Path analysis

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Path analysis

Hesham Abdalla
Dear all

I have a Path analysis model with nominal and orderinal (Likert) independent
variables. how can i deal deal with these variables in the model?
thanks in advance.
--
Hesham Abdel-Meguid Abdalla
Statistics Dept.
Assiut University

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Path analysis

Garry Gelade
Hesham

It depends on the role these variables play in your model.  For example, a
nominal variable like gender can be coded 0,1 and used as an exogeneous
variable (path going from it but no path going to it). There is nothing
special you need to do in this case, and it can be done in AMOS. A nominal
variable like success or failure which is endogenous (has a path leading to
it) can also be coded 0,1 but you would need a program like Mplus that
handles logistic regression to estimate the path coefficient - I don't know
if AMOS can do this. Mplus also handles multinomial logistic regression so
your endogenous variable can also have more than two catgories.  Another way
nominal variables can be used is to define different groups in a multi-group
analysis, which AMOS and MPlus can do.

Ordinal endogenous variables are handled in Mplus using ordered logistic
regression - in AMOS you might have to treat them as continuous.  Hope this
helps.


Garry Gelade
Business Analytic Ltd

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hesham Abdalla
Sent: 18 October 2008 20:59
To: [hidden email]
Subject: Path analysis

Dear all

I have a Path analysis model with nominal and orderinal (Likert) independent
variables. how can i deal deal with these variables in the model?
thanks in advance.
--
Hesham Abdel-Meguid Abdalla
Statistics Dept.
Assiut University

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

__________ NOD32 3538 (20081020) Information __________

This message was checked by NOD32 antivirus system.
http://www.eset.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