大學生基本素養測驗的發展及信度效度分析

Evaluation of the Psychometric Properties of the General Literacy Test for University

陳柏熹(Po-Hsi Chen) ； 黃馨瑩(Hsin-Ying Huang) ； 陳郁欣(Yu-Hsin Chen) ； 葉泰廷(Tai-Ting Yeh) ； 蘇少祖(Shao-Tsu Su)

《教育科學研究期刊》 60卷3期 (2015/09) Pp. 95-126

https://doi.org/10.6209/JORIES.2015.60(3).04

摘要

本研究目的旨在發展大學生基本素養測驗並進行信度與效度評估。藉由分析國內大專院校的通識教育目標和核心素養，並參考ATC21S提出的21世紀現代學生需具備的10項基本素養，歸納出大學生基本素養測驗的九項素養，分別為：溝通合作、美感素養、科學思辨、資訊素養、終身學習、創新領導、問題解決、公民社會及生涯發展。測驗形式為線上多媒體情境式題型，每個題本均包含九項素養的內容，每項素養皆有二至三個題組。研究對象為全國大專校院一至四年級學生，研究樣本來自20校10,958名大學生。由效度評估結果可知，大學生基本素養測驗的題組效果不大，可以忽略，並採用部分計分模式來估計，幾乎所有試題與模式都能適配，顯示建構效度良好。試題發展過程均歷經嚴謹修審題程序，取得良好專家效度證據。此外，不同性別和年級的學生在各素養的表現上差異不大，和過去的文獻相符合，具有良好的效標關聯效度。信度證據方面，各素養能力估計誤差約在 .20～.60 logit之間，單一題本的信度高於.69，顯示本測驗題數雖少，但信度大致良好。整體而言，大學生基本素養測驗具良好的信度與效度。

關鍵字

信度 ； 效度 ； 素養評量 ； 情境式測驗 ； 線上測驗

並列摘要

This study evaluated the psychometric properties of the General Literacy Test for University Students. To develop the assessment framework, the educational objectives of general literacy courses of universities in Taiwan as well as the core competencies of Assessment and Teaching of 21st Century Skills were all reviewed and considered. The general literacy test is composed of nine literacy domains: communication and collaboration, esthetics, information, lifelong learning, career, leadership, problem solving, social concerns and citizenship, and scientific thinking. The items of the general literacy test were developed into a multimedia assessment. In addition, booklets including two to three testlets on each of the nine literacy domains were organized. Overall, 10,958 students from 20 universities participated. The results demonstrated that testlet effect could be ignored and the partial credit model could be used to analyze data. In addition, the items fit the partial credit model closely and the test demonstrated favorable construct validity. In addition, small differences existed between gender and grades. The results were consistent with those of previous studies and showed that the test's criterion-related validity was acceptable. Furthermore, the standard error of ability ranged from .20 to .60 logit and the reliabilities of the booklets were more than .69 in the nine literacy domains. The evidence supports the high degree of reliability and validity of basic literacy assessment.

並列關鍵字

reliability ； validity ； literacy assessment ； scenario based test ； online test

參考文獻

王保進（2011）。以學生學習成效品質保證為核心之第二週期系所評鑑作業規範。評鑑雙月刊，33，7-11。【Wang, B.-J. (2011). Student learning outcomes associating with quality assurance mechanism for department and graduate institute evaluation. Evaluation Bimonthly, 33, 7-11.】

陳儀蓉、黃芳銘（2006）。組織公民行為量表在男女員工群體上之測驗恆等性檢定。測驗學刊，53（2），297-326。【Chen, Y.-J., & Hwang, F.-M. (2006). Assessing measurement invariance of organizational citizenship behavior citizenship behavior scale between male and female staff. Psychological Testing, 53(2), 297-326.】

Bock, R. D., & Mislevy, R. J. (1982). Adaptive EAP estimation of ability in a microcomputer environment. Applied Psychological Measurement, 6(4), 431-444.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New Jersey, NY: Lawrence Erlbaum Associates.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155-159.