An Introduction to Factor Analysis

Let’s take a look at a few different methods that can be used to extract factors from a data set:

• Principal Component Analysis: Principle component analysis, or PCA, is the most common method of factor analysis used by researchers. PCA involves extracting the maximum variance between the variables and then putting it into the first factor. Then, the variance explained by the first factor is removed after which the maximum variance for the second factor is extracted. This process is repeated until the last factor.  

• Maximum Likelihood Method (ML): In the maximum likelihood method, researchers can test the statistical significance of factor loadings and calculate the correlations among different factors so that the confidence intervals for these parameters can be computed. ML is best suited when working with data that is normally distributed because it facilitates the computation of a wide range of indexes. 

• Common Factor Analysis: After PCA, common factor analysis is the most preferred method used by researchers. It involves extracting the fewest factors that can account for the common variance within a set of variables.