I Order of "Extracted Factors" in SPSS Factor Analysis

[WWGD]

Hi All,
I am doing some FA on SPSS. I entered 9 columns from an Excel file into
SPSS. I am just not sure of how the order of the initial columns corresponds to the order
of the components that have been extracted; specifically, in the "total Variance Explained" section of the outputs. Would someone please please help me figure it out?

 

[FactChecker]

The order of the columns of input data should have no effect on the order of the factors determined by FA. The factors are determined strictly by a statistical calculation that has nothing to do with the order of columns in the input data file. To test that, try changing the order of columns of the input file and run the SPSS program again. The FA process should not change.

 

[WWGD]

Thanks, but, aren't the factors a subset of the input data? EDIT: I mean, don't we keep, out of the input factors, those that explain the most variability ( or whose eigenvalue in the correlation matrix is 1 or greater)?

 

[FactChecker]

I should have been more careful in how I said things. If the order of the input columns are changed, the FA results should remain statistically unchanged, although the FA results may refer to columns of data using the new column positions. That may not be answering your question. I should probably review how SPSS presents its results before I try to say more.

 

[FactChecker]

The listing of the factors in the "Total Variance Explained" table are listed in order of greatest to least. When you select the top one, it should trace back to the same information regardless of the order of columns in the input file. That factor can be traced back to the columns by referring to either the "Component Matrix" table, the "Pattern Matrix" table, or the "Structure Matrix" table, depending on the particular method you are using. Large numbers in the table indicate the columns of input data that make up most of that factor.