Cross-national analyses often violate assumptions of non-independence due to connections between nations, inflating false positive rates. This paper demonstrates that cross-national research in economics and psychology insufficiently accounts for this non-independence. A review of highly cited studies reveals rare controls for non-independence, and simulations show that commonly used methods are inadequate. Reanalyses of prior studies demonstrate that half of the estimates are compatible with no association after controlling for non-independence. The authors strongly encourage social scientists to adequately control for non-independence in cross-national research.
Publisher
Nature Communications
Published On
Sep 18, 2023
Authors
Scott Claessens, Thanos Kyritsis, Quentin D. Atkinson
Tags
cross-national research
non-independence
economics
psychology
false positives
social sciences
statistical analysis
Related Publications
Explore these studies to deepen your understanding of the subject.
