Redefine or justify? Comments on the alpha debate.

Area of Science:

• Psychology
• Research Methodology

Background:

• The reproducibility of findings in psychological research is a significant concern.
• Current null hypothesis significance testing (NHST) practices, using an alpha level of .05, may not provide sufficient empirical evidence.
• A debate exists on how to improve the evidential value of statistical results.

Purpose of the Study:

• To evaluate the arguments against lowering the alpha level for NHST.
• To propose lowering the alpha level to .005 as a practical solution to the replication crisis in psychology.

Main Methods:

• Critical analysis of existing proposals regarding alpha levels in NHST.
• Argumentative synthesis of the debate between Benjamin et al. and Lakens et al.

Main Results:

• The arguments presented by Lakens et al. against lowering the alpha level to .005 are not convincing.
• Lowering the alpha level to .005 is deemed the most effective current strategy to address the replication crisis within the NHST framework.

Conclusions:

• The proposal to lower the alpha level to .005 offers a viable path to enhance research reproducibility.
• Given the persistence of the NHST paradigm, this adjustment is the most pragmatic approach to improve evidential standards in psychology.