數學後設認知量表之發展與信效度考驗

An Evaluation of the Reliability and Validity of the Mathematical Metacognition Scale

本研究旨在發展適用中小學生的數學後設認知量表（Mathematical Metacognition Scale， MMS）， 本研究分預試與正式施測兩階段，兩階段皆採立意取樣的叢集抽樣方式，各抽取中小學生457 人與861 人。預試資料進行探索性與驗證性因素分析，驗證性因素分析結果獲得三個因素：後 設認知知識、計劃與監控、評估，其內部一致性Cronbach’s α 係數分別為 .87、.87、.72，12 題 的總量表Cronbach’s α 係數為 .90，顯示本量表具有良好的信度。正式施測資料，進行測量恆 等性考驗，考驗結果發現不同性別具測量恆等性，顯示本研究發展的數學後設認知量表，具有 良好的建構效度。

The main purpose of this study was to develop the Mathematical Metacognition Scale (MMS) on 5th – 8th graders in elementary and junior high schools. The present study included two phases: pretest and test. By adopting purposeful sampling and cluster sampling in both phases， 457 and 861 students were selected for the pretest and the test， respectively. Exploratory factor analysis and confirmatory factor analysis were conducted with the pretest data. The confirmatory factor analysis resulted in three factors including metacognitive knowledge， planning and monitoring， and evaluation. The Cronbach's alphas of the MMS subscales were .87， .87， .72， and .90 for the total score， which indicated good reliability. Measurement invariance was conducted to the test and indicated gender invariance. Therefore， it showed that the MMS has good construct validity.