to construct the model. In addition, the model is tested with data generated from part of the constructed items in college entrance examination chemistry Jiangsu paper. The predicted value of the model and the measured value have a good degree of fit, indicating the application value and promotion significance of the model.
LIU Fang, WANG Wei-Qun, WU Xing. Construction of Item Difficulty Model of Senior High School Academic Achievement Test[J]. Chinese Journal of Chemical Education, 2022, 43(21): 43-47.