96-100學年度大學入學考試化學考科試題分析研究 A Study of Item Analysis of Chemistry Test of DRT 2007-2011

Date
2012
Authors
吳國良
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究利用大學入學考試中心,96-100學年度指考化學考科的試題進行分析研究,主要的方向包括不同類別考生成績的表現、試題類型、試題難度、學生作答上較困難試題與非選擇題考生作答分析。 在不同類別考生成績的表現,是以全體參與96-100學年度的考生為比較的基礎;而試題類型、試題難度、學生作答上較困難試題與非選擇題考生作答分析,則採取一組9所學校的學生,分布區域在臺灣的北、中、南三個地區,並含有指考化學考科高、中、低三種成績組合。以100學年度而言,此組學校學生人數共計1,038位,對照全體考生,其成績分布卡方檢定值為17.03小於23.68,該組合學校的學生樣本,可以代表全體考生的分布。不同類別考生成績表現而言,採敘述統計方式;就試題類型,是採因素分析、階層性集群分析與專家判斷法;試題難度採多元迴歸分析;學生作答上較困難試題調查則採專家問卷調查方式;考生非選擇題作答則採類型分析方式。 本研究結果顯示,指考化學考科的成績表現,就不同類別考生而言,是男生優於女生、非應屆優於應屆、公立優於私立、學校規模愈大,考生成績愈好。其中,男女與畢業別的差異,效果量小於.2、學校別差異的效果量介於.2~.5、學校規模差異的效果量則是介於.5~.8。以試題類型而言,指考化學考科有三種主要的試題類型,其試題難度呈現階層性關係並具有顯著性差異;就影響試題難度的因素而言,本研究利用四種試題的屬性,對試題難度變異的解釋力可達64%;利用難度預估的迴歸方程式,計算化學考科5年113題所得的難度值與實測值之間比較,若以四種難度類別作區分,大部分是屬於同一類別或差別一個難度類別,僅1題差別兩個難度類別;關於學生作答上較困難的原因,與上述影響試題難度的變項相近,有些則是屬於學生的迷思概念;考生非選擇題的作答分析發現,不同年度的非選擇題,考生的作答表現,可以分類成不同的層級,並可找出考生實際作答的例子進行驗證。未來可嘗試不同科目或不同測驗的試題難度研究,以及教學上如何增進學生較高階認知能力等方向進行。
The purpose of this study was to investigate the items in chemistry of 2007-2011 Department Required Test (DRT) of College Entrance Examination Center (CEEC). The main purposes of this study are to compare the relationships of the student’s characteristics and chemistry test grade of DRT, to analyze the item formats, item difficulty, student’s poor performance items and response patterns of open-ended questions. Except comparing the total students of 2007-2011 involving the DRT chemistry test, the researcher also sampled a group of 9 school students, which including the northern, middle, and southern areas of Taiwan, and with different DRT chemistry achievements. According to chi-square test, this sampling was an appropriate combination. By the methods of factor analysis, cluster analysis, multiple regression analysis, questionnaire analysis, and response patterns analysis, this study had following results. Firstly, the performance of students varied with gender, retaken the test or not, and school type. The male students outperformed than the female ; the retaken students got better scores than the first-taken ones; the public school students performed better than those of private schools; the larger the schools were, the better performance were the students. But, the difference of gender and taking times did not have small effect size, and the difference of school type had small to medium, and the difference of school size had medium to large effect size. Secondly, there were three item formats in various years’ tests, and the passing rate also showed hierarchical relationships of these three item formats. Thirdly, the multiple regression analysis showed that there were four item attributes with great effect on item difficulty which accounted for 64% variance. Lastly, the reasons of student’s poor performance items were similar to those item attributes, besides that some student’s poor performance due to the misconceptions, and according to the response patterns of open-ended questions, the student’s performance has divided with different procedural and conceptual knowledge levels. Taken together, this study has provided methodology of item analysis and an example for other subjects. Furthermore, the item analysis also gives great help for school instruction and future item-writing.
Description
Keywords
指考, 化學考科, 試題類型, 試題難度, 試題屬性, 迷思概念, DRT, chemistry test, item format, item difficulty, item attribute, misconception
Citation
Collections