Methodological Advances in Consumer Research. Advances in Consumer Research
André, Quentin. Methodological Advances in Consumer Research. Advances in Consumer Research. 2021, Vol. 49, p821-826.ÌýÌýÌýÌý
Noise in the Process: A Meta-Analysis of Mediation Effects in Marketing Journals Aaron Charlton, College of Business, Illinois State University, USA Amanda Montoya, Department of Psychology, University of California Los Angeles, USA John Price, WU, Vienna University of Economics and Business, Austria Joe Hilgard, College of Arts and Sciences, Illinois State University, USA Paper #2: The same year, Simonsohn, Simmons and Nelson's "False-Positive Psychology" (2011) showed that common practices in behavioral research (e.g., not reporting all conditions and measures, or deciding when to stop data collection) allowed researchers to provide significant evidence for impossible results (e.g., that listening to "When I'm 64" by The Beatles can lower people's age by more than a year). Like the afore-mentioned meta-analyses, the current research examines the average power of mediation tests through analysis of reported statistics-specifically confidence intervals from mediation tests. To simulate true effects, we created a series of 3-variable datasets with a weak population-level indirect effect (ß=.O4) that fully mediated the relationship between X and Y. We settled on ß=.O4 because it allowed us to achieve desired power levels using sample sizes that are similar to those reported in marketing journals.ÌýÌýÌýÌý
https://www.proquest.com/docview/3090688396?pq-origsite=gscholar&fromopenview=true&sourcetype=Conference%20Papers%20&%20Proceedings