• The Korean Journal of Applied Statistics
• /
• v.27 no.4
• /
• pp.513-521
• /
• 2014

If the target error of an estimator at the present time is greater than the coefficient of variation(CV) of the estimator at the previous time, sample size at this point should be decreased. Various papers have researched sample size determination methods using the CV of an estimator at the previous time, variation of population size and target error of the estimator at this time in sampling on successive occasions. We research a new sample size determination method additionally using change of population CV. We compare the proposed method with existing ones in various simulation settings.

• Communications for Statistical Applications and Methods
• /
• v.17 no.6
• /
• pp.791-801
• /
• 2010

This study reviews the concept of disclosure, disclosure risks, and uniqueness. The number of uniqueness in the population is of great importance in evaluating the disclosure risk of micro data files. We approach this problem by considering some basic superpopulation models including the Multinomial-Dirichlet model, the Poisson- Gamma model of Bethlehem et al. (1990) and Takemura (1997), and the Modified Multinomial-Dirichlet model. We decided the critical population size of each superpopulation model for four different superpopulation models.

• The Korean Journal of Applied Statistics
• /
• v.9 no.2
• /
• pp.161-169
• /
• 1996

In the design of the sampling, it is important to make a decision about the size of the sample to be selected from the population. We often have a problem to get the optical size of the sample to be considered for cost and time expended for selecting sample unit from highly skewed population. In this case, we give a graphical criterion with Take-all Stratum rate to choose a method and also illustrate the efficiency between the Neyman allocation and the cut-off method with real data.

• Survey Research
• /
• v.5 no.1
• /
• pp.93-101
• /
• 2004

In business survey, cut-off sampling is usual, The contribution from cut-off part of the population is at least small in comparison with the remaining population. In this case, part of the target population is excluded from the selection and parameter estimations are only based on Take-all and Take-some stratum. It may be tempting not to use resources on enterprises that contribute little to the overall results of the survey. And this reduces the response burden for these small enterprises. But, the size of cut-off stratum has been increased as a way to manage reduced budgets. This leads to additional bias. In this study, the population have been separated as three stratum, cut -off, take-some, take-all, and we will estimate cut-off part using auxiliary variable.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.566-568
• /
• 2003

본 논문에서는 SCORM을 기반으로 한 e-Learning 시스템에서 학습자의 학습 활동을 트래킹하여 학습자의 수준을 적응적으로 판단하는 기법을 제시하였다. 제시된 기법에서는 모집단의 크기가 작을 경우 교수자가 지정한 난이도를 이용하여 학습자의 수준을 판단하고, 모집단의 크기가 충분히 클 경우에는 문항반응이론을 적응한 난이도에 의해 학습자의 수준을 판단하였다. 문항반옹이론을 적용할 시점에서 교수자가 지정한 난이도가 문항반응이론에서 추정한 난이도와 차이가 날 경우, 교수자가 지정한 난이도를 문항반응이론의 난이도로 수정하는 적응적인 기법을 제시하였다. SCORM의 트래킹 기능을 이용하여 실험한 결과 문제를 푼 학습자의 수가 적을 경우에는 학습자 수의 변화에 따라 학습자의 수준이 계속 바뀌는 문제점이 있음을 알 수 있었다. 따라서 모집단의 크기가 작을 경우, 본 논문에서 제안한 방법에 의해 교수자가 지정한 문항의 난이도를 이용하여 학습자의 수준을 판단하는 것이 효과적이었다.

• The Korean Journal of Applied Statistics
• /
• v.5 no.1
• /
• pp.9-17
• /
• 1992

In this paper we give a Bayesian approach to problems of estimation for the ratio in finite populations. Adopting the Ericson's superpopulatin approach in which the finite population of size N is viewed as arising form a random sample of N units from some superpopulation. We derive the exact posterior of the ratio under the noninformative prior on superpopulation parameters. Based on our results we compute an exact Bayesian confidence interval and compare this with the existing methods.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.1157-1161
• /
• 2009

본 연구에서는 부트스트랩(Bootstrap) 기법을 이용하여 측우기 강우량 관측계열(CWK)과 근대우량계 강우량 관측계열(MRG)에 대해 동질성 분석을 실시하였다. 서로다른 두 자료계열에 대한 전통적인 통계적 동질성 검정 방법은 모집단의 분포형을 알고 있어야 검정결과가 유효하였기 때문에 모집단의 분포가 복잡한 기상자료들은 이러한 전통적 방법을 사용하여 동질성을 파악하는 것이 매우 어려웠고 결과로 제시된 통계적 유의성에 대해서도 의심의 여지가 있었다. 이러한 이유로 본 논문에서는 모집단을 가정하지 않아도 되는 비모수적 모의 방법인 부트스트랩 기법을 이용하여 두 자료계열간의 동질성 검정을 실시하였다. 분석 결과 M20의 CWK와 MRG는 미소한 기후의 경년변화 (Trend)의 영향을 제외하면 동질성을 가진 자료로 볼 수 있었으나, 갈수기의 경우는 월강우량의 크기에 변화가 있으며 호우기의 경우는 일강우량의 크기 및 호우의 형태에 변화가 있는 것으로 나타났다.

• Journal of Educational Research in Mathematics
• /
• v.21 no.1
• /
• pp.17-32
• /
• 2011

This study investigated pre-service teachers' understanding of statistical sampling. The researchers categorized major topics related to sampling into representativeness of samples, sampling variability, and sampling distribution, and selected concepts connected to each topic. Findings on this study are as follows: Even though most of the pre-service teachers considered the random sampling bringing unbiased outcomes as a proper sampling method, only 64% of them recognized that sample is a quasi-proportional, small-scale version of population; Few pre-service teachers understood that more important is the size of sample, not the portion of sample to population, and half of them appreciated that the number of sampling has a powerful effect on drawing of reliable results than the size of sample; Few pre-service teachers understood that sampling distribute is irrelevant to the shape of population and has a symmetrical bell-shape.

• The Korean Journal of Applied Statistics
• /
• v.13 no.2
• /
• pp.207-224
• /
• 2000

Generally cluster size is predetermined when we use the stratified two-stage cluster sampling But in case that the sizes of clusters vary greatly one may want to make the sizes to be about equal. In this paper we study the optimal cluster size in stratified twostage cluster sampling. Also we find the optimal primary sampling unit sizes and optimal secondary sampling unit sizes under the given cost restriction.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.587-597
• /
• 2018

Procedures, such as sampling technique, survey method, and questionnaire preparation, are required in order to obtain sample data in accordance with the purpose of a survey. An important procedure is the decision of the sample size formula. The sample size formula is determined by setting the target error and total cost according to the sampling method. In this paper, we propose a sample size formula using population changes over time, estimation error of the previous time and response rate of past data when the target error and the expected response rate are given in the simple random sampling. In actual research, we use estimators that apply complex weights in addition to design-based weights. Therefore, we induce a sample size formula for estimators using design-based weights and nonresponse adjustment coefficients, that can be a formula that reflects differences in response rates when survey methods are changed over time. In addition, we use simulations to compare the proposed formula with the existing sample size formula.