The DHS Program

Multilevel Modeling Using DHS Surveys: A Framework to Approximate Level-Weights (English) 

Document Type

Methodological Reports

Publication Topic(s)

Survey Methodology, Data Quality, Analysis

Language

English

Recommended Citation

Mahmoud Elkasabi, Ruilin Ren, and Thomas W. Pullum. 2020. Multilevel Modeling Using DHS Surveys: A Framework to Approximate Level-Weights. DHS Methodological Reports No. 27. Rockville, Maryland, USA: ICF.

Download Citation

RIS format / Text format / Endnote format

Publication Date

August 2020

Publication ID

MR27

Abstract:

This report responds to the demand from The Demographic and Health Surveys (DHS) Program data users for sampling weights that correspond to each stage of sampling in DHS surveys. Such weights, described as level-weights, are required for multilevel modeling with data from complex surveys that involve multistage sampling, unequal sampling probabilities, and stratification. Like most survey datasets available for public use, level-weights are not included in DHS data because of concerns about disclosure risk. This report develops and illustrates a framework for estimating or approximating level-weights in DHS surveys. The proposed framework requires data that is included in the publicly available DHS datasets and final reports. The same framework can be used to approximate level-weights for other household surveys, such as the Malaria Indicator Surveys (MIS) and the Multiple Indicator Cluster Surveys (MICS).