• Can you run a rank-order segmentation in SPSS? (SImilar to a Fasclus in

    From Erica Z@21:1/5 to All on Fri Feb 14 13:40:33 2020
    Hi there,

    My team is currently split in tool usage - half of us use SPSS and half of us use SAS. There are a few processes we'd like to be able to do in SPSS. Does anyone know of an easy way to do the following?

    - Factor Analysis (similar to a Varclus in SAS): we run factor analysis but are looking for something similar to the Varclus function that prevents items from loading onto multiple factors.

    - Rank Order Segmentation (similar to a Fasclus in SAS): we run Cluster analysis on SPSS but often end up with a segment that's high on every input and one that is low on every input. We've been doing Fasclus in SAS but is there a way to do this in SPSS?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Rich Ulrich@21:1/5 to erica.zimmerman814@gmail.com on Sat Feb 15 01:40:20 2020
    On Fri, 14 Feb 2020 13:40:33 -0800 (PST), Erica Z <erica.zimmerman814@gmail.com> wrote:

    Hi there,

    My team is currently split in tool usage - half of us use SPSS and half of us use SAS. There are a few processes we'd like to be able to do in SPSS. Does anyone know of an easy way to do the following?

    - Factor Analysis (similar to a Varclus in SAS): we run factor analysis but are looking for something similar to the Varclus function that prevents items from loading onto multiple factors.

    - Rank Order Segmentation (similar to a Fasclus in SAS): we run Cluster analysis on SPSS but often end up with a segment that's high on every input and one that is low on every input. We've been doing Fasclus in SAS but is there a way to do this in SPSS?


    [cribbed from an older message by Bruce]

    The SPSS mailing list (http://spssx-discussion.1045642.n5.nabble.com/)
    is a lot more active than this forum these days. You might want to
    join that list (if not already a member) and post your question there.
    Via the page given above, click on -more options- near the top for
    info on how to subscribe. HTH.

    My own, random comments -
    I have done a lot of factor analyses of correlation matrices,
    almost always for items of rating scales. My advice is not so
    relevant if your best correlations are outside of the 0.35 - 0.65
    range.

    After a period of experimenting, I settled on looking (rarely)
    at PC; and always at FA (communalities on the diagonal),
    usually with Varimax rotation. I got used to using those results,
    and they served me well.

    I just now looked up Varclus - it looks to me like it would
    take a while to get used to using those results. Do you
    have reference to someone's impressive results, from using
    that? (Or Fasclus?)

    --
    Rich Ulrich

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)