Crafting Disaster-Driven Statistics: A Strategic Sampling Model

Abstract

This article details the development and implementation of a strategic sampling methodology aimed at enhancing disaster-related statistics in Bangladesh. The study focuses on creating a specialized sampling frame by conducting a comprehensive census of enumeration areas (mouzas) affected by natural disasters. Employing a two-stage random sampling technique, the methodology incorporates stratification at district and disaster-type levels to capture diverse disaster occurrences. The Kish allocation method is utilized for sample allocation, addressing disparities in district sizes. Through meticulous trial and error simulations, the study ensures minimum sample sizes within each domain while employing inverse probability weights to estimate parameters. This strategic approach adopts robust estimations, enriching insights into disaster-related statistics.

International Journal of Statistical Sciences, Vol.24(1), March, 2024, pp 49-64