Automated reproducibility assessments in the social and behavioral sciences using large language models
2026-06-11 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors tested whether large language models (LLMs) can help check if social and behavioral science studies are reproducible without needing lots of human work. They used 76 studies and found that the LLM could closely match original effect sizes in 41% of them and agreed with the study conclusions 96% of the time. Humans matched effect sizes 34% of the time and agreed with conclusions 74% of the time. The authors suggest LLMs can be a helpful, scalable way to automatically assess reproducibility in research.
ReproducibilityLarge Language ModelsEffect SizeCohen's dBehavioral SciencesSocial SciencesAutomated AnalysisResearch ReplicationQualitative Conclusion
Authors
Tobias Holtdirk, Pietro Marcolongo, Anna Steinberg Schulten, Felix Henninger, Stefan Rose, Sarah Ball, Bolei Ma, Frauke Kreuter, Markus Weinmann, Stefan Feuerriegel
Abstract
Reproducibility in the social and behavioral sciences is typically evaluated by independent researchers who reanalyze the original data to assess whether the published findings can be recovered. However, such approaches are resource-intensive and difficult to scale. Here, we show that large language models (LLMs) can automate reproducibility assessments. Using N=76 published studies with predefined claims from the behavioral and social sciences, we compare LLM-generated analysis with the original findings and human reanalysis. For 7 studies, the LLM could not produce a viable effect size estimate. For the remaining studies, our LLM pipeline recovered the original effect sizes in 41% of studies using a +/-0.05 tolerance in Cohen's d. Further, our LLM pipeline reached the same qualitative conclusion as the original study in 96% of cases, where conclusions indicate whether the reanalysis supports the original claim. For comparison, human reanalysts recovered the original effect sizes in 34% of studies and reached the same qualitative conclusion in 74% of cases. Together, these results show that LLMs can serve as a scalable tool for automated reproducibility assessment and provide a foundation for systematic auditing of empirical results in the social and behavioral sciences.