Design-Based versus Model-Based Methods: a Comparative Study Using Longitudinal Survey Data

• Activity Number: 76
• Type: Contributed
• Date/Time: Sunday, August 6, 2006 : 8:00 PM to 9:50 PM
• Sponsor: General Methodology
• Author(s): Sunita Ghosh*+ and Punam Pahwa and Geert Molenberghs
• Companies: University of Saskatchewan and University of Saskatchewan and Limburgs Universitair Centrum
• Keywords: survey methods ; GEE ; NPHS ; marginal model ; multi-stage sampling

Abstract

Survey data analysis using complex sampling designs ought to account for clustering, stratification and unequal probability of selection. Design-based and model-based methods are two commonly used routes taken to account for such survey designs. Several studies of cross-sectional survey designs have shown that these two approaches provide similar results when the model fits the data well. The present paper aims at comparing these two approaches using the National Population Health Survey (NPHS) dataset. The NPHS is an ongoing longitudinal study, and a stratified multi-stage sampling design was used. A marginal modeling approach proposed by Rao (1998) will be used by way of a design-based method. The Generalized Estimating Equation (GEE) method, proposed by Liang and Zeger (1986), will be used as a typical model-based approach. Results obtained from these methods will be compared.