SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 


#Analytics
#SPSSStatistics
#Analyticstools
 View Only

  • 1.  Factor Analysis in SPSS

    Posted 06/12/24 09:28 AM
    Hi all,

    I have a two-part question:

    1) If I'm conducting a factor analysis can the variables I use have different numbers of points on their scales?  For example, can I use some variables with a seven-point agree/disagree scale, and some with a six-point "completely describes me" scale?  Or do I need to somehow standardize the scales before I do a factor analysis?

    2) If I need to standardize, how do I do that in SPSS?

    Thanks!
    Bob 


  • 2.  RE: Factor Analysis in SPSS

    Posted 06/13/24 09:46 AM
    Edited by Kirill Orlov 06/13/24 09:47 AM

    It is all right with FA based on variables measured on different scales and/or having different variance.

    But you should decide if you are going to base FA on correlations or covariances. You may want to check these threads: https://stats.stackexchange.com/q/62677/3277 ; https://stats.stackexchange.com/q/53/3277.

    Doing the analysis on correlations (a more frequent decision) saves you from thinking about prior linear rescaling both your scales to a common one, such as range to 0-1 scale - because Pearson r is insensitive to linear transforms of data.



    ------------------------------
    Kirill Orlov
    ------------------------------



  • 3.  RE: Factor Analysis in SPSS

    Posted 06/13/24 10:15 AM
    Another option that might sometimes be useful would be to standardize robustly first.  The STATS PREPROCESS extension command has options for that.--



    Original Message


Related Content