Insights in Hypothesis Testing and Making Decisions in Biomedical Research

Area of Science:

• Biostatistics
• Statistical Inference
• Research Methodology

Background:

• P values are widely employed for statistical inference across biomedical and social sciences.
• Misinterpretation and misuse of p values are prevalent issues in research.
• Confidence intervals offer a more robust alternative for statistical inference.

Purpose of the Study:

• To review the application of p values in hypothesis testing.
• To discuss resampling methods, such as the bootstrap, as alternatives for calculating confidence intervals.
• To highlight the advantages of confidence intervals over p values for robust inference.

Main Methods:

• Review of existing literature on p values and hypothesis testing.
• Discussion of resampling techniques, specifically the bootstrap method.
• Exploration of confidence interval calculation using these methods.

Main Results:

• P values, while common, are prone to misuse and misinterpretation.
• The bootstrap method provides a robust approach to calculating confidence intervals.
• Confidence intervals offer a more informative measure for statistical inference compared to p values.

Conclusions:

• Resampling methods like the bootstrap are recommended for calculating confidence intervals.
• Confidence intervals provide more reliable and interpretable results than p values.
• Adopting confidence intervals can improve the rigor of statistical inference in research.