A neural network approach for assessing quality in technical education: an empirical study

The diverse nature of requirements of stakeholders in a Technical Education System (TES) makes it extremely difficult to decide on what constitutes quality. Hence, identification of common minimum quality items suitable to all stakeholders will help to design the system and thereby improve customer satisfaction. To address this issue, a measuring instrument known as EduQUAL is developed and an integrative approach using neural networks for evaluating service quality is proposed. The dimensionality of EduQUAL is validated by factor analysis followed by varimax rotation. Four neural network models based on back-propagation algorithm are employed to predict quality in education for different stakeholders. This study demonstrated that the P-E gap model is found to be the best model for all the stakeholders. Sensitivity analysis of the best model for each stakeholder was carried out to appraise the robustness of the model. Finally, areas of improvement were suggested to the administrators of the institutions.