Abstract

CONDORI HUANCA, Ronal Edwin. An approach to the optimal sampling design. Revista Varianza [online]. 2022, n.19, pp. 17-26. ISSN 9876-6789.

Abstract In the absence of the development and quantification of a greater diversity of precision measures for proven designs in the Bolivian context. This document addresses the development of the experimental objective of comparability and choice of different potential sampling designs to be applied to quick counts for electoral purposes, for the same information from the 2020 general elections in Bolivia was used, selecting up to eighteen possible proven scenarios, and evaluating their standard errors, coefficients of logarithmic variation, and other measures. Considering the ratio estimator, it was found that the design with stratification based on enclosure size with a sequential type selection is the most optimal design with a logarithmic coefficient of 0.042, standard errors close to 0.018, and other more accurate measures.

Keywords : Quick Count; Logarithmic coefficient of variation; Optimal sampling design; Elections.