Key concepts:
- Observed variable: They are variables that is directly observed or measured in a study.
- Factors: Factors are deduced from the interconnections of observed variables rather than being directly observed or measured. Other names include constructs, latent variables, or latent construct.
- Factor loadings or loadings: Factor loadings are numerical values that represent the magnitude and direction of the relationship between each observed variable and each factor. The presence of high factor loadings indicates a substantial association between an observed variable and a factor.
- Communalities: Communalities are numerical values that represent the proportion of variance in each variable that is accounted for by the underlying factors.
- Scree plot: It is a graphical representation of eigenvalues.
- Eigenvalues: Eigenvalues provide a measure of the proportion of variance accounted for by each factor. Researchers frequently employ eigenvalues and the scree plot to determine the optimal number of factors to retain in analysis.
- Rotation: In exploratory factor analysis (EFA), it is common practice to employ factor rotation techniques in order to enhance the interpretability of the extracted factors. Examples of rotation commonly used in EFA include varimax, oblimin, promax etc.
- Model fit: Model fit measures the degree to which the hypothesized factor model aligns with the observed data.