Evaluating the performance of capture procedure and capture percentage in reproducibility research: a similation study
Loading...
Date
2018
Authors
Cheng, Yongtian
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In the era that researchers have been aware of the importance of replication studies, it is necessary for researchers to provide a criterion for effect size (ES), a commonly reported parameter in psychological study, to make a decision about whether it is successfully replicated. These days, a considerable amount of replication studies use the 95% confidence interval of effect size to determine whether the effect size has been successfully replicated, which is known as capture percentage (Cumming & Maillardet, 2006). Specifically, if the ES in the original study falls within the confidence interval of ES in a replication study, then researchers believe that the ES in the original study is successfully replicated. In this study, the two major uses of this criterion are discussed and evaluated. Based on the simulation results, the capture procedure cannot be used to determine the successful replication of the effect in a study.
Description
Keywords
capture percentage, replication crisis, confidence interval
Citation
APA