Evaluating the Use of Confidence Intervals in Quantifying Uncertainty and Improving Statistical Reporting Practices

Posted: Aug 08, 2019

• Ava Rodriguez
• Ava Smith
• Charlotte Davis

Abstract

Statistical inference forms the backbone of scientific research across disciplines, yet the methods for quantifying and communicating uncertainty remain contentious and often misunderstood. Confidence intervals have been widely promoted as superior alternatives to point estimates and null hypothesis significance testing for conveying the precision of estimates and the uncertainty inherent in statistical inference. Despite decades of advocacy from statisticians and methodological experts, the implementation and interpretation of confidence intervals in practice continue to exhibit significant shortcomings. This research addresses the critical gap between statistical theory and practical application by systematically evaluating how confidence intervals are used, interpreted, and reported across scientific domains.