TY - JOUR
AU - Mulaik, Stanley A.
AB - Multir~ut.iure Bchu\'ior-a1 Rrsear.ch, 25 (I), 53-59 Copyright 0 1990. Lawrence Erlbaum Associates, Inc. Blurring the Distinctions Between Component Analysis and Common Factor Analysis Stanley A. Mulaik Georgia Institute of Technology The Differences Bemleen Common Factor Analysis and Component Analysis I feel there are crucial differences between common factor analysis and component analysis overlooked by Velicer and Jackson (1990). I also disagree with their suggestion that something may be wrong with the latent variable concept implicit in common factor analysis and structural equation modeling, that component analysis models are to be preferred because they are determinate whereas common factor models based on latent variables are flawed because they are indeterminate. The aim of common factor analysis is not simply to represent the observed variables in terms of a smaller number of variables but rather to represent the observed variables as functions of other, latent variables, some of which, the common factors, are solely responsible for the covariation among the observed variables, and the remainder, the unique factors, are responsible for variation unique to each respective observed variable. Potentially a common factor variable can have theoretical applications independent of any particular set of observed variables by being, say, the common cause 
TI - Blurring the Distinctions Between Component Analysis and Common Factor Analysis
JF - Multivariate Behavioral Research
DO - 10.1207/s15327906mbr2501_6
DA - 1990-01-01
UR - https://www.deepdyve.com/lp/taylor-francis/blurring-the-distinctions-between-component-analysis-and-common-factor-0XEFTW0Q1h
SP - 53
EP - 59
VL - 25
IS - 1
DP - DeepDyve
ER -