Statisticians have been arguing for many years over the nuances of different techniques for examining the interrelatedness of data points. There are many mathematically different strategies, such as Principle Component Analysis (PCA). Others include principal axis factor, maximum likelihood, generalized least squares, and unweighted least squares. An exploration of these is beyond the scope of this text. Just be aware of these names so that when you see them, you can know that it is a species of factor analysis. All of them are designed to result in a *simple structure*. **Simple structure** is statistician speak for a pattern of results where each variable loads highly onto one and only one factor. Simple structures are often achieved through a piece of mathematical wizardry known as **rotation**, and there are several methods of accomplishing this.

The table below represents a typical journal presentation of factor analysis:

Note that the sequentially numbered questions have been reordered so that the heavy loadings on the salient factor are clustered together. This facilitates an easy interpretation of the table. Also, note that the rotation method is provided in the table title. The author has omitted factor loadings below a .400. This is why it appears that each question loads on one and only one factor. In reality, every question will likely have some factor loading value on every factor.

Last Modified: 02/14/2019