Predicting learner performance with precision is critical within educational systems, offering a basis for tailored interventions and instruction. The advent of big data analytics presents an opportunity to employ Machine Learning (ML) techniques to this end. Real-world data availability is often hampered by privacy concerns, prompting a shift towards synthetic data generation. This study presents an empirical comparison of real, synthetic, and hybrid (real + synthetic) datasets in forecasting learner performance, deploying an array of regression-based ML algorithms, including Random Forest, Gradient Boosting, Support Vector Regression, XGBoost, and K-nearest Neighbor. Our methodology encompasses the generation of synthetic data via generative model, followed by the application of these algorithms to each dataset. The models are evaluated using precision metrics to assess their predictive accuracy. The study reveals that synthetic data can match real data in terms of predictive performance, with hybrid datasets achieving an accuracy of up to 87.76%, highlighting the effectiveness of combining both data types. These findings highlight the potential of synthetic data as an effective alternative when access to actual data is limited, promoting progress in educational technology and ML. [Note: The page range (108-129) shown on the PDF is incorrect. The correct page range is 108-131.]