Exploring the Application of Logistic Regression in Predicting Binary Outcomes Across Health and Social Sciences

Posted: Jul 18, 2022

• Amelia Green
• Aria Clark
• Aria Lewis

Abstract

Logistic regression represents one of the most widely employed statistical methodologies for binary classification problems across numerous disciplines. Despite its mathematical elegance and interpretability, the cross-disciplinary application of this technique has received insufficient scholarly attention. This research addresses this gap by systematically examining the performance, interpretability, and practical utility of logistic regression models across three distinct domains: healthcare diagnostics, educational attainment prediction, and social behavior forecasting. The fundamental research question guiding this investigation concerns whether the consistent mathematical foundation of logistic regression translates to consistent performance and utility across diverse application domains, or whether domain-specific characteristics necessitate substantial methodological adaptations.