P-Curve: Statistical Significance Distributions for Meta-Analysis

P-curve analysis addresses publication bias and p-hacking by examining the distribution of statistically significant p-values across a set of studies. The method was introduced by Uri Simonsohn of the University of Pennsylvania. It is designed to help determine whether reported effects reflect true findings or selective reporting.

Use Cases

• Detecting publication bias based on the distribution of p-values.
• Assessing the evidential value of a research literature based on reported statistical significance.
• Testing for p-hacking in a collection of study results.
• Conducting meta-analytic research on research practices.

Strengths

• Method introduced by a named author from a specific institution (Uri Simonsohn, University of Pennsylvania).
• Addresses a well-defined and significant problem in research methodology (publication bias, p-hacking).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

Uri Simonsohn, University of Pennsylvania

Collection Method

Likely contains statistical results or p-values from a collection of studies, as described in the methodology.