Statistics in Infection and immunity revisited

Area of Science:

• Microbiology
• Immunology
• Biostatistics

Background:

• A 2003 review highlighted significant statistical analysis and reporting errors in over half of Infection and Immunity (IAI) articles.
• This issue can compromise the validity and reproducibility of published research findings.

Purpose of the Study:

• To update the assessment of statistical methods used in recent Infection and Immunity (IAI) publications.
• To identify and discuss common statistical reporting errors and provide examples of best practices.
• To enhance the IAI "Instructions to Authors" criteria for statistical analysis and reporting.

Main Methods:

• Systematic review of statistical methodologies employed in recent IAI articles.
• Analysis of reporting quality, identifying common errors and exemplary cases.
• Comparison with existing IAI statistical guidelines.

Main Results:

• Common statistical methods in IAI include t-tests, ANOVA, chi-squared tests, and regression analyses.
• Frequent errors involve incorrect application of tests, insufficient detail in reporting, and P-value misinterpretation.
• Examples of both sound and flawed statistical reporting are identified.

Conclusions:

• Despite previous reviews, statistical reporting errors remain prevalent in Infection and Immunity (IAI) research.
• Clearer guidelines and improved author education are crucial for enhancing statistical rigor.
• Adherence to enhanced statistical criteria will improve the quality and interpretability of published studies.