There is no "forced" n-factor solution. There are always k factors
for k observed variables, no matter you like it or not. You have only the
choice of extracting one, two or as many as you want, from one to k factors
or components, thus explaining less or more variance with the extracted
components.
You do not have either the choice of extracting any "minor" factor
before you extract the "major" factors that go before it: the procedure
starts by extracting first the component corresponding to the highest
eigenvalue, explaining the most variance, then the second one (explaining
the maximum amount within the remaining variance) and so on.
Hector
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Eugene
Sent: 13 November 2007 15:25
To:
[hidden email]
Subject: Forced one-factor solution.
Hi All,
I'm using SPSS 15.0 for Windows. Is it possible to run a forced
one-factor
solution with this program?
I've tried using principal components analysis (Analyze -> Data
Reduction ->
Factor -> Descriptives) under the impression that choosing 1 as the
number
of factors gave a forced one-factor solution. Instead, it seems to
just
give me the first principal component.
Any help is much appreciated.
Thank you,
Eugene
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text
except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
