Assessing Construct Validity: The Utility of Factor Analysis

應用因素分析法以檢驗建構效度

呂正雄(Cheng-Hsiung Lu)

《測驗統計年刊》 15期_上 (2007/06) Pp. 79-94

https://doi.org/10.6773/JRMS.200706.0079

摘要

根據一九五四年美國心理學會所研擬之效度標準，總共有四種效度分類，可供不同研究之本質而採用，其種類包括內容效度、預測效度、同時效度、以及建構效度。為檢驗建構效度，本研究使用探測性(exploratory)因素分析法及驗證性(confirmatory)因素分析法，來探討實證資料之因素序列。經由因素分析法之操作，重要的分析結果包括因素負荷量、轉軸後之簡明結構、特徵值陡坡圖示、因素模式圖，以及模式適配指數表。由這些因素分析法所得到之資料，本研究獲得建構效度之實證檢驗。

關鍵字

建構效度 ； 探測性因素分析法 ； 驗證性因素分析法

並列摘要

According to validity standards of American Psychological Association in 1954, four types of validity were identified for accomplishing different aims of research, namely content validity, predictive validity, concurrent validity, and construct validity. For the justification of the construct validity, this study utilizes exploratory factor analysis and confirmatory factor analysis to examine factors of the empirical data. Through the operation of the empirical data analysis, important results are accordingly illustrated by the exploratory factor analysis such as factor loading, the rotated simple structure, the plot of scree test; also by the confirmatory factor analysis such as specifying a priori factor pattern, and indices for goodness of model fit. Consequently, the obtained solutions from the factor analysis procedures provide reasonable justification to construct validity of the researched.

並列關鍵字

construct validity ； exploratory factor analysis ； confirmatory factor analysis

參考文獻

Anderson, J. C.,Gerbing, D. W.(1988).Structural equation modeling in practice: A review and recommended two step approach.Psychological Bulletin.103(3),411-423.

Bartz, A. E.(1999).Basic statistical concepts.Upper Saddle River, NJ:Merrill.

Bentler, P. M.(1990).Comparative fit indexes structural models.Psychological Bulletin.107(2),238-246.

Bollen, K. A.(1990).Overall fit in covariance structure models: Two types of sample size effects.Psychological Bulletin.107(2),256-259.

Courville, T.,Thompson, B.(2001).Use of structure coefficients in published multiple regression articles: Beta is not enough.Educational and Psychological Measurement.61,229-248.