Utilizing Ensemble Machine Learning Models to Analyze the Disproportionate Impact of Imposter Syndrome on Underrepresented Demographic Groups in Software Engineering

• Adaan Ahsun

  Covenant University

  Author

Imposter Syndrome (IS) affects over half of software engineers, with significantly higher prevalence among women and racial minorities, yet no validated predictive framework exists to identify at-risk individuals within specific underrepresented groups. This study addresses this gap by developing and validating an ensemble machine learning framework capable of analyzing the disproportionate impact of Imposter Syndrome on underrepresented demographic groups in software engineering. Using data from 624 software engineers across 26 countries, we constructed a stacking ensemble combining Random Forest, XGBoost, and Support Vector Machine classifiers, with logistic regression as the meta-learner. The framework achieved 89.4% accuracy in predicting severe Imposter Syndrome, with 87.2% precision and 91.3% recall, substantially outperforming single-model baselines. Feature importance analysis identified organizational environment, perceived competence evaluation, and demographic intersectionality as the strongest predictors. The study provides a replicable framework for organizations to identify vulnerable populations and implement targeted interventions, contributing to both the machine learning and software engineering literature on mental health prediction.

• pdf

Copyright (c) 2026 Adaan Ahsun (Author)

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