SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Health Commun. Res. 2016; 15: 47-97.

Vernacular Title

메타분석 결과에서 통계적 오류의 영향: 헤지스․올킨 방법과 헌터․슈미트 방법의 비교 분석

Copyright

(Copyright © 2016, Korea Health Communication Association)

DOI

unavailable

PMID

unavailable

Abstract

This study is a meta-analytical methodological analysis of the approaches and features of the two methods of meta-analysis, Hedges & Olkin's approach and Hunter & Schmidt's approach. . In the meta-analytic methodology, we have examined the fundamental assumptions and the differences between the two methods in terms of major issues. We focused on the mutual comparison of the specific calculation process and the results. Respectively. Specifically, it is a research paper applying planned behavior theory, which is one of representative social cognition theory to explain human behavior performed in Korea until 2015, based on Hedges and Olkinmeta analysis method and Hunter and Schmidt meta analysis method. The results are presented.

METHODological differences such as sampling error and measurement error correction, confidence interval and confidence interval, Fisher z conversion value use, effect size comparison, and homogeneity test were also discussed.


This study attempted to review and compare two major meta-analytic approaches, the Hedges and Olkin approach and the Hunter and Schmidt approach, in the aspects of meta-analytic methodology. The basic methodological assumptions and characteristics of the two approaches were also discussed in terms of the analytic techniques, methodological procedures, and interpretations of the results. In addition, the two meta-analyses of the theory of planned behavior, one of the major social cognitive theories, were conducted. Based on the results, Fisher's z-transformation, effect size estimation, heterogeneity assessment, correction for sampling error and measurement error, and confidence interval and creditability interval were discussed.

본 연구는 메타분석의 대표적인 두 가지 방법인 헤지스와 올킨(Hedges & Olkin)이 제시하 는 방식과 헌터와 슈미트(Hunter & Schmidt)의 방식간의 접근방법과 특징에 대해 메타분석 방법론적 측면에서 비교분석하기 위해 진행하였다. 메타분석 방법과 관련하여 두 방식간의 방 법론적 기본 가정과 차이점을 주요 쟁점별로 살펴보았으며, 구체적인 산출 과정과 그 결과를 상호 비교하는데 초점을 맞춰 상이한 분석방법과 절차, 그리고 결과 해석의 차이점과 유의점 등을 검증하였다. 구체적으로 국내에서 2015년까지 수행된 인간 행동을 설명하기 위한 대표적 인 사회인지이론 가운데 하나인 계획된 행동이론을 적용한 연구논문을 대상으로 헤지스와 올킨 메타분석방법과 헌터와 슈미트 메타분석방식에 기초하여 그 결과를 제시하였다. 표집오차 및 측정오차 보정, 신뢰구간과 확신구간, 피셔(Fisher) z 변환값 사용 여부, 효과크기 비교, 동질성 검사 여부 등 방법론적 차이에 대해서도 논의하였다.


Language: ko

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print