跳到主要內容區塊 :::
   
:::
首頁/研究主軸/績效責任/研究計畫/從TASA 與TIMSS 資料庫探究臺灣學生數學學習表現

從TASA 與TIMSS 資料庫探究臺灣學生數學學習表現

  • 資料類型

    研究計畫

  • 計畫編號

    NAER-103-12-B-2-04-00-1-05

  • GRB編號

  • 計畫名稱

    從TASA 與TIMSS 資料庫探究臺灣學生數學學習表現

  • 計畫主持人

    曾建銘

  • 共同主持人

    吳慧珉

  • 經費來源

    國家教育研究院

  • 執行方式

    自行研究(本院經費-本院人員)

  • 執行機構

    國家教育研究院

  • 執行單位

    測驗及評量研究中心

  • 年度

    2014

  • 期程(起)

    2014-01-01

  • 期程(迄)

    2015-04-30

  • 執行狀態

    已結案

  • 關鍵詞

    大型測驗,數學,臺灣學生學習成就評量,國際數理成就趨勢調查

  • Keywords

    Large scale assessment,MATH,TASA,TIMSS

  • 研究主軸

    中小學生成就長期追蹤

  • 一、研究內容:

      本研究是一年期計畫,使用TASA 資料庫和TIMSS 資料庫,透過文件分析、專家會議及統計分析等方法,整合分析臺灣學生在不同資料庫之問卷表現與評量內容表現,及透過不同資料庫之量化數據,以了解臺灣學生之數學學習情形。綜上得到以下結果:

    (一)若考量TASA 要與國際做比較,Propensity Score Matching(PSM)的方法是一可行的統計方法,但背景問卷的問題內容宜盡量相同,若只是意義相近,難免還是會造成受試者─學生解讀有異,而產生作答信度降低,影響效度。

    (二)因為測驗目的不同,所以在評量架構領域內容的比率也不同,TASA 代 數的比重較重遠高於TIMSS,多了13.6%;而在資料與機率部份TIMSS 則比TASA 多了10.4%,約是二倍。

    (三)以兩測驗的評量架構與試題內容的比較,TASA 比較強調代數與幾何的數學知識,而TIMSS 測驗雖然也是偏向各國課程妥協的設計,但是還是可以看出,其偏向日常生活的應用,尤其是在數、資料與機率兩內容領域的比重明顯大於TASA。

    (四)由PSM 挑出兩考試中背景相似的兩群人,在TIMSS 與TASA 的試題難度排序結果發現,在TIMSS 中覺得較困難的試題,幾乎是TASA 評量架構 中或實際的九年一貫能力指標中沒有的部分。可見對於臺灣學生而言,對於資料詮釋的推理且屬於建構反應試題是感到困難的,若本國數學課程擬隨世界潮流強調素養能力,可能可以考慮加入日常生活類似TIMSS 的數與資料詮釋內容。

    二、對教育政策之建議與貢獻:

    建議一:TASA 應參考國際測驗問卷設計之理論,設計一份具有理論基礎並符合本土教育社會與文化的學生、教師、學校與家長問卷,以符應臺灣的教育政策與現場教學及供學者教育研究用。

    建議二:若擬進行國際教育測驗資料比較,首先要注意目的、評量架構與問卷內容的差異,每個測驗皆有其目的,但在達成目的的前提,若擬進行國際比較,最好能對評量架構的領域與內容及定義就擬比較部分調成一致。

    建議三:進行跨國測驗資料的比較,不能單純由學生作答反應的結果,來進行比較、作相關建議,還是要看評量架構與實施之課程內容。

    建議四:我國擬實施十二年基本國教,而我國所參與的國際大型測驗結果,可供課程研擬做參考。

  •   This study is proposed for one-year period of research implementation. The aim of this study is to provide useful information for investigating and comparing the students’ performance from TASA and TIMSS databases, based on an in-depth look at the content of the respective frameworks and assessment items. The study has the following results:

    (A) If you want to compare TASA with international large scale assessment, Propensity Score Matching (PSM) is an available method of statistical methods. But the content of items in background questionnaire have to be the same as soon as possible, if only similar meaning, that will inevitably result in students have different interpretation, arising reliability was reduced, and affecting the validity.

    (B) Because of the different test purposes, therefore, the area content proportion of framework is different. The proportion of Algebra area in TASA is much higher than in TIMSS, more than 13.6 percent; and TIMSS is much higher than TASA 10.4%, about twice.

    (C) Comparison the assessment framework and content of TASA and TIMSS, TASA more emphasis on Algebra and Geometry, and although the TIMSS test is curriculum design which is a compromised among nations. TIMSS is tend to the daily life of the application especially in the number as well as data and chance two content areas.

    (D) By the PSM selected out two groups with similar backgrounds, the sort of item difficulty in TIMSS and TASA was found that the more difficult items in TIMSS, almost absent in TASA assessment framework or Competence Indicators of Grade 1-9 curriculum. The reasoning and data interpretation items which belong to the construction response type are troublesome for Taiwan students. If the new national mathematics curriculum will follow the world trend to emphasize literacy competency, it may consider joining the daily life and data interpretation content areas which are similar to TIMSS content.

    Recommendations for policy:

    (A) TASA should refer to the theory of international test questionnaire design to design students, teachers, schools and parents questionnaire which are with theoretical basis and correspond to local education community and culture in order to accord with Taiwan's education policy, on-site instruction, and to provide for education scholars use.

    (B) If the international education test data were compared, the first is to note the differences of purpose, framework and questionnaire between assessments. Each assessment has its unique purpose. There is a precondition for the purpose, if the research wants to compare different national data. It is the best to adjust the framework, content and definition into a consistent for the comparison section.

    (C) It has risk to compare international test data cannot simply look at the results of students’ response data and make recommendations. The more important is to see the assessment framework and implementation curriculum.

    (D) The12-year compulsory education will be implemented in recent years. The test results of the international large scale test could be the reference for curriculum elaboration.

top
回首頁 網站導覽 FAQ 意見信箱 EN
facebook youtube