Python Adaptive Conjoint Module

[zstatman]

Anyone of you Python gurus have or are intrigued into developing an ADC
module?

Hoping,
WMB
Statistical Services

=========================================
mailto:[hidden email]
http://home.earthlink.net/~info.statman
=========================================

=====================
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

[Pearmain, Michael]

Hi All,

Merry xmas to everyone, I was recently was given a report which contains
factor scores and cluster centres for a study,

I've re-run the survey using the same questions that the original study
used, does anyone know of a way I can use the Factor scores in the first
study, and apply them onto the second wave?

Mike

=====================
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

[Art Kendall]

These are a few of the possibilities.

If you can get the data from the original study you can re-run it using
all of the original data but have it compute factor scores for the full
set of old and new cases.

Ask the author of the original study for the coefficients (s)he used.

If this was a conventional scale creation that used unit weights, all
items retained go on only one scale.  Simply reflect those items that
need it, and then sum the items that are assigned to a scale.

Try a confirmatory factor analysis putting items that go together on a
proposed factor.

Art Kendall
Social Research Consultants

Pearmain, Michael wrote:

> Hi All,
>
> Merry xmas to everyone, I was recently was given a report which contains
> factor scores and cluster centres for a study,
>
> I've re-run the survey using the same questions that the original study
> used, does anyone know of a way I can use the Factor scores in the first
> study, and apply them onto the second wave?
>
> Mike
>
> =====================
> 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

[Hector Maletta]

         If you have the output of the factor analysis that gave origin to
the factor scores in the first study, that output included a matrix of
factor score coefficients. These coefficients define a factor score as a
linear function of observed variables, and can be used to compute factor
scores based on the same analysis but on another data set. However, if you
run a fresh factor analysis in the second data set, the resulting factor
scores are likely to be different, though they might be correlated with the
ones coming from the first dataset.
         If you do not have the factor analysis output from the first study,
just the factor scores of the cases, I'm afraid there is little you can do
to apply "the same factor scores" (i.e. compute factor scores based on the
same coefficients) in the second study. You need the coefficients to do
that.
         Hector

         -----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Pearmain, Michael
Sent: 20 December 2007 19:45
To: [hidden email]
Subject: Applying factor scores to new data.

         Hi All,

         Merry xmas to everyone, I was recently was given a report which
contains
         factor scores and cluster centres for a study,

         I've re-run the survey using the same questions that the original
study
         used, does anyone know of a way I can use the Factor scores in the
first
         study, and apply them onto the second wave?

         Mike

         =====================
         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