A two-stage SEM-neural network analysis to predict drivers of m-commerce in India

Abstract

The rapid developments in the field of mobile technologies and deep penetration of smartphones have created tremendous opportunities for m-commerce worldwide. The purpose of this study is to investigate factors that predict consumer's intention to adopt m-commerce. The study identifies variables relevant for m-commerce environment and empirically establishes their influence on m-commerce adoption intention. A two-stage analysis comprising of structural equation modelling (SEM) and neural network (NN) technique is employed to test the proposed model. The results obtained from SEM analysis observed that perceived risk is the strongest predictor of m-commerce adoption decision, followed by performance expectancy, variety of services and perceived critical mass. Effort expectancy is found to be statistically insignificant. The significant factors from SEM were used as inputs to NN model and the results established performance expectancy to be the most important input variable in predicting m-commerce adoption intention followed by variety of services, perceived risk and perceived critical mass. The findings of this study are useful for m-commerce marketers and service providers, in developing suitable marketing strategies to scale up their business. This study is one of the few empirical studies conducted in India to examine the adoption intention of m-commerce.