Clearer Thinking with Spencer Greenberg Should we trust papers published in top social science journals? (with Daniel Lakens)
88 snips
Jul 24, 2024 
In this conversation with Daniel Lakens, an experimental psychologist at Eindhoven University of Technology, he tackles the trustworthiness of social science research. They discuss how p-hacking can distort findings and the need for rigorous methodologies. Daniel critiques psychology's fragmented theories, advocating for broader frameworks akin to physics. He also explores the limitations of peer review and suggests integrating diverse expertise to enhance research integrity. The episode finishes with insights on collaboration and the educational shifts needed in future research.
AI Snips
Chapters
Transcript
Episode notes
Trust in Social Science Papers
- Trust in social science papers depends on the specific paper.
- Newer papers tend to be better due to improved practices, but older, less rigorous methods can still appear.
Evaluating Paper Trustworthiness
- Look for falsifiable hypotheses, clear data analysis plans, and large datasets when assessing papers.
- Consider if the theory is theoretically sound and if the predictions are constrained.
Large Datasets and P-hacking
- Large datasets increase estimate accuracy, but don't eliminate p-hacking.
- P-hacking is still possible with flexible theories, statistical models, or data manipulation.