這句話實在是很模糊,即使翻譯成中文仍不太能理解。再多看幾篇文獻深入探究後,將我目前瞭解的重點彙整如下:
1. 此處的信度是指「內在信度」。量表的信度可分兩類:「外在信度」(external reliability) 與「內在信度」(internal reliability)。外在信度是指再測信度與施測者間信度。而內在信度則是指內在一致性,亦即組成量表題項的一致程度如何。
2. Cronbach's alpha是基於"essentially tau-equivalent model"之假設所設計。
3. essentially tau-equivalent model假設一量表中,每個題項都是測量相同的潛在特質、具相同scale (計分方式與概念同、每一題的標準差相似),但每一題的精準度以及變異可不同。亦即每一題項所測量到的True score (T)都一樣,Observed score (O)的差異只受題目精準度影響(例如:比較specific或general的問法,所得分數的精準度不同)。
4. 但現實情況中,量表難以完全符合essentially tau-equivalent model之假設,亦即每一題分數之變異不同。又由信度之概念公式:reliability=Var(T)/Var(O),當每一題的Observed score變異數不一,使得量表之Observed score總變數增加,信度將下降。即若違反essentially tau-equivalent model之假設,將低估信度。
綜合上述可知:以Cronbach's alpha容易低估量表的內在信度,換言之,以Cronbach's alpha計信度較保守,是估計信度的最低限度。
References:
Graham, J. M. (2006). Congeneric and (Essentially) Tau-Equivalent Estimates of Score Reliability: What They Are and How to Use Them. Educational and Psychological Measurement, 66, 930-944.
Gu, F., Little, T. D., & Kingston, N. M. (2013). Misestimation of Reliability Using Coefficient Alpha and Structural Equation Modeling When Assumptions of Tau-Equivalence and Uncorrelated Errors Are Violated. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 9, 30-40.
Raykov, T. (1997). Scale Reliability, Cronbach's Coefficient Alpha, and Violations of Essential Tau-Equivalence with Fixed Congeneric Components. Multivariate Behavioral Research, 32, 329-353.
請跟姚老師確認
回覆刪除好的,我會再找機會請教姚老師
刪除