Before you start
Before proceeding, please read the PowerPoint on Factor Analysis.

A personality study
Ten participants are given a battery of personality tests, comprising the following items: Adventure; Agoraphobia; Anxiety; Arachnophobia; Extraversion; Sociability. The purpose of this project is to ascertain whether the correlations among the six variables can be accounted for in terms of comparatively few latent variables, or factors.
Preparing the data set
The data are shown in Table 1. Ensure that the values in the Decimals column are 0. Click the Data View tab to open Data View

Procedure for the factor analysis of the raw data
Follow the procedure we saw earlier, requesting the Univariate descriptives, Initial solution, Coefficients, Reproduced, Scree plot, and Varimax options.

Interpretation of the results
After a table of descriptive statistics, there is a table labeled Correlation Matrix. Is there any evident pattern that would suggest that the matrix might be accounted for in terms of relatively few factors (components)? Examine the remainder of the output in the manner outlined in the PowerPoint, considering the scree plot, the table labeled Component (factor) Matrix listing the unrotated loadings for each factor, the residuals and the final rotated matrix in the table labeled Rotated Component Matrix.
How might the patterns among the correlations in the matrix be explained psychologically? Look at the table Rotated Component Matrix and make a list of the loadings that are greater than about 0.5 on each factor (component).

revised 09-15-2008