A Long Walk of Survey Sampling with Modern Methodologies

Abstract

Survey sampling emerged as a crucial methodology to overcome the impracticality and high cost of full censuses. This article aims to synthesise the historical trajectory of survey sampling and review the rationale and application of modern techniques. A narrative review methodology was employed this study to trace the evolution from early probability-based methods, which prioritised representativeness, to contemporary non-probability methods that address challenges in big data and hard-to-reach populations. Key findings highlight that while probability sampling remains the gold standard for unbiased inference, modern methods like network, encounter, and composite sampling offer practical, cost-effective alternatives, though they require advanced techniques for bias correction. The study provides a comparative framework to guide researchers in selecting context appropriate sampling strategies across fields such as agriculture, ecology, and social sciences.