한국수학교육학회지시리즈A:수학교육 (The Mathematical Education)
- 제46권4호
- /
- Pages.407-421
- /
- 2007
- /
- 1225-1380(pISSN)
- /
- 2287-9633(eISSN)
대학수학능력시험 수리 영역 문항 난이도 예측을 위한 회귀모형 추정
Estimating the regression equations for predicting item difficulty of mathematics in the College Scholastic Ability Test
- Lee, Sang-Ha (Korea Institute of Curriculum & Evaluation) ;
- Lee, Bong-Ju (Korea Institute of Curriculum & Evaluation) ;
- Son, Hong-Chan (Korea Institute of Curriculum & Evaluation)
- 발행 : 2007.11.30
초록
The purpose of this study is to identify the item characteristics that are supposed to affect item difficulty and to estimate the regression equations for predicting item difficulty of mathematics in the College Scholastic Ability Test(CSAT). We selected six variables related to item characteristics based on learning theories: contents, cognitive domain, novelty, item type, number of concepts, and the amount of computation. With data of the CSAT mathematics test administered in 2004-2006, item difficulty was regressed on the six variables, the location of an item, and the item writer's judgment on difficulty. The novelty of an item was found to be a statistically insignificant variable in explaining item difficulty. Four regression equations with different sets of independent variables could explain
키워드