Principal Components and Factor Analysis

Select the Scores tab of the Factor Analysis Results dialog box to access the options described here.

Factor score coefficients. Click the Factor score coefficients button to produce a spreadsheet with the factor score coefficients. For principal components analysis, exact coefficients can be computed from the (rotated or unrotated) factor loadings. For principal factors analysis (with estimated communalities), regression estimates of the factor score coefficients are computed; see Harman (1976, p. 368).

Factor scores. Click the Factor scores button to produce a spreadsheet containing factor scores, based on the factor score coefficients computed as described above. If casewise or pairwise deletion of missing data is selected, then factor scores can only be computed for complete cases (without missing data on any of the variables in the current analysis); if mean substitution is selected, then missing data will be substituted by the respective variable means in the computations (since factor scores are computed from normal z values, missing data will in effect be substituted by zero). The Factor scores button is only available if a raw data file is currently being analyzed.

Save factor scores. Click the Save factor scores button to save the factor scores, which are computed as described above. This button is only available if a raw data file is currently being analyzed.