Consumer Behavior Research Methods

Factor Analysis

Methodology - PCA

  • The aim of the factor analysis is to identify a reduced number of linear combinations of the original variables that explain most of the variance of the same variables:
     
 
  • Every linear combination is a function of all the original variables, but it is correlated in particular to some of them
     
  • The components or factors are not correlated among each other and their maximal number is equal to the number of original variables (p)

  • The components explain a decreasing percentage of data variance (the first components explain more info than the last ones)

  • The higher the correlation of the original variables, the higher is the power of the factor analysis, and the less information you loose by performing it

Diskussion