An Efficient Sampling Approach for Village Elections: Quick Count Using Stratified Systematic Cluster Sampling

Abstract

Quick counts are widely used to estimate election outcomes before official results are announced. However, their accuracy depends on the quality of the sampling method used. This study applies the Stratified Systematic Cluster Random Sampling method in the 2019 village head election in Panembangan Village, Indonesia, to provide an efficient and statistically sound quick count process suitable for rural contexts with limited resources. The method integrates stratification (dividing voters by hamlet), clustering (grouping by polling stations), and systematic sampling (selecting polling stations at fixed intervals). Out of 10 polling stations, 5 were systematically selected after stratification. All valid votes from these polling stations were used for estimation. The results show that candidate Untung Sanyoto received 59.16% of the vote, while his opponent received 40.84%. The margin of error was ±0.69% at a 95% confidence level, and the estimates closely matched the official election results, demonstrating the method’s reliability. This study shows that combining statistical rigor with practical sampling strategies can produce fast, accurate, and cost-effective results. It provides a scalable model for local elections, especially in regions facing geographic or logistical challenges, and contributes to the development of transparent and trustworthy democratic practices.