Rural Technology Adoption and Use Model in Rural Africa: A Predictive Approach to Telephony Acceptance

Introduction

Today, wireless telephony is an indispensable technology for humankind. More than a tool for communication and information sharing, it brings many opportunities and shows the potential to change human life in many areas. With the development of information and communication technologies and the recent growth of the mobile telecommunications sector, there has been an unprecedented increase in digital inclusion in low- and middle-income countries. However, mobile phone networks and users’ subscriptions are far from common (GSMA, 2019). The digital gap is a reality related to the coverage gap, coming from the difficulties and challenges operators encounter in extending their networks in some areas.

In many cases, even when the deployment challenge has been overcome with lightweight and low-cost technologies, the users’ subscription to the services is not guaranteed to ensure a return on investment. Thus the usage gap, which threatens the viability of the network. In 2019, according to GSMA reports, the usage gap problem was estimated to affect 44% of the world's population, i.e., people who have access to technology but do not use it (GSMA, 2020). Especially in rural areas, the telephony market is critical and unprofitable because the usage gap is even higher in these areas. With the usage gap, the rural telephony market is mainly a market with low purchasing potential. It is then relevant to explore the factors that make people in rural areas that are not yet covered adopt and use the future networks that will be put in place to bridge the coverage gap. This will help anticipate the needed policies to ensure a better subscription to the planned services. That's the purpose of this study, which uses a methodological approach based on Structural Equation Modelling (SEM). This approach makes it possible to evaluate a set of causal relationships between several factors. It is widely applied as a popular method of data analysis (Birgit et al., 2019; Joseph F. Hair et al., 2019; Sabiu et al., 2019; M Sarstedt et al., 2019; Umrani et al., 2017) and is well suited for predictive applications(Chin, 1998). Several studies evaluated the acceptance of technologies, and several of them addressed mobile telephony, but few studies have dealt with the particular context of unconnected rural areas. Determining the factors that encourage technology adoption in this context is of significant interest to the operators' strategic plan for extending its network to those areas. For this reason, this article is structured around the following questions:

• •

  What are the fundamental elements or factors that can play a role in encouraging the adoption of telephony in rural areas?

• •

  By ensuring that the inhabitants adopt the technology, we also ensure the real use in the future?

• •

  What are the factors that determine the use of technology in these environments?

Based on the existing literature, in a methodological approach, this paper proposes a model for adoption based on two widely used models, TAM and UTAUT. By offering a predictive analysis of future users' behavior of a yet to be deployed network, this study makes an essential contribution to enriching the existing literature. First, in this paper, we present the theoretical development of our research model. Then, we describe the research methodology and give the results. Finally, the results are analyzed and discussed, along with their implications and future work.

Theoretical Development

Modeling techniques are commonly adopted in empirical work in many research works. As structural equation models, these techniques are, for the most part, approaches based on statistical methods that allow testing hypotheses dealing with the relationships between observed and latent variables. The partial least squares method is particularly interesting because of its rather confirmatory predictive nature (Mourre, 2013). In this respect, they are more advantageous than the various traditional analysis methods such as simple and multiple regressions, simple correlation tests, and canonical analyses.

This study considers two theories based on structural equation modeling, from recent information and communication technologies literature, widely used as the main theoretical framework for understanding and explaining adoption and use behavior.