本研究的目的在於瞭解台灣學生在TIMSS 2003試測的數學成就表現。研究方法完全遵照國際測驗中心的程序,受測樣本從全國小學依隨機抽樣抽出50班,共1601位四年級學生。本研究的主要發現:全部的229題試題中,有74題的數學內容尚未正式列入我國八十二年版的四年級數學課程中。台灣學生的平均通過率由高至低為數、幾何、資料、代數、測量分別為55.8%、54.15、52.5%、48.6%、48.0%。除了代數之外,台灣學生在其他四個主題的平均通過率皆高於其他國家學生的平均通過率。在這次試測中,分別有15,17,8,3,6題是學生學過的試題,而我國學生的成就表現低於國際平均通過率。本研究發現影響學生數學成就的可能因素包括:評量的形式(選擇題、填充題),評量的內容(操作性的圖卡)、題目本身命題不適當、題目的語意不清,題目的情境不合台灣國情文化、學生沒有學過相關的數學內容、解題時忽略真實情境的考量。本研究的發現可以提供給我國規劃九年一貫數學課程的參考,評量的方法及形式可以提供給教師評量學生的參考。
The study was intended to investigate students' achievement of mathematics in TIMSS 2003 field test. The methodology including sampling, data processing followed the processes of TIMSS 2003 field test controlled by International Study Center. The1601 fourth graders from 50 classes were randomly drawn from 25 elementary schools. The main results of the Study included: The students' performance in five mathematics areas from high to low in order was number, geometry, data, algebra, and measurement. The only exception was algebra. Taiwan students were achieved better in other four areas than International students. The mathematics contents referred to 74 assessment items were not learned in the 1993 version of mathematics curriculum standards of Taiwan. Students were not performed well in the five areas, it's proportion corresponding to(the equation is abbreviated), respectively. The factors influenced students' achievement including: format of assessment (multiple choice v.s constructed responses), content of assessment (number, algebra, measurement, geometry, data), items itself (fitness, clarity, context), students' ignorance of authentic context, the contents involving in items students unlearned.