It is data reduction tool or structure detection method which helps in removing redundancy or repetition from a set of correlated variables. This analysis tool represents correlated variables with a smaller set of “derived” variables. It helps in forming and arriving at factors that are relatively independent of one another. With the help of this method, we can simply categorize mainly in two types of “variables”: latent variables/factors and the other one is observed variables.
Factor analysis is being applied in many cases – e.g. through this approach, different groups can be identified which permits us to select one variable that is representing many. It also allows us to obtain insights into categories through identification of underlying factors. It also assists us to categorize people or objects depending on their factor score. Different steps involved in exploratory factor analysis includes – collecting and exploring data and choosing relevant variables from it, followed by extracting initial factors and choosing number of factors to retain. This step is being followed by choosing an estimation method, then rotation and interpretation. On the basis of this step, decision is being made to implement few changes in the form of either dropping an item or including an item. Then the last step is to construct scales and use in further analysis.