數學 學力評價法의 比較硏究

Abstract

數學의 學力을 評價함에 있어서 選多型問項型式의 評價方法과 論文型評價方法의 장·단점과 有意性을 統計的으로 究明하고자 한다. 이에 서울에 있는 인문계 고교중 평균화된 여자고등학교 문과 120명 이과 120명 되합 480명을 標本으로 論文型과 選多型의 問項 15個로 調査하였다. 여기에서 論文型과 選多型은 有答肢와 無答肢의 차이로서 答肢를 選多型에는 四個주어 四肢選多型으로 作成하였다. 선다형, 논문형은 答肢의 有無의 差異가 있을뿐 그외의 問題의 性質은 同質인 것이다. 本論文에서는 2가지 型式에 대해서 정답율, 정답율의 差의 有意性, 相關係數, 相關係數의 有意性을 檢定하여 問項別 辨別力을 査定하여보고자 한다. 檢定結果 學生의 수준이 문항에 비해 떨어져서 상관계수의 차의 有意性과 回歸係數의 有意性의 分析이 意味가 없어 正答率의 差의 有意性을 檢定하여 各問項의 分布상태 및 辨別力을 검정하였다. ;The purpose of this thesis is to examine the advantages and disadvantages, and significance that the multiple-choice items or the essay-type items have in evaluation the achievement in mathematics, For the purpose of ascertaining them, a total of 480 students(that is, 240 in the department of liberal arts and 240 in the science department) from a girl's high school was sampled, Each of 480 students was examined with the 15 multiple-choice items(four-option items) and the 15 essay-type: items respectively. Each of two tests mentioned above is two different types requiring the same answers. The discriminatice power of two type tests -- the multiple-choice items and the essay-type items --- was measured and analyzed by examining whether or not the difference between the right answer rate of the multiple-choice items and that of the essay-type items is significant. The result is as follows: (1) The multiple-choice items are nearly Normal Distribution. The essay-type items are L-type Distribution, and this figures show that most students have no ability to solve problems by themselves. (2) In analysing (1), we can guess that most students are familiar with the multiple-choice items, and the guessing factors for the right answers have a great influence upon them. And it was found that the essay-type items, which is a good method for measuring the logical thinking ability, are not Normal Distribution. It means that the educational system in force doesn't help develpp the logical thinking ability. (3) The discriminative power of the multiple-choice items is similar to that of the essay-type items, in the problems which measure "the memory and understanding of basic comceptions." (4) Regardless of the multiple-choice items or the essay-type items, the difficult problems have poor discriminative power. (5) In the problems which mea sure "the faculty of calculation" or "the conceptions and applications of principles", the difference between the right answer rate of the multiple-choice items and that of the essay-type items is significant.