Also known as principal component analysis, designed to reduce a large set of variables into small artificial variables known as the principal components, which account for all the other variables.
- From the main menu- click on Analyze- choose Data Reduction –Factor.
- The factor analysis dialogue box opens
- Drag all the variables you want included in the analysis into the variables box
- Click on the Descriptive button
- The factor analysis: descriptive dialogue box
- In addition to the options that have been checked by default, also check Coefficients, KMO and Bartlett’s test of Sphericity, Reproduced and Anti-image from the Correlation Matrix area
- Click the Continue button to return to the factor analysis dialogue box
- Click the Extraction button and the Factor Analysis: Extraction dialogue box
- Keep all the defaults and check the Scree plot in the Display area.
- Click the continue button.
- Click the Rotation button, in the dialogue box select the Direct Varimax option in the Method area.
-The other common alternative is the Oblimin option.
- This activates the Rotate Solution option in the Display area and is always checked by default if not make sure to check it.
- Also select Loading plot(s) in the Display area.
- Click the continue button to go back to the Factor analysis dialogue box
- Click the scores button, check the Save as Variables option, and keep the Regression option selected. Click continue
- Click the Options button. Check the Sort by Size and Suppress Small Coefficients Option. Change Absolute values below from”10” to “3”.
- Click continue
- Click the Ok button to generate the output.
The output from the above analysis is very extensive but normally you will need to run the analysis multiple times to arrive at a final solution.
You need to focus on the Initial Eigenvalues to generate an initial sense of the major components and to measure how much of the total variation each component explains.