In an event that there are several variables in a particular research design, it can be very advantageous to try and reduce the variables to a smaller set of factors. This technique has no dependent variable making it an independent technique. The research analyst is therefore looking for the basic structure of the data matrix.
These independent variables are always normal and continuous having three to five variables which are loading into a factor. There should be at least five observations per variable leading to over 50 observations. Generally what is preferred between the variables is multicollinearity, this because correlations are very key to towards data reduction. MSA measures the degree of the predictability of every variable. There are two major factor analysis methods. This first one is referred to as common factor analysis and the second referred to as principal component. They perform different functions. (William C. Black, 1998).