Replication of Chan and Albarracín (2023) "A meta-analysis of correction effects in science-relevant misinformation"
Chan M.-P.S. and Albarracín D. (2023) examined the impact of correction strategies on mitigating science-relevant misinformation, and found debunking effect was not significant. We replicated their study using the provided replication package, including raw data and analysis code. Our computational reproducibility confirmed the main results. However, our RoBMA analysis showed BF10 values of 1.422 with outliers and 1.437 without outliers, differing from the original study, suggesting lower publication bias. In our robustness reproducibility checks, we found discrepancies. While the misinformation correction effect became significant with different meta-analytic methods, cumulative meta-analysis showed no significant changes over time or with increasing sample sizes. Additionally, the WAAP-WLS analysis indicated no publication bias. These results suggest that while some findings align, the robustness of the debunking effect varies with analytical methods.
This paper received a response:
Chan, M-P. S., and D. Albarracín. 2025. Response to Replication of Chan and Albarracín (2023) "A meta-analysis of correction effects in science-relevant misinformation".