Integrating Ensemble ML Imposter Syndrome Risk Metrics into Agile Project Management Systems

• Adaan Ahsun

  Covenant University

  Author

Agile project management environments, characterized by rapid iteration, continuous evaluation, and ambiguous role definitions, create conditions that amplify impostor syndrome among software professionals. Despite growing recognition that over half of software engineers experience frequent impostor feelings, existing project management tools lack integrated mechanisms for detecting and mitigating this psychological risk in real time. This study addresses this gap by developing and validating a framework that integrates ensemble machine learning impostor syndrome risk prediction into Agile project management workflows. Using the OSMI Mental Health in Tech Survey 2016 dataset (n=1,432) and a retrospective analysis of sprint-level project metrics, we implemented a RandomForestClassifier-based risk detection system alongside XGBoost, Support Vector Machine, and Logistic Regression baselines. The ensemble model achieved 89.4% accuracy in identifying elevated impostor syndrome risk, with precision of 87.2% and recall of 91.6%, outperforming all baselines. Feature importance analysis identified sprint review anxiety, retrospective participation hesitation, and estimation confidence gaps as the strongest organizational predictors. The framework demonstrates that psychological risk metrics can be embedded into existing Agile ceremonies without disrupting workflow, enabling early intervention at the team level. These findings offer practitioners a validated approach for monitoring team psychological health while maintaining individual privacy, and contribute to academic literature on integrating mental health informatics with project management systems.

• pdf

Copyright (c) 2026 Adaan Ahsun (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.