|
Activity Number:
|
145
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #307225 |
|
Title:
|
Hierarchical Linear Modeling Using Complex Survey Data Based on the Pre-Elementary Education Longitudinal Study (PEELS)
|
|
Author(s):
|
Hyunshik Lee*+ and Frank Jenkins and Sunyeong Heo
|
|
Companies:
|
Westat and Westat and Changwon National University
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20854,
|
|
Keywords:
|
multi-level modeling ; HLM 6.0 ; growth curve analysis ; simulation study
|
|
Abstract:
|
PEELS studies children with disabilities using a multistage survey design. Children are followed for 5 waves at which numerous background and cognitive measures are gathered. The goal is to identify correlates of cognitive development for the population. The study will gauge development by estimating cognitive growth curves for individual children and the average growth for key subgroups. The HLM 6 program will be used for the analysis. It is not known how well the program estimates growth and variance parameters for data with complex sampling, missing data points, and imputed values for missing variables. This study evaluates the performance of HLM by analyzing simulated data with sample and missing data characteristics varying over a plausible range while on average similar to the actual PEELS data. Bias and precision of estimates as functions of sample factors will be determined.
|
