# Section 8.5: EFA Interpretation

Learning Objectives

At the end of this section you should be able to answer the following questions:

•  How do we assign a name or label to factors?
•  How many variables or items can be recommended per factor?

## Pattern Matrix & Structure/Factor Matrix

The next piece of information is the pattern and the structure matrixes. These provide useful information on the factor loadings of the items. The pattern matrix shows the unique contribution of a variable to a factor, and is generally simpler to interpret. The structure matrix shows the shared variance that is ignored in the pattern matrix, and is more complicated to interpret. Note: Jamovi automatically produces the pattern matrix.

PowerPoint: Matrices

Please have a look at the below output for a Pattern Matrix:

## Number of Variables per Factor

The replicability and strength of a component are determined by the number of variables per factor/component. A good rule is that a minimum of 4 variables is recommended per factor. All items that load on to factor should have a score of greater than .40 on the pattern matrix. Just be aware that items can “crossload” (i.e. have scores great than .40 on more than one factor).