A hybrid SEM-neural network method for modeling the academic satisfaction factors of architecture students

Abstract

With the increasing development of universities and institutions of higher education, student satisfaction is one of the most critical issues for their acceptance to new areas, the main feature of which is the competition to attract and retain students. Despite numerous studies on student satisfaction, more conceptual and methodological understanding is needed to increase the explanatory and predictive power of student satisfaction modes. The first objective of this study is to use artificial neural networks (ANN) to measure the components of student satisfaction. The second goal is to explain the specific components of academic students' Academic Satisfaction through the SEM-ANN combination. Data were collected from 420 graduate students at Tabriz Islamic Art University. The research model was Obtained through a multi-analytic Including structural equation modeling (SEM) and ANN. We used the results of SEM as input to the ANN model to develop a predictive model of Academic Satisfaction of architecture students. Based on the SEM results, we found that the components of educational services (0.99), sociocultural (0.95), and perceptions (0.85) are the most critical components affecting the Academic Satisfaction of architecture students. At the same time, the economic-entrepreneurial component (0.37) has little effect on the Academic Satisfaction of architecture students. The analytical results in ANN using the multi-layer perceptron model led to the explanation of a model with high predictability of components of Academic Satisfaction of architecture students. The findings of this study can be critical, and valuable for universities in predicting students' Academic Satisfaction. The results of this study enable universities to predict the extent of its impact on student satisfaction before formulating their programs, and to cover the stimuli affecting students' Academic Satisfaction with appropriate measures. As a result, it is increasing students' Academic Satisfaction.

Keywords

• Academic satisfaction
• Artificial neural networks
• Architecture
• Structural equations model