search for




 

Sample-size determination for complex sample survey based on statistical power criteria
Journal of the Korean Data & Information Science Society 2021;32:593-604
Published online May 31, 2021;  https://doi.org/10.7465/jkdi.2021.32.3.593
© 2021 Korean Data and Information Science Society.

Inho Park1 · Hyo Joo Lee2

1Department of Statistics, Pukyong National University
2Nakdonggang River Basin Head Office, Korea Water Resources Corporation
Correspondence to: This research was supported by a Research Grant of Pukyong National University (2019) of the first author. Also, this paper is a condensed form of the second author’s master thesis from the Pukyong National University, Busan, Korea.
1 Corresponding author: Associate professor, Department of Statistics, Pukyong National University, Busan 48513, Korea. E-mail: ipark@pknu.ac.kr
2 Researcher, Nakdonggang River Basin Head Office, Korea Water Resources Corporation, Daegu 41914, Korea.
Received March 31, 2021; Revised May 7, 2021; Accepted May 12, 0021.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
The determination of sample size is one of the most important aspects in planning a survey. Several criteria for precision such as (relative) standard error of an estimate are often adopted for deciding the sample size to satisfy the descriptive purpose of the survey in the literature. In this study, we consider power calculation and sample size determination issue when designing a survey the main objective of which is for the analytical purpose. We first carry out a simulation study to show how complex design features can distort the performance of the power when testing a hypothesis on the mean and proportion using the IID-based theory. Furthermore, we propose a method of the sample size determination through a design effect formula for complex sample survey to adjust for the IID-based theory in order to meet a detectable difference requirement of the hypothesis testing.
Keywords : Analytic comparison, complex sample survey, design effect model, power calculation, sample size.