Retract p < 0.005 and propose using JASP, instead

Area of Science:

• Psychology and life sciences research
• Scientific methodology and statistical analysis

Background:

• Research reproducibility is a significant challenge in science.
• A proposal suggests a stricter p < 0.005 threshold for statistical significance to enhance discovery claims.
• This proposal has generated considerable debate among researchers.

Purpose of the Study:

• To evaluate the proposed stricter p < 0.005 significance standard.
• To analyze counterarguments regarding the proposal's philosophical and methodological implications.
• To present a simpler, more suitable alternative for assessing statistical evidence.

Main Methods:

• Critical reflection on the original p < 0.005 proposal and its critiques.
• Exploration of Bayes factors analysis as a complementary statistical tool.
• Discussion of integrating frequentist error statistics with Bayesian beliefs.

Main Results:

• The stricter p < 0.005 standard faces significant philosophical and methodological challenges.
• Bayes factors analysis provides a framework for understanding both frequentist and Bayesian perspectives.
• Parallel use of Bayes factors analysis can enhance interpretation of statistical evidence.

Conclusions:

• The proposed p < 0.005 threshold may not be the optimal solution for reproducibility.
• Bayes factors analysis offers a more integrated approach to statistical inference.
• Utilizing Bayes factors analysis (e.g., via JASP) is recommended for a nuanced understanding of research findings.