Marginal effect: SEM

[John F Hall]

I have had an enquiry about SPSS from Juan Albertorio of Centers for Disease Prevention and Control: can anyone point him in the right direction?

 

1:  Do you have any documentation of how calculate marginal effect on a complex survey with imputed (missing) data in SPSS?

2:  Do you have any documentation on the use of structural equation modelling using survey data?

 

 

John F Hall

[Retired academic survey researcher]

IBM-SPSS Academic Author 9900074

 

Website:          http://surveyresearch.weebly.com/

SPSS course:   http://surveyresearch.weebly.com/1-survey-analysis-workshop-spss.html

Research:        http://surveyresearch.weebly.com/3-subjective-social-indicators-quality-of-life.html

 

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

[Mike]

On Thursday, September 21, 2017 11:48 AM,  John F Hall wrote:
>I have had an enquiry about SPSS from Juan Albertorio of
>Centers for Disease Prevention and Control: can anyone
>point him in the right direction?
>
>1:  Do you have any documentation of how calculate marginal effect
>on a complex survey with imputed (missing) data in SPSS?

I'll leave this to someone else to answer.

>2:  Do you have any documentation on the use of structural
>equation modelling using survey data?

This is a somewhat strange request because structural equation
modeling (SEM) has been used to analyze survey data for
decades and a search of either scholar.google.com or
Jstor.org (many statistical journals here) should be done to
determine which research is relevant to the specific issues
being dealt with.  An important question is what software
will be used to do the SEM analyses (different programs
have different strengths).  The program Mplus have been
used for complex surveys and one entry point to Mplus
is the following link:
https://www.statmodel.com/features4.shtml

But one has to ask what kind of analysis does one want to
do becuase it possible that an SEM analysis could be done
with readily available software like SAS or Stata. AMOS is
an add-on to SPSS and one's institution may not have bought
this option, requiring one to buy or get access to an some other
SEM program, like Mplus, which has a large group of users
who are active in developing the software.  If one is going to
invest time and effort in learning an SEM program, I'd suggest
Mplus unless one has to use SAS or Stata or some other
program that is readily available.

There is also a SEM mailing list that one might want to join
after one has gotten some basic knowledge about SEM; see:
https://listserv.ua.edu/cgi-bin/wa?A0=semnet


-Mike Palij
New York Uiversity
[hidden email]

 

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

[John F Hall]

Mike
Thanks for this: hope it finds what he needs.  Juan copied in to this reply.
John

John F Hall
[Retired academic survey researcher]
IBM-SPSS Academic Author 9900074

Website:          http://surveyresearch.weebly.com/ 
SPSS course:
http://surveyresearch.weebly.com/1-survey-analysis-workshop-spss.html 
Research:
http://surveyresearch.weebly.com/3-subjective-social-indicators-quality-of-l
ife.html


-----Original Message-----
From: Mike Palij [mailto:[hidden email]]
Sent: 21 September 2017 19:35
To: John F Hall <[hidden email]>; [hidden email]
Cc: Michael Palij <[hidden email]>
Subject: Re: Marginal effect: SEM

On Thursday, September 21, 2017 11:48 AM,  John F Hall wrote:
>I have had an enquiry about SPSS from Juan Albertorio of Centers for
>Disease Prevention and Control: can anyone point him in the right
>direction?
>
>1:  Do you have any documentation of how calculate marginal effect on a
>complex survey with imputed (missing) data in SPSS?

I'll leave this to someone else to answer.

>2:  Do you have any documentation on the use of structural equation
>modelling using survey data?

This is a somewhat strange request because structural equation modeling
(SEM) has been used to analyze survey data for decades and a search of
either scholar.google.com or Jstor.org (many statistical journals here)
should be done to determine which research is relevant to the specific
issues being dealt with.  An important question is what software will be
used to do the SEM analyses (different programs have different strengths).
The program Mplus have been used for complex surveys and one entry point to
Mplus is the following link:
https://www.statmodel.com/features4.shtml

But one has to ask what kind of analysis does one want to do becuase it
possible that an SEM analysis could be done with readily available software
like SAS or Stata. AMOS is an add-on to SPSS and one's institution may not
have bought this option, requiring one to buy or get access to an some other
SEM program, like Mplus, which has a large group of users who are active in
developing the software.  If one is going to invest time and effort in
learning an SEM program, I'd suggest Mplus unless one has to use SAS or
Stata or some other program that is readily available.

There is also a SEM mailing list that one might want to join after one has
gotten some basic knowledge about SEM; see:
https://listserv.ua.edu/cgi-bin/wa?A0=semnet


-Mike Palij
New York Uiversity
[hidden email]

 

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