手法を比較

	確認的因子分析（CFA）	収束的妥当性
分野	心理測定学	心理測定学
系統	Latent structure	Latent structure
提唱年	1969	1959
提唱者	Karl Gustav Jöreskog	Donald T. Campbell & Donald W. Fiske
種類	Hypothesis-testing latent variable model	Validity evidence / construct validation
原典	Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34(2), 183–202.	Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56(2), 81–105.
別名	CFA, confirmatory FA, measurement model, restricted factor analysis	convergent construct validity, convergence validity, AVE-based convergent validity
関連	4	4
概要	Confirmatory factor analysis tests a researcher-specified factor structure against observed data. Unlike exploratory approaches, the researcher decides in advance which indicators load on which latent factor, and the model is evaluated by how closely the implied covariance matrix reproduces the sample covariance matrix. CFA is central to scale validation, construct validity assessment, and measurement invariance testing.	Convergent validity is the degree to which multiple indicators that are theoretically expected to measure the same construct actually correlate with one another. It is one of the two complementary forms of construct validity identified by Campbell and Fiske (1959) and is now routinely assessed via factor loadings and the Average Variance Extracted (AVE) statistic in SEM-based scale validation.