Prescriptive Faculty Performance Analysis: A Case at the Onset of Covid-19 Pandemic

ABSTRACT

The challenges of internationalization, the diversion to outcome-based education, and the emergence of the COVID-19 pandemic triggered a growing demand for quality educators. Hence, educational institutions shall ensure continuous evaluation of faculty performance and use its data as a tool to capacitate learning providers and enhance instruction in the classroom. Using the identified performance indicators, this study aims to elicit insights from the dataset extracted from the Faculty Performance Evaluation System (FPES) of the Camarines Sur Polytechnic Colleges (CSPC) to understand how the students perceived their respective instructors' performance levels prior to and at the onset of the COVID-19 pandemic. Generated patterns were uncovered using descriptive analysis based on the students' ratings. Meanwhile, the students' comments, suggestions, and recommendations were analyzed using Sentiment Analysis through TextBlob. The same dataset was further examined to recommend a prescribed action using a supervised learning method (Decision Tree Algorithm). With 98% model accuracy, faculty performance testing dataset were provided with prescribed actions with the following rules Outstanding & Very Satisfactory Ratings = Re-Hire/No Action Needed;Satisfactory = Mentorship;Unsatisfactory & Poor = Re-Training & Re-Evaluation. The study discovered a decline in the faculty performance evaluation results at the onset of the COVID-19 pandemic. However, the students' sentiments were considerate to the faculty's endeavor as most of its polarity scores fell under "positive." Recommendations to strengthen and boost faculty performance were incorporated based on the findings of the prescriptive analysis.