The announcement below is posted on behalf of Dr. Jenn-Yun Tein
([email protected]). 

  

 

The Arizona State University Program for Prevention Research is hiring a
postdoctoral Research Associate for NIMH funded projects to evaluate a
long-term intervention for bereaved children and apply advanced
methodologies in prevention research, particularly as applied to studies
of program effectiveness and dissemination.  This position will assist
faculty with the application of advanced methodologies in prevention
research, methodological development and the writing of manuscripts, and
conduct statistical analyses with structural equation modeling,
longitudinal modeling techniques and multilevel modeling.  Required
Qualifications:  PhD in Psychology, Sociology, Education or other social
science discipline.  Experience with advanced statistical analyses,
including structural equation modeling, longitudinal modeling techniques
and multilevel modeling.  Desired Qualifications:  Experience with
advanced statistical applications.  Experience supervising others.
Application procedure: fax or mail letter of interest, curriculum vitae,
and three letters of reference to Dr. Jenn-Yun Tein, Program for
Prevention Research, Arizona State University, P. O. Box 876005, Tempe,
AZ 85287-6005.  Application Deadline: November 30, 2005, or if not
filled, every two weeks thereafter until search closed.  This is a 2
year position and includes allotted time to take advanced statistical
seminars.  Salary range: $45,000-$50,000.  Background check will be
conducted prior to employment.  AA/EOE.

 

 

------------------------------------------------

Scott R. Weaver, Ph.D.

Postdoctoral Research Fellow

Program for Prevention Research

Arizona State University

P.O. Box 876005

Tempe, AZ 85287-6005

(480) 727-6155 (O)

(480) 965-5430 (F)

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
http://lists.utsouthwestern.edu/pipermail/impute/attachments/20051109/5416a363/attachment.htm
From Craig.Enders <@t> asu.edu  Fri Nov 11 19:34:43 2005
From: Craig.Enders <@t> asu.edu (Craig Enders)
Date: Fri Nov 11 19:34:49 2005
Subject: [Impute] MI with multiple cohort data
Message-ID: <[email protected]>

I'm curious if anyone has used MI to handle missing data by design in a 
longitudinal study with multiple age cohorts - the ultimate goal is to perform 
a growth curve analysis, e.g., using PROC MIXED.? Because certain individuals 
were never assessed at particular ages, it seems that both age (the "time 
score") and the outcome is missing.? Some of the outcome scores are missing by 
design (e.g., someone in the 15 year old cohort was never assessed at age 13), 
and some of the outcome scores are missing for other reasons (e.g., attrition 
occurred within each cohort).? 

It seems that there are at least three different options:

1) Impute the time scores (i.e., the ages that the subjects were never assessed 
at due to the missing by design data collection strategy) and the missing 
outcome values.? This solution makes me very uneasy.

2) Fill in the missing time scores with age values that could have 
hypothetically been observed, had the data collection schedule used a single 
cohort, then impute the missing outcome scores - this seems like a better 
solution, and would result in a fully balanced data set.

3) Use MI to fill in missing outcome scores separately within each age cohort - 
this doesn't seem to provide any advantage over ML estimation using PROC MIXED, 
sans the imputation phase.

Obviously, the growth curve model could be estimated using PROC MIXED, but it 
would be useful to incorporate auxiliary variables that might be related to 
attrition.? The auxiliary variable issue is straightforward if ML estimation is 
implemented using SEM software (e.g., Graham, 2003), but I'm curious about the 
application of MI in this situation.

I would appreciate hearing different thoughts on this,

Craig Enders

*****************************************************************************

Dr. Craig Enders
Assistant Professor
Quantitative Psychology Concentration
Department of Psychology
Arizona State University
Box 871104
Tempe, AZ 85287-1104
Office?(480) 727-0739
[email protected]
http://www.asu.edu/clas/psych/people/faculty/cenders.htm

*****************************************************************************



*****************************************************************************

Dr. Craig Enders
Assistant Professor
Quantitative Psychology Concentration
Department of Psychology
Arizona State University
Box 871104
Tempe, AZ 85287-1104
Office?(480) 727-0739
[email protected]
http://www.asu.edu/clas/psych/people/faculty/cenders.htm

*****************************************************************************

Reply via email to