A neural network analytical model for predicting determinants of mobile learning acceptance

Abstract

User acceptance of technology is considered as one of the core fields in Human Computer Interaction (HCI) domain. The rapid development of mobile technologies during the last decade is playing a great role in the evolution of mobile learning applications. Hence, the main purpose of this study is to empirically explore and predict determinants that affect students' behavioural intention to accept m-learning using multi-analytical analyses: neural network and Multiple Linear Regression (MLR) models. This study applied neural network to provide further understanding of m-learning adoption based on a non-linear model. The data were collected through an online survey distributed to the undergraduate and graduate students at the University of Jordan. According to the analyses, the findings of this research show that the neural network model is a better choice than the multiple regression model to predict determinants of m-learning adoption and captured the non-linear relationship. The Artificial Neural Network (ANN) model results indicate that all the determinants are significant including, at some level, demographic variables. The MLR model results indicate that among the determinants only performance expectancy, efforts expectancy and social influence are significant to predict m-learning adoption.