Longitudinal data analysis

[whatsinaname]

I am looking to analyze data with a time span of 5 years (2005 - 2010).  The data is from the same individuals over this time period, and is thus a within subjects repeated measure.  The data is also nested by year.  My outcome variable is a measure of performance, which is ordinal (high performer, average performer, low performer).  I have several independent variables that I want to test if they predict the outcome variable of performance.

What is the appropriate statistical model for this data?  My candidates are 1) generalized estimating equations (GEE) using ordinal logistic regression, 2) structural equation modeling, and 3) a lagged analysis that explained performance ratings as a function of the independent variables and performance ratings the previous year.  Any preferences, or model that someone found more valid / useful?  Bonus points if you can point me to a research article that uses any of these models similar to the way I intend to.

Thanks in advance for any assistance you would be able to provide!

[Adel Daoud]

 

 

Hi

 

Sounds as if you will need multilevel method to accomplish what you want to do.

 

Check out the following book as a reference

http://www.amazon.com/Applied-Longitudinal-Data-Analysis-Occurrence/dp/0195152964

 

Or,

http://www.wiley.com/WileyCDA/WileyTitle/productCd-EHEP002272.html

 

 

If you want to include SEM, then you will need multilevel SEM. See

http://www.amazon.com/Introduction-Multilevel-Techniques-Quantitative-Methodology/dp/1841697567

 

 

 

 

hth,

Adel

On Thu, May 9, 2013 at 1:11 PM, whatsinaname <[hidden email]> wrote:

I am looking to analyze data with a time span of 5 years (2005 - 2010).  The
data is from the same individuals over this time period, and is thus a
within subjects repeated measure.  The data is also nested by year.  My
outcome variable is a measure of performance, which is ordinal (high
performer, average performer, low performer).  I have several independent
variables that I want to test if they predict the outcome variable of
performance.

What is the appropriate statistical model for this data?  My candidates are
1) generalized estimating equations (GEE), 2) structural equation modeling,
and 3) a lagged analysis that explained performance ratings as a function of
the independent variables and performance ratings the previous year.  Any
preferences, or model that someone more valid / useful?  Bonus points if you
can point me to a research article that uses any of these models similar to
the way I intend to.

Thanks in advance for any assistance you would be able to provide!



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Longitudinal-data-analysis-tp5720129.html
Sent from the SPSSX Discussion mailing list archive at Nabble.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

--
Adel Daoud, PhD
Department of Sociology and Work Science,  University of Gothenburg
Box 720
405 30 Göteborg, SWEDEN
Visiting address: Sprängkullsgatan 25, room K109
[phone]
[hidden email]

[David Greenberg]

There are many options. Look at econometrics textbooks that have a chapter on panel data; and at Todd Little's new book on longitudinal data analysis using structural equation modeling. David Greenberg, Sociology Department, NYU

On Thu, May 9, 2013 at 7:11 AM, whatsinaname <[hidden email]> wrote:

I am looking to analyze data with a time span of 5 years (2005 - 2010).  The
data is from the same individuals over this time period, and is thus a
within subjects repeated measure.  The data is also nested by year.  My
outcome variable is a measure of performance, which is ordinal (high
performer, average performer, low performer).  I have several independent
variables that I want to test if they predict the outcome variable of
performance.

What is the appropriate statistical model for this data?  My candidates are
1) generalized estimating equations (GEE), 2) structural equation modeling,
and 3) a lagged analysis that explained performance ratings as a function of
the independent variables and performance ratings the previous year.  Any
preferences, or model that someone more valid / useful?  Bonus points if you
can point me to a research article that uses any of these models similar to
the way I intend to.

Thanks in advance for any assistance you would be able to provide!



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Longitudinal-data-analysis-tp5720129.html
Sent from the SPSSX Discussion mailing list archive at Nabble.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