A SEM-Neural Network Approach for Predicting Antecedents of Factors Influencing Consumers' Intent to Install Mobile Applications

A SEM-Neural Network Approach for Predicting Antecedents of Factors Influencing Consumers' Intent to Install Mobile Applications

Yakup Akgül (Alanya Alaaddin Keykubat University, Turkey)
DOI: 10.4018/978-1-5225-4029-8.ch012

Abstract

This chapter explores the present gap in the literature regarding the acceptance of mobile applications by investigating the factors that affect users' behavioral intention to use apps in Turkey. First, structural equation modeling (SEM) was used to determine which variables had significant influence on intention to install. In a second phase, the neural network model was used to rank the relative influence of significant predictors obtained from SEM. The results reveal that habit, performance expectancy, trust, social influence, and hedonic motivation affect the users' behavioral intention to use apps.
Chapter Preview
Top

Introduction

Recently, people are progressively eager to adopt new technologies in their daily lives by accepting and using modern technologies is common habit, making technology, now more than ever become a part of our usual activities (Islam, Low and Hasan, 2013). As of May 2017, Apple’s App Market contained 2.2 million and Google’s Google Play Market contained 2.8 million apps (Statista, 2017a). Consumers have downloaded apps at a astounding percentage. As of May 2017, Apple has had 140 billion total downloads from its market since its origination in 2016 (Statista, 2017b). In the firstquarter of 2016 alone, Google had an estimated 64 billion downloads and Apple had 100 million downloads (Android Authority, 2017). Mobile applications are designed to extend the capabilities of the mobile devices for mobile device operating systems, which end-user soft ware apps, (Purcell, 2011). Mobile applicatios are defined as software or programmes that are designed to perform specific tasks, which can usually be downloaded onto users’ mobile devices (Kwon et al., 2013; Mozeik et al., 2009; Wang et al., 2015). As well, users can install different kinds of mobile applications, such as game, music, shopping, bank payment applications and so forth, which are delivered by the third-party software providers (Grotnes, 2009; Islam, Islam and Mazumder, 2010; Taylor, Voelker and Pentina, 2011). By the installation of these applications, the functions of the mobile devices are expanded. The number of mobile apps has been rising, and this rise contributed to the arising range of consumer needs that are being served by mobile apps (Kim, Yoon & Han, 2014). Because of the disadvantages of websites due to their restrictive functionality, many companies provide mobile apps to the customers (Magrath and McCormick, 2013). Compared to mobile websites, mobile applications are preferred by consumers primarily because they are perceived as more convenient, faster and easier to browse (Mobile Apps: What Consumers Really Need and Want, 2016). Many recent works that have examined the factors involved in a consumer’s adoption of distinct mobile device services, such as mobile payments (Lu, Yang, Chau and Cao, 2011; Zhou, 2013), mobile banking (Chen, 2013), financial services (Chemingui and Ben lallouna, 2013), health services (Deng, Mo and Liu, 2014), mobile learning (Callum, Jeffrey and Kinshuk, 2014), mobile data services (Al-Debei & Al-Lozi, 2014) and mobile games (Jiang, Peng and Liu, 2015). Regardless, there have not been many works that have assessed the determinants that affect the installation of mobile apps on mobile devices in Turkey context. One of the main drawbacks of conventional statistical techniques used for the prediction of users’ behavior is that they usually examine only linear relations among variables. In order to overcome this issue, relative importance of significant variables will be determined using neural networks, capable to model complex non-linear relationships.

Complete Chapter List

Search this Book:
Reset