手法を比較

	ポリトマス探索的因子分析	Graded Response Model (GRM)
分野	心理測定学	心理測定学
系統	Latent structure	Latent structure
提唱年	1978	1969
提唱者	Bengt Muthén	Fumiko Samejima
種類	Latent variable / dimension reduction	Item response theory / polytomous IRT model
原典	Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9(4), 466–491.	Samejima, F. (1969). Estimation of Latent Ability Using a Response Pattern of Graded Scores. Psychometrika Monograph Supplement, No. 17.
別名	EFA for ordered-categorical data, polychoric EFA, ordinal exploratory factor analysis, polytomous factor analysis	Samejima's GRM, Derecelendirilmiş Tepki Modeli (GRM), graded IRT model
関連	4	7
概要	Polytomous exploratory factor analysis extends standard EFA to ordered categorical (Likert-type) response data by replacing the Pearson correlation matrix with a polychoric correlation matrix. It recovers the latent continuous variable that each polytomous item is assumed to reflect, yielding more accurate factor loadings and better-defined factor structures than treating ordinal scores as if they were continuous.	The Graded Response Model is an item response theory model developed by Fumiko Samejima in 1969 for ordered polytomous items such as Likert-type scales. It estimates both the discriminating power of each item and a set of threshold parameters marking the boundaries between adjacent response categories, while simultaneously placing persons on a continuous latent trait scale.