Correlation
Principal component analysis
Decomposes a set of correlated variables into fewer components, from their correlation matrix.
Variables
- VariablesContinuous · Ordinal — at least 3
Options
- Components retained (0 = all)
What it reports
Eigenvalue, share of variance explained and loadings for each component; KMO and Bartlett test.
scipy · Open the accuracy report
When does this apply?
For decomposing at least three interrelated quantitative variables into fewer components; the decomposition runs on their correlation matrix. No hypothesis is tested: the output is the eigenvalues, the proportions of variance explained, the loadings, KMO and Bartlett’s test of sphericity. How many components to retain is the researcher’s decision — all are shown by default and the software imposes no cutoff such as “eigenvalue > 1”. With two variables there is nothing to decompose: Correlation analysis covers that case.