大學生評鑑教學量表：編製及效度考驗

Student Evaluation of Effective University Teaching in Taiwan: Scale Development and Factor Structure

本研究以「有效教學」概念為核心，設計出適於台灣大學生採用的總結性教學評鑑量表。正式測驗包含北部某私立大學22000餘名學生在2670班的評鑑資料。根據有效教學的假設模式把量表的15個題目分屬四個第一階層潛在因素：敬業精神、教學方法、授課內容、及學習成效，第一階層因素再匯整成為單一個第二階因素：教學。隨機抽取614班進行驗證性因素分析，發現全體學生所秉持的評鑑概念與假設模有所不同，其原因可能是不同學系的老師達成優良教學有不同的方式，或學生對不同學門課程採用不同的評鑑標準，加上評鑑干擾因素的作用。如果把受群體限制為相近的課程範圍，如大一英文及大一經濟學等多班級科目，則學生對「有效教學」的概念能符合教育學者提出的二階層因素結構，因而肯定學生能以合理的量尺來評鑑教學。建議以「大學生評鑑教學量表」進行總結性教學評鑑時，不應進行全校性評比，而應將比較範圍限制為任教於同一類型科目的老師，則學生所秉持的評鑑量尺才能顯示出合理的結構。

This study present s a reliable and valid， summative instrument for evaluating university teaching by Taiwan students， namely Student Evaluation of University Teaching in Taiwan （SEUTT）. The initial items were constructed with careful analysis of the definition of effective university teaching to make it appropriate for Taiwan universities. The hypothetical model designated 15 items to 4 first─order factors：commitment， instructional method， instructional content， and learning effect. Then the first─order factors were clustered as a single second─order factor， teaching.A subject pool with 22，000 undergraduates of 2670 classes was included in the formal test. A random sample of 614 classes was selected form the subject pool I the validity examination. Although reliability indices were high， a CFA showed the hypothetical model was significantly different from the model derived from the observed data. In examining effects of the unrelated variables to effective teaching（e.g. colleges， classes size， perceived grading justice， grade expected perceived workload， reason to take course.）， all of t hem substantially influenced evaluation. Besides， ideal instruction would be very different. Therefore， other CFAs were performed， but limited to multisection courses. Freshman English and Introduction to Economics were used as examples， and the result showed responses in both multisection courses did confirm to the hypothetical model. In using SEUTT， a comprehensive comparison including all faculty members may not be adequate. The evaluation scale in students’ minds will show effectiveness only when we limit the comparison for teachers in teaching the same multisection course.