99% Impossible: a Valid, Or Falsifiable, Internal Meta-Analysis
Researchers are increasingly relying on internal meta-analysis (IMA) to document cumulative evidence. We demonstrate that even minor violations of the assumptions underlying IMA invalidate its results—and once established—a false-positive IMA is practically impossible to falsify. We conclude that IMA should be used exclusively for pre-registered many-lab replications.
Citation:
Joachim Vosgerau, Uri Simonsohn, Leif D. Nelson, and Joseph P. Simmons (2019) ,"99% Impossible: a Valid, Or Falsifiable, Internal Meta-Analysis", in NA - Advances in Consumer Research Volume 47, eds. Rajesh Bagchi, Lauren Block, and Leonard Lee, Duluth, MN : Association for Consumer Research, Pages: 212-216.
Authors
Joachim Vosgerau, Bocconi University, Italy
Uri Simonsohn, University of Pennsylvania, USA
Leif D. Nelson, University of California Berkeley, USA
Joseph P. Simmons, University of Pennsylvania, USA
Volume
NA - Advances in Consumer Research Volume 47 | 2019
Share Proceeding
Featured papers
See MoreFeatured
O11. Have Less, Compromise Less: How the perception of resource scarcity influences compromise decisions
Kate Kooi, University of Miami, USA
Caglar Irmak, University of Miami, USA
Featured
“It’s Not You, It’s Me”: How Corporate Social Responsibility Decreases Customer Citizenship Behavior
Sofia Batista Ferraz, EAESP-FGV
Andres Rodriguez Veloso, University of São Paulo, Brazil
Diogo Hildebrand, Baruch College, USA
Featured
A Conceptual Framework of Violation of Trust and Negative Emotional Responses during Brand Transgressions
Karthik Selvanayagam, Indian Institute of Technology, Madras
Varisha Rehman, Indian Institute of Technology, Madras