Significant Results from Dr. Bassem Mohsen Yehia’s DBA Research at IBAS Switzerland

School dropout is still one of the most pervasive and detrimental issues facing the country’s educational system, even after the Egyptian government made significant efforts to make basic education mandatory. The problem has significant social and economic ramifications in addition to its educational ones. Dropouts experience increased unemployment, fewer possibilities in life, and social marginalization, which contributes to a difficult-to-break cycle of poverty.

Dr. Bassem Mohsen Yehia offers a potent, data-driven strategy to deal with this dilemma in his ground-breaking Doctorate of Business Administration (DBA) thesis at the International Business Academy of Switzerland (IBAS). His work provides a predictive model that helps identify students at risk of dropping out—early enough to make a difference—and goes beyond simply detecting symptoms.

The study, technique, and insights of Dr. Bassem Mohsen Yehia are examined in this paper, which also provides a useful framework for legislators, school officials, and data scientists to address Egypt’s dropout problem.

The Dropout Dilemma: A Threat to National Progress

Even though all Egyptian children are required by law to get a basic education, hundreds of them nevertheless leave the system too soon. These dropouts are actual people whose futures are at risk because they left school, not simply statistical statistics.

“Dropout students often lack the skills and attributes required to compete in the modern labor market. Their exclusion from education restricts both individual potential and national progress.” – Dr. Bassem, 2024

School dropouts frequently suffer from:

• Low academic motivation
• Behavioral and psychological challenges
• Poor home support
• Long-term unemployment and economic vulnerability 

Understanding—and more importantly, predicting—who is most at risk becomes essential if Egypt is to meet its educational development goals.

A Predictive Model for Early Intervention

The main contribution of Dr. Bassem is the development of a logistic regression model that can predict which students are most likely to drop out. The study addresses class imbalance, a problem in data science when the proportion of dropout instances in a dataset is much lower than that of non-dropout cases.

Instead of ignoring this imbalance, Dr. Bassem confronts it with a robust comparative approach using advanced sampling techniques:

• Over-sampling: Adding synthetic data to minority (dropout) cases
  

• Under-sampling: Reducing data from majority (non-dropout) cases
  

• Hybrid methods: Combining both to balance datasets without losing information
  
  

The objective? to ensure minimum Type II error—when the model is unable to identify a true dropout case—and to enhance the model’s capacity to forecast actual dropouts without incorrectly identifying pupils.

“Balancing the dataset through hybrid sampling—especially combining ROS with NearMiss-3—produced the best model performance with high AUC and F-score and the lowest Type II error.” – Dr.Bassem, 2024

Why Class Imbalance Matters in Dropout Prediction

Extremely unbalanced datasets are problematic for many machine learning methods. Because there are few dropout cases in the realm of education, standard models usually perform badly. Dr. Bassem demonstrates that resampling is required and not optional.

Key findings include:

• The dataset’s structure heavily influences which sampling methods are most effective.

• Combined sampling techniques outperform individual ones.

• Logistic regression, when paired with balanced data, is highly effective in educational contexts.

These insights lay the foundation for building accurate early warning systems in Egypt’s schools, allowing for proactive interventions instead of reactive measures.

Core Predictors of Dropout in Egyptian Basic Education

Beyond the model, the study identifies five major factors that significantly increase a student’s likelihood of dropping out:

1. Chronic Health Conditions

Chronically unwell students frequently miss class, lag behind academically, and lose interest in their studies. They have a higher danger of dropping out permanently if they don’t receive healthcare help.

2. Parental Illiteracy

Parents find it difficult to assist their children’s educational journeys when they are illiterate. Early school departure is silently predicted by this disparity in home participation.

3. Low Academic Performance

Students who do poorly on a regular basis are more prone to get disillusioned and lose confidence, which might ultimately result in withdrawal. Failure in school is frequently ignored until it is too late.

4. Lack of Teacher Support

At-risk pupils feel alone when teachers don’t offer them academic or emotional assistance. The difference between persistence and dropout is frequently a teacher’s kind demeanor.

5. Co-Education Limitations

Limited access to mixed-gender educational environments in certain regions can restrict opportunities, particularly for girls, leading to early discontinuation of education.

“These five factors are not just data points—they are lives hanging in the balance. Early action on these predictors can save thousands from long-term exclusion.” – Dr.Bassem, 2024

From Insight to Action: Policy Recommendations

Dr. Bassem’s argument goes beyond analysis. It offers a methodical road map for turning forecasted information into legislative actions. His suggestions cover management, policy, and technological areas.

1. Launch National Early Warning Systems

The Ministry of Education can keep an eye on children who are at danger in real time by using prediction models like Dr. Bassem’s. Schools can get alerts for prompt academic help, medical referrals, or counseling.

2. Invest in Teacher Training and Mentorship

Teachers are the front line. Empower them with:

• Emotional intelligence training

• Skills to identify early signs of disengagement

• Tools for inclusive classroom practices
  

This not only boosts retention but also strengthens teacher-student relationships.

3. Support for Parents

By strengthening the support structure at home, community outreach initiatives, adult literacy courses, and seminars for parental engagement can address parental illiteracy.

4. Healthcare Partnerships

Through public-private partnerships, basic healthcare services may be offered in schools, assisting children with chronic diseases and reducing absenteeism caused by preventable medical issues.

5. Educational Technology and Digital Inclusion

Where traditional schooling faces challenges, digital tools can bridge gaps. Mobile apps, SMS alerts, and digital homework platforms can keep students connected to learning even in hard-to-reach areas.

Methodological Strength: Why This Research Matters

The research methodology is especially strong because it:

• Uses a logistic regression framework—trusted for transparency and interpretability.

• Applies comparative sampling strategies, offering more than one pathway to optimization.

• Anchors conclusions in practical, real-world education policy relevance. 

While the findings are based on a specific Egyptian dataset, the model has the potential to be adapted for use in other developing countries facing similar educational challenges.

“Predictive analytics must not stay in academia—they must be integrated into national strategy and school practice.” – Dr. Bassem, 2024

Limitations and Future Research

Every robust study acknowledges its boundaries. Dr. Bassem,notes that:

• The dataset reflects a specific educational occurrence in Egypt and may not represent all regions equally.

• Further testing is needed in urban vs. rural schools.

• Psychological and social data, though critical, were not deeply embedded in the current dataset. 

Future research could explore:

• Comparative studies across Middle Eastern and African countries

• Longitudinal tracking of students post-intervention
• The role of AI and machine learning in continuous dropout risk monitoring
  

Final Reflections: Education as a Cornerstone of Development

Dropping out is a systemic failure, not just a personal decision. Dr. Bassem Mohsen Yehia’s research provides a data-supported, practical approach to address this issue, and if incorporated into national policy, it has the potential to change the course of millions of children’s lives and further Egypt’s socioeconomic development objectives.

“Education must not just be accessible—it must be sustainable. Prediction gives us the power to act before it’s too late.” – Dr. Bassem, 2024

Be Part of the Transformation

Are you an educator, policymaker, or data specialist passionate about reshaping public systems with advanced tools?

Join the DBA program at IBAS–VERN and master the skills needed to turn research into real-world solutions. Gain access to world-class faculty, hands-on data projects, and a global network of change-makers.

🔗 Apply now and lead change in education, one data point at a time.