Factor analysis - 4 separate surveys scenarios

[Orlando Villella]

I am working with a doctoral student who has created a survey insturment
administered. This survey is completed 4 times by each of her participants.
Each time, the participant is to respond to a different situational scenario
so the insturment intended to measure their reaction to each individual
scenario. Would it be appropriate to do 4 separate factor analyses to
analyze factor loading on the survey instrument with each scenario? Would it
make sense to do a single factor analysis using the variables form all 4
scenarios?
Thanks for your help.
Orlando

=====================
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]

It depends on the theory being explored or tested. If the scenarios are
supposed to operate upon the same underlying psychological traits, one
single factor analysis may suffice, using dummies for the scenarios (three
of them, one being the default). If the scenarios would cause the
fundamental traits or underlying factors to change, then run different
factor analysis for each. If you're not sure, do both and study the results
closely until you understand better what's going on.

Hector



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Orlando Villella
Sent: 08 August 2008 13:28
To: [hidden email]
Subject: Factor analysis - 4 separate surveys scenarios

I am working with a doctoral student who has created a survey insturment
administered. This survey is completed 4 times by each of her participants.
Each time, the participant is to respond to a different situational scenario
so the insturment intended to measure their reaction to each individual
scenario. Would it be appropriate to do 4 separate factor analyses to
analyze factor loading on the survey instrument with each scenario? Would it
make sense to do a single factor analysis using the variables form all 4
scenarios?
Thanks for your help.
Orlando

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

[Dale Glaser]

Hector, do you think conducing a dynamic factor analysis might be plausible?...there are some classic articles in Psychometrika (I think as early as the 50s) that address growth models and EFA............assuming for now there is no a priori factorial structure to the instrument, there might be a way to set a dynamic EFA structure in Mplus.................dale

Hector Maletta <[hidden email]> wrote:  It depends on the theory being explored or tested. If the scenarios are
supposed to operate upon the same underlying psychological traits, one
single factor analysis may suffice, using dummies for the scenarios (three
of them, one being the default). If the scenarios would cause the
fundamental traits or underlying factors to change, then run different
factor analysis for each. If you're not sure, do both and study the results
closely until you understand better what's going on.

Hector



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Orlando Villella
Sent: 08 August 2008 13:28
To: [hidden email]
Subject: Factor analysis - 4 separate surveys scenarios

I am working with a doctoral student who has created a survey insturment
administered. This survey is completed 4 times by each of her participants.
Each time, the participant is to respond to a different situational scenario
so the insturment intended to measure their reaction to each individual
scenario. Would it be appropriate to do 4 separate factor analyses to
analyze factor loading on the survey instrument with each scenario? Would it
make sense to do a single factor analysis using the variables form all 4
scenarios?
Thanks for your help.
Orlando

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



Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

=====================
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]

This is not a dynamic analysis, but a comparative static one: four scenarios
to be compared.

Hector



  _____

From: Dale Glaser [mailto:[hidden email]]
Sent: 08 August 2008 16:18
To: Hector Maletta; [hidden email]
Subject: Re: Factor analysis - 4 separate surveys scenarios



Hector, do you think conducing a dynamic factor analysis might be
plausible?...there are some classic articles in Psychometrika (I think as
early as the 50s) that address growth models and EFA............assuming for
now there is no a priori factorial structure to the instrument, there might
be a way to set a dynamic EFA structure in Mplus.................dale

Hector Maletta <[hidden email]> wrote:

It depends on the theory being explored or tested. If the scenarios are
supposed to operate upon the same underlying psychological traits, one
single factor analysis may suffice, using dummies for the scenarios (three
of them, one being the default). If the scenarios would cause the
fundamental traits or underlying factors to change, then run different
factor analysis for each. If you're not sure, do both and study the results
closely until you understand better what's going on.

Hector



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Orlando Villella
Sent: 08 August 2008 13:28
To: [hidden email]
Subject: Factor analysis - 4 separate surveys scenarios

I am working with a doctoral student who has created a survey insturment
administered. This survey is completed 4 times by each of her participants.
Each time, the participant is to respond to a different situational scenario
so the insturment intended to measure their reaction to each individual
scenario. Would it be appropriate to do 4 separate factor analyses to
analyze factor loading on the survey instrument with each scenario? Would it
make sense to do a single factor analysis using the variables form all 4
scenarios?
Thanks for your help.
Orlando

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




Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

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