• Title/Summary/Keyword: 메타분석

Search Result 812, Processing Time 0.449 seconds

Robust Bayesian meta analysis (로버스트 베이지안 메타분석)

  • Choi, Seong-Mi;Kim, Dal-Ho;Shin, Im-Hee;Kim, Ho-Gak;Kim, Sang-Gyung
    • Journal of the Korean Data and Information Science Society
    • /
    • v.22 no.3
    • /
    • pp.459-466
    • /
    • 2011
  • This article addresses robust Bayesian modeling for meta analysis which derives general conclusion by combining independently performed individual studies. Specifically, we propose hierarchical Bayesian models with unknown variances for meta analysis under priors which are scale mixtures of normal, and thus have tail heavier than that of the normal. For the numerical analysis, we use the Gibbs sampler for calculating Bayesian estimators and illustrate the proposed methods using actual data.

A Bayesian Method to Semiparametric Hierarchical Selection Models (준모수적 계층적 선택모형에 대한 베이지안 방법)

  • 정윤식;장정훈
    • The Korean Journal of Applied Statistics
    • /
    • v.14 no.1
    • /
    • pp.161-175
    • /
    • 2001
  • Meta-analysis refers to quantitative methods for combining results from independent studies in order to draw overall conclusions. Hierarchical models including selection models are introduced and shown to be useful in such Bayesian meta-analysis. Semiparametric hierarchical models are proposed using the Dirichlet process prior. These rich class of models combine the information of independent studies, allowing investigation of variability both between and within studies, and weight function. Here we investigate sensitivity of results to unobserved studies by considering a hierachical selection model with including unknown weight function and use Markov chain Monte Carlo methods to develop inference for the parameters of interest. Using Bayesian method, this model is used on a meta-analysis of twelve studies comparing the effectiveness of two different types of flouride, in preventing cavities. Clinical informative prior is assumed. Summaries and plots of model parameters are analyzed to address questions of interest.

  • PDF

A Meta-Analysis of the Effects of Aromatherapy Hand Massge (아로마테라피 손마사지 효과에 대한 메타분석)

  • Park, Yeon-Suk
    • Journal of Digital Convergence
    • /
    • v.13 no.1
    • /
    • pp.469-479
    • /
    • 2015
  • Purpose: The purpose of this study is to explore the effect of nursing intervention of aromatherapy hand massage. Method: In order to conduct a meta-analysis, a total of 124 studies were retrieved. From these studies, fifteen studies met the inclusion criteria with a total of 733 participants. And these studies published from January 2004 to July 2013 were included in the journal articles published in Korea. Results: In fifteen studies, Random effect model for the aromatherapy hand massage that 95% confidence level, significant level of p-valve on less than 0.001. Standardized Mean Difference reported systolic blood pressure(d=-9.5320), diastolic pressure(d=-8.8730), pulse(d=-9.1241), anxiety(d=-0.1469), and depression(d=-0.4489) All measurement variables were significantly decreased in the negative direction. Conclusion: This study suggests that aromatherapy hand massage can reduce the level of blood pressure(systolic, diastolic), pulse, anxiety and depression.

A Meta-Analysis of Research Trends in Mathematics Learning Disabilities (수학학습장애 연구 동향 메타분석)

  • Jeon, Yoon-Hee;Chang, Kyung-Yoon
    • Journal of Educational Research in Mathematics
    • /
    • v.26 no.3
    • /
    • pp.543-563
    • /
    • 2016
  • This study was designed as a meta-analysis to investigate the research trends in mathematics learning disabilities(MLD) area. The results of this study were as follows: The 201 researches targeted for the analysis can be categorized 4: characteristic of students with MLD, screening students with MLD, interventional teaching for students with MLD, and et cetera. Also, the outcomes of researches regarding intervention in MLD determined to have a large effect resulted in a total average of 0.958. Especially, as a result of analysing the effect size in accordance with teaching method variables in group-case designed researches, the effect was largest when direct instruction and strategy instruction was given. The effect was largest when the frequency of intervention was over 16 and under 20. The results in this study be summed up as follows. MLD can be served as a foundation in setting a direction for further research to improve in Korea.

A Meta-analysis of Relationship among Satisfaction, Trust, and Loyalty in E-commerce (전자상거래 연구에서 만족, 신뢰 그리고 충성도 간에 관계에 관한 메타분석)

  • Nam, Soo-Tai;Yang, Ki-Seol;Jin, Chan-Yong
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.19 no.7
    • /
    • pp.1711-1718
    • /
    • 2015
  • A Meta-analysis refers to a statistical literature synthesis method from the quantitative results of many known empirical studies. Recently, the convergence of knowledge-based society and, information telecommunication technologies has a rapid impact on politics, economics and various fields. A meta-analysis was conducted to identify the relationships among satisfaction, trust, and loyalty in e-commerce studies. A total of 57 research papers published in Korean academic journals during 2002 and 2013 were reviewed and the causal relationships among satisfaction, trust, and loyalty in e-commerce were established. The result of the meta-analysis might be summarized that the highest effect size (r =. 591) was in the path from the satisfaction to the trust. The second biggest effect size (r =. 554) was found in the path between the satisfaction to the loyalty. The third biggest effect size (r =. 552) was found in the path between the trust to the loyalty. By the way, the smallest effect size (r =. 484) was found in the path between the trust to the satisfaction.

A Meta-analysis and Review of External Factors based on the Technology Acceptance Model : Focusing on the Journals Related to Smartphone in Korea (기술수용모델 선행요인에 관한 문헌적 고찰 및 메타분석)

  • Nam, Soo-Tai;Shin, Seong-Yoon;Jin, Chan-Yong
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.18 no.4
    • /
    • pp.848-854
    • /
    • 2014
  • A Meta-analysis refers to a statistical literature synthesis method from the quantitative results of many known empirical studies. We conducted a meta-analysis and review of external factors based on the technology acceptance model for Smartphone-related researches. This study surveyed 106 research papers that established causal relationships in the technology acceptance model published in Korean academic journals during 2008 and 2013. The result of the meta-analysis might be summarized that the playfulness has the highest effect size in the path from external factors to the perceived usefulness, with the effect size(0.536). Also the self efficacy showed the highest effect size(0.626) in the path from external factors to the perceived ease of use. Based on these findings, several theoretical and practical implications were suggested and discussed with the difference from previous researches.

A Meta-Analysis on the Solution Focused Group Program Effects for Elementary School Children (초등학생을 위한 해결중심 집단상담프로그램의 효과에 관한 메타분석)

  • Park, Jung-Im
    • The Journal of the Korea Contents Association
    • /
    • v.14 no.11
    • /
    • pp.476-485
    • /
    • 2014
  • This study was to examine the effectiveness of the Solution Focused Group Counseling Program for elementary school students by meta-analysis. For the purpose of the study, master's theses, doctoral dissertations, and journal articles published in Korea up to May, 2014 were systematically reviewed. As a result, a total of 20 studies were eligible for the inclusion criteria. The mean effect sizes and test for homogeneity of effect size (Q-statistic) were analyzed by using Comprehensive Meta-Analysis software 2.0. The main findings of the study were as follows: First, the average effect sizes for Solution Focused Group Counseling Program were ES=1.61 in self-esteem, ES=1.35 in school adjustment capacity, ES=1.07 in interpersonal relationship and ES=1.03 in self-efficacy, Second, the statistically significant moderate variables of the effect size for the Solution Focused Group Counseling Program were 'sessions' of self-esteem and 'sessions', 'running times' in school adjustment capacity. Based on the study results, the research and practice implications were discussed.

Meta-Analysis on the Effects of Programming Education using Educational Programming Languages (교육용 프로그래밍 언어를 사용한 프로그래밍 교육의 효과에 관한 메타분석)

  • Yang, Changmo
    • Journal of The Korean Association of Information Education
    • /
    • v.18 no.2
    • /
    • pp.317-324
    • /
    • 2014
  • In this paper, we investigate the effects of programming education using educational programming languages. We select 70 papers which contain 101 individual experimental studies we can estimate the effect size from. The average effect size of the studies is .641, which means that programming education has modest effects on the learners. The average effect size of the learners' cognitive and affective domains are 0.632 and 0.666, respectively. The average effect size by the controlled variables such as school level, experimental periods, and programming languages have no statistically significant difference. The results of our study show that the experimental group receiving programming education, can be expected perform about 23.9% better compared to the control group having received no programming education.

The Effect of Internet Addiction Prevention Program on Adolescents' Self-regulation -A Systematic Review and Meta-analysis- (인터넷중독 예방프로그램이 청소년의 자기통제력에 미치는 효과 -체계적 문헌고찰 및 메타분석-)

  • Chung, Myung Sill;Seomun, GyeongAe
    • Journal of Digital Convergence
    • /
    • v.17 no.8
    • /
    • pp.347-355
    • /
    • 2019
  • The purpose of this study was to investigate the effect of internet addiction prevention program on adolescents' self-regulation using systematic review and meta-analysis. Databases od domestic literature were used, Library of Congress, KERIS, KISS, domestic Journal, until 2018, eleven articles were used for meta-analysis. As a result of the study, there was statistically significant difference in the overall effect of self-regulation on the internet addiction prevention program, but the heterogeneity was high and subgroup analysis was conducted. The effects of self-regulation on the Internet addiction prevention program according to subjects were statistically significant in both elementary and middle & high school students. The effect size of self-regulation by internet addiction was statistically significant..Therefore, it's necessary to develop an Internet addiction prevention program according to the characteristics of elementary and middle & high school students.

Evidence-based Medicine and Systematic Review (근거중심의학과 체계적 고찰)

  • Kim, Soo Young
    • Quality Improvement in Health Care
    • /
    • v.12 no.2
    • /
    • pp.35-40
    • /
    • 2006
  • 체계적 고찰은 명확히 정의된 문헌검색에 기반하고, 결정된 기준(criteria)을 이용하여 선택된 문헌의 질을 평가하며, 타당성이 입증된 방법으로 분석을 시행하는 리뷰의 한 형태이다. 체계적 고찰은 (1)답을 얻고자 하는 질문을 공식화 하고 이에 대한 프로토콜을 작성하고, (2)해당 질문에 대한 일차 연구를 찾아서 포함시키고 (3)자료의 질을 평가하고, (4)자료를 추출하여 뽑고, (5)자료를 분석하며, (6)결과를 해석하고 보고서를 작성하는 순서로 이루어진다 하지만 체계적 고찰 혹은 메타분석이 모든 문제에 대한 해결책은 아니다. 메타분석이 수행되었다고 하더라도 질이 나쁜 연구가 포함되거나 질 문제를 무시한 경우, 이질성에 대해 적절한 주의를 기울이지 못할 때, 무차별적인 자료 병합으로 결론이 정확하지 않은 경우, 출판 비뚤림, 시간 지체 비뚤림, 이중 출판 비뚤림, 언어 비뚤림, 결과 보고 비뚤림과 같은 보고 비뚤림이 문제가 되는 경우에는 메타분석 혹은 체계적 고찰이 잘못된 결론으로 도달하도록 할 수 있다. 따라서 올바른 방법론에 기초한 체계적 고찰만이 현존하는 모든 근거를 결합하여 올바른 결론으로 이끌 수 있다.

  • PDF