Term: Principal Components Analysis
Last Updated: 2009-08-27
A method of factor analysis in which the total variance in a data set of many variables is analysed. That is, every variable contributes all of its variance (the sum of each score's squared difference from the mean) in an attempt to identify an underlying factor, or latent variable, that is responsible for the values on the observed variables. This can be contrasted with Principal Factor Analysis, in which only a subset of the total variance in a data set of many variables is analysed.