In 2020, I developed an automated system to evaluate faculty performance and generate personalized reports across an entire university, involving hundreds of professors and thousands of students. The project integrated academic and administrative data to compute key performance indicators (KPIs) and automatically deliver individual feedback reports.
Data was collected from multiple internal sources, including attendance records, grades, student feedback, grade distributions, pass rates, and administrative attributes. Both faculty and student records were structured into a relational database, enabling large-scale cross-referencing for every class and academic group.
The system organized data by academic term and consolidated each professor’s performance across several dimensions. Some of the calculated KPIs included:
Once KPIs were computed, the system generated individual reports using R Markdown, combining quantitative analysis with visualizations and auto-generated narratives. Each report included:
Reports were exported directly to PDF and sent to each professor, providing clear, actionable feedback without manual effort.
This solution enabled the university to implement data-driven faculty evaluations at scale without additional operational cost. It also increased transparency in evaluation criteria and helped identify areas for academic and pedagogical improvement.
Project developed in 2020 and deployed institutionally during the 2020-I academic term.