No evidence for nudging after adjusting for publication bias
Maier, Bartos, Stanley, Shanks, Harris and Wagenmakers reply to the Mertens meta-analysis by arguing that its finding of moderate publication bias is the real headline. They reanalyse the same database using robust Bayesian meta-analysis, which combines selection models estimating the relative probability of publication with models of the relationship between effect size and precision, and lets the models that predict the observed data best carry the inference. The result is an absence of evidence for an overall effect, with evidence against an effect in the information and assistance intervention categories, while the evidence for structure interventions, the family that contains defaults, came out undecided. Almost every category also shows evidence of heterogeneity, which the authors note implies some nudges might work even where there is evidence against an average effect. The specificity of this result matters more than the headline.
Maier, Bartos, Stanley, Shanks, Harris and Wagenmakers reply to the Mertens meta-analysis by arguing that its finding of moderate publication bias is the real headline. They reanalyse the same database using robust Bayesian meta-analysis, which combines selection models estimating the relative probability of publication with models of the relationship between effect size and precision, and lets the models that predict the observed data best carry the inference. The result is an absence of evidence for an overall effect, with evidence against an effect in the information and assistance intervention categories, while the evidence for structure interventions, the family that contains defaults, came out undecided. Almost every category also shows evidence of heterogeneity, which the authors note implies some nudges might work even where there is evidence against an average effect. The specificity of this result matters more than the headline.
Written from the abstract, Semantic Scholar.
What our sources record
Reproduced as it was written when this source was checked, figures and all.
Maier, M., Bartos, F., Stanley, T. D., et al. (2022). 'No evidence for nudging after adjusting for publication bias.' PNAS, 119(31), e2200300119. doi:10.1073/pnas.2200300119
Where we use it
- The Psychology You Learned Didn't Replicate, Why We Act
Reliability of this reference
The link below was matched against the publisher's record on title, first author and year, and all three agree.
The source
https://doi.org/10.1073/pnas.2200300119
DOI: 10.1073/pnas.2200300119