手法を比較
選択した手法を並べて確認できます。異なる行はハイライト表示されます。
| 多層確認的因子分析 (MCFA)× | 確認的因子分析（CFA）× | |
|---|---|---|
| 分野 | 心理測定学 | 心理測定学 |
| 系統 | Latent structure | Latent structure |
| 提唱年≠ | 1994 | 1969 |
| 提唱者≠ | Bengt O. Muthen | Karl Gustav Jöreskog |
| 種類≠ | Latent variable model / measurement model | Hypothesis-testing latent variable model |
| 原典≠ | Muthen, B. O. (1994). Multilevel covariance structure analysis. Sociological Methods & Research, 22(3), 376–398. DOI ↗ | Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34(2), 183–202. DOI ↗ |
| 別名 | MCFA, multilevel measurement model, two-level CFA, hierarchical CFA | CFA, confirmatory FA, measurement model, restricted factor analysis |
| 関連≠ | 6 | 4 |
| 概要≠ | Multilevel confirmatory factor analysis tests a pre-specified factor structure while simultaneously accounting for the non-independence of observations caused by clustered data. It decomposes item variance into within-group and between-group components, fitting a separate measurement model at each level, making it the standard tool for validating psychometric scales administered within natural groups such as classrooms, clinics, or organisations. | 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. |
| ScholarGateデータセット ↗ |
|
|