With Great Power Comes Great Responsibility: Common Errors in Meta-Analyses and Meta-Regressions in Strength &...

Area of Science:

• Sports Science
• Biostatistics
• Research Methodology

Background:

• Meta-analyses and meta-regressions are influential in practice.
• Statistical errors in meta-analyses are widespread and can lead to flawed conclusions.
• This study focuses on highly cited meta-analyses in strength and conditioning research.

Purpose of the Study:

• To review common statistical errors in meta-analyses.
• To document the frequency of these errors in highly cited strength and conditioning meta-analyses.

Main Methods:

• Identified five common statistical errors in meta-regression.
• Quantified the frequency of these errors in 20 highly cited meta-analyses from strength and conditioning research over the past 20 years.

Main Results:

• 85% of the 20 most highly cited meta-analyses contained statistical errors.
• 45% mistakenly calculated effect sizes using standard error instead of standard deviation, leading to exaggerated results.
• 45% failed to account for correlated observations within studies.

Conclusions:

• Statistical errors are prevalent in strength and conditioning meta-analyses.
• Highlights five critical errors for authors, editors, and reviewers to check.
• Emphasizes the need for rigorous statistical review to ensure reliable research findings.