Bayesian evidence synthesis is no substitute for meta-analysis : a re-analysis of Scheibehenne, Jamil and Wagenmakers (2016)

• Scheibehenne, Jamil, and Wagenmakers (2016; SJW) recently introduced Bayesian evidence synthesis (BES). They applied it to a set of original studies that examined the influence of social norms on towel reuse at hotels. While most of the original studies provided nonsignificant results (p > .05), BES provided “strong support” (p. 3) for the effect. Due to methodological limitations, we think that this conclusion is wrong and that BES suffers from several problems. Combining frequentist and Bayesian approaches, we: illustrate the perilsof pooling data; assess publication bias, and conduct a Bayesian meta-analysis.

Subject headings

SAMHÄLLSVETENSKAP  -- Ekonomi och näringsliv -- Nationalekonomi (hsv//swe)

SOCIAL SCIENCES  -- Economics and Business -- Economics (hsv//eng)

Keyword

Economics

Nationalekonomi

Publication and Content Type

vet (subject category)

rap (subject category)