Benchmarking higher education institutes using data envelopment analysis: capturing perceptions of prospective engineering students

In today’s exceptionally demanding education domain, there is a critical need for higher education institutes to continuously improve their credibility and skill offerings for prospective students. More specifically, institutes need to know how they are evaluated and perceived by the prospective students. Though there are plethora of rating agencies that perform an independent evaluation, their assessment is primarily focused on internal institute practices rather than how students’ needs and perceptions be captured. The objective of this study is to evaluate how students rank a higher education institute for taking admission that is the most crucial decision while starting their long-term career. Academic experts and prospective engineering students are surveyed using interviews and convenience sampling method. As a result, the problem evolves as a multi-criteria decision-making problem that considers numerous criteria simultaneously. Considering the relevant criteria as inputs and outputs, the Data Envelopment Analysis is employed to evaluate the weights corresponding to each criterion/factor. The model is then deployed as a linear programming formulation that is tackled using professional linear programming solver. The engineering stream, fees, location, and perceptions of employment opportunities are found as the top parameters that drive the decision of a prospective engineering student. The identified important factors, interests, and capabilities of the student are amalgamated for the selection of the appropriate engineering stream and institute, optimally. The insights are important for engineering institutes to strategize and align their offerings and marketing approach as per the needs and perceptions of prospective students. The study will also be helpful for international universities looking ahead for collaborative and individual opportunities in the Indian education sector.