key: cord-0059913-bca5keoz
authors: Ntsiful, Alex; Kwarteng, Michael Adu; Jibril, Abdul Bashiru; Popesko, Boris; Pilik, Michal
title: Factors Driving the Adoption of Mobile Banking App: An Empirical Assessment in the Less Digitalized Economy
date: 2020-11-10
journal: Re-imagining Diffusion and Adoption of Information Technology and Systems: A Continuing Conversation
DOI: 10.1007/978-3-030-64861-9_11
sha: 52820f8e382afb32dcb745ee85195777fd8117ec
doc_id: 59913
cord_uid: bca5keoz

Financial institutions remain one of the key industries making use of information and communication technologies to transform their products and services and their business in general. As the Internet becomes pervasive, most banks, in addition to their traditional mobile banking, have introduced mobile banking apps, which ride on smart devices and the Internet to offer banking services remotely to customers. However, given some contextual factors such as unstable Internet, adoption of this innovation may be a challenge in an emerging economy. Thus, the study sought to find out the factors, which could affect the adoption of mobile banking app in a less digitalized environment. To this end, data was sourced via the intercept approach on a sample of the entire population in Ghana for the analysis. By using the partial least square structural equation modelling (PLS-SEM) technique, the study found that performance expectancy and hedonic motivation were the key factors that influence mobile banking app adoption. Contrary to our expectation, effort expectancy, perceived transaction cost, and privacy and information concerns were found to have no effect on consumers’ intention to adopt mobile banking app. Implications for further research and practice are presented and discussed.

Financial institutions remain one of the key industries making use of information and communication technologies, such as the Internet and wireless technologies to transform their products and services and their business in general. With and through mobile technologies, banks can meet customers' expectations by offering them different services (Singh and Srivastava 2018) . The banks started with the introduction of mobile banking, which according to Barnes and Corbitt (2003) can be defined as the channel allowing customers to use their digital devices to interact with the bank. As the Internet becomes more pervasive, the banks have now developed their applications to augment the mobile banking service. Unlike mobile banking which can be done without smartphone and Internet, services offered by the bank on mobile banking app can only be accessed by smartphones with the Internet. To this end, the mobile banking app is less useful in developing economies where the Internet is a challenge. For instance, research shows that in rural India where there is less access to the Internet, mobile banking is easily accessed, and remains the preferred channel of banking services (Singh and Srivastava 2018) . Whilst studies on mobile banking abound (see Chhonker et al. 2018; Grover and Kar 2020; Singh et al. 2017 ) research on mobile banking apps remain scant. Moreover, the few studies we have on mobile banking apps have also been conducted in Internet-endowed economies (e.g. Munoz-Leiva et al. 2017; Wijland et al. 2016 ); with few in developing economies such as Ghana (e.g. Ansong and Synaepa-Addision 2019) .

In studying the adoption of the predecessor of mobile banking apps, Owusu et al. (2020) note that although there is a high rate of mobile device penetration in Africa (ITU 2007) , the diffusion of mobile banking technologies is far below expectation. We thus argue that there are myriad of factors in developing economies, which make direct applications of the previous findings contextually irrelevant in those economies. For instance, in a less digitized environment such as Ghana where consistent pervasive stable Internet is a dream and even in the capital city, Accra, where one expects to have a stable Internet in place, is also not the case. Thus, we argue that an introduction of a mobile banking app to attract a massive adoption may be a mirage. We also contend beyond downloading of the mobile banking app, customers in such growing economies would have other issues such as privacy/cybersecurity or perceived transaction cost, as the modalities and how the Internet of things work are not popular in these environments. In other words, the study argues that the issues on mobile banking apps are not well researched and empirically tested. As these apps are at their infancy stage, there is the need to examine thoroughly to know which issues could influence customers' adoption behaviour. Following these arguments, the key objective of the present study is to address the factors or issues, which could affect customers' adoption of mobile banking app in Ghana. To achieve this objective, the study uses the modified version of the unified theory of acceptance and use of technology (UTAUT2) (Venkatesh et al. 2012) . Specifically, we examine the relationships between performance expectancy, effort expectancy, and hedonic motivation (UTAUT2), and intention to use mobile banking apps. Further, the study also adds perceived transaction cost and privacy and information concern to the constructs of the said theories in examining the relationships. We make significant contributions to literature with this paper. First, we extend UTAUT2 with two additional constructs to help understand the issues customers may have in adopting mobile banking apps. Second, contextually, we add to the scanty of research output on mobile banking app adoption in a less Internet-penetrated territory. These, we believe, would offer insights to managers and policymakers in organizations especially those in the financial sector as a guide in the development and introduction of their banking apps on the markets.

We organize the rest of the paper as follows: While the next Sect. 2 presents the theoretical background and hypotheses development, Sect. 3 explains our empirical study. This is followed by Sects. 4, which presents the discussion and conclusion of the study.

The theory from which our study takes inspiration is the modified version of the unified theory of acceptance and use of technology (UTAUT2) by Venkatesh (2012) . As the name suggests, UTAUT is an integration of eight information system theories and models. The first four of these theories and models include the theory of reasoned action-TRA (Fishbein and Ajzen 1975) , social cognitive theory-SCT (Bandra 1986), technology acceptance model-TAM (Davis 1989) , and theory of planned beha-viourTPB (Ajzen 1991) . The other four theories/models include the model of PC utilization (MPCU-(Thomsposn et al. 1991) ), the motivational model-MM (Davis et al. 1992) , the decomposed theory of planned behaviour-DTPB (Taylor and Todd 1995) and the innovation diffusion theory-IDT (Rogers 1995) . Thus, the first UTAUT (Venkatesh et al. 2003) posits that for factors such as performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating conditions (FC) are the key determinants of a person's intention to adopt a technology. This means for a person to use a certain technology, the technology should meet performance expectations, and should not require excessive effort to use. The theory also acknowledges that social norms or influence from other people the individual value and also certain conditions such as technical and human support can affect the person's intention to use the technology. Thus, PE, EE, SI, and FC were the key four constructs of UTAUT. In his study, Venkatesh moderated these variables with age, experience, gender, and voluntariness of use.

However, in 2012, Venkatesh and his colleagues modified the UTAUT to UTAUT2 by including new constructs: price value, habit, and hedonic motivation, and dropped the voluntariness of use as a moderator and concluded that UTAUT2 better explain consumer intention. We also design our present study in line with the UTAUT2 but specifically adopt only the performance expectation (PE), effort expectancy (EE), and hedonic motivation (HM). We also extend the UTAUT2 with perceived transaction cost (PTC), and privacy and information concern (PIC). We argue that in an environment such as Ghana, as mobile banking apps use the Internet, customers may have problem Internet and transaction costs. We further contend that in such an environment where technology acceptance is at infancy, customers could be more sceptical over cybersecurity issues regarding personal information if they should use the mobile banking app, which uses the Internet. To this end, we explain performance expectation, effort expectancy, hedonic motivation, perceived transaction cost, and privacy and information concern, as well as our dependent variable, intention to adopt mobile banking app in details as follows: We summarize previous works on our study's constructs in Table 1 below: (Venkatesh et al. 2003) Extant studies have found that PE affects users behavioural intention to adopt new technology (see Shareef et al. 2018; Alalwan et al. 2017; Venkatesh et al. 2003) . For instance, Alalwan et al. 2017 found that performance expectancy has a positive impact on customers behaviour intention to mobile food ordering apps in Jordan Effort expectancy (EE) EE is the degree to which users of a technology perceive it to be easy to use. Thus, in this study, we define effort expectancy as the extent to which consumers perceive mobile banking apps to easy to use (Venkatesh et al. (2012) Several studies have confirmed the relationship between EE and customers' intention to adopt new technology (Thusi and Maduku 2020; Albashrawi et al. 2017; Alalwan et al. 2017) . For example, Albashrawi et al. (2017) noted that EE influence customers intention in using mobile banking app in US banks Hedonic motivation (HM) HM refers to the pleasure gained from using an innovation (Venkatesh et al. 2012) HM has been found as a good predictor of behavioural intention to use innovation (Thusi and Maduku 2020; Okumus et al. 2018; Yeo et al. 2017) . For instance, Okumus et al. (2018) found that HM positively affects customers' intention to adopt smartphone diet apps to order food at restaurants Perceived transaction cost (PTC) Perceived transaction cost refers to the perception that customers hold that using innovation or new technology will attract financial charges. Mobile banking app costs may include initial purchase price, transaction cost, subscription charges (Singh and Srivastava 2018) Overall, PTC is a major obstacle to the adoption of mobile technology (Lubua and Following the theoretical review and the previous empirical works summarized in Table 1 above, we argue that the three constructs of UTAUT2(performance expectancy, effort expectancy, hedonic motivation) plus perceived transaction cost, and privacy and information concern will directly influence usage intention to use the mobile app (our dependent variable) and are therefore conceptualized as independent variables. Accordingly, we derive the following hypotheses: The methodology used to test the proposed model is outlined in the following section.

The study made use of both probability and non-probability sampling techniques for data collection. In effect, the non-probability sampling technique was adopted to select the target banks (unit of our analysis) comprising universal, rural, investment management firms/banks. On the other hand, the probability sampling technique (random sampling-which we consider to be most practical in this case) was then adopted via the intercept approach to gathering data for our analysis. Initially, we aimed to recruit 600 participants for the study. However, the influx of the COVID-19 pandemic did not allow us, therefore we ended up getting 327 respondents of which 291 was used for our analysis, after cleaning the data. We must emphasize that the first author was on the ground for data collection and recruited research assistants from his previous 

Definition Some findings Privacy and information concern (PIC) PIC denotes the extent to which a user of technology feels sceptical about personal data Prior studies have found that PIC negatively affects technology adoption (Hayikader, et al. 2016; Tang and Liu 2015; Alkhater et al. 2014; Gupta et al. 2013) Intention to use a mobile banking app An intention is defined as the plan of action to engage in a certain behaviour (Armitage and Christian 2003) . In this study, the intention to use is conceptualized as a dependent variable and is influenced by PE, EE, HN, PTC, and PIC University to help gather the entire data (A token of reward from the funds of our research project was given to these research assistants). Data collection was undertaken in the months between November 2019 and January 2020. On average, the questionnaire took seven minutes to fill.

The mainstream of sampled respondents was male (65%), aged between 21-35 (39.6%), and a bachelor's degree (63.2%). In this study, statistical analyses were performed using both IBM SPSS and ADANCO 2.0 version (Henseler et al. 2014) . Lastly, the research constructsexcept for demographic variables -were measured using a five-point scale (ranging from completely disagree to completely agree). Besides, our work is consistent with most of the views expressed in Podskaoff et al. (2003) regarding the minimization of common method bias.

By way of improving the face and construct validity, we curled the study's constructs from the existing literature. Specifically, the measures for Performance Expectancy (PE), Effort Expectancy (EE), Hedonic Motivation (HM) and Intention to adopt mobile banking app (INT) were adapted from Farah et al. (2018) , while the measures, Perceived transaction cost (PTC) and Privacy/information concern (PIC) were modified from Hanafizadeh et al. (2014) and Albashrawi and Motiwalla (2019) respectively. Moreover, the research hypotheses were tested by using the PLS-path modelling technique and precisely using mode A algorithm for path analysis while mode B was used to measure control variables on the dependent variable. The inspected composite reliability scores pertaining to the reflective measurement model were recorded as follows Fornell and Larcker (1981) , discriminant validity was documented for the constructs.

Model fit and quality criteria were assessed based on Standardized Root Mean Squared Residual (SRMR). We obtained 0.0562 (SRMR value and thus acceptable since it is less than 0.1) (Henseler et al. 2014) . The SRMR is calculated as the square root of the sum of the squared differences between the model-implied correlation matrix and the empirical correlation matrix, i.e. the Euclidean distance between the two matrices. Hence, the lower the SRMR, the better the theoretical model's fit. As a rule of thumb, a value of 0 for the SRMR indicates a perfect fit, and, mostly, an SRMR value less than 0.05 indicates an acceptable fit (Byrne 2013). However, a recent replication study shows that even totally correctly specified models can yield SRMR values of 0.06 and higher (Henseler et al. 2014) . Therefore, a baseline value of 0.0562, as proposed by Hu and Bentler (1999) , appears to be better for variance-based SEM in our case depicting a model fit. Table 2 shows an empirical analysis of the proposed model of this study. We performed a structural path analysis using a statistical software: ADANCO 2.0 version as earlier disclosed. Reflecting on the recommendation of pioneer scholars (Hair et al. 2014; Henseler et al. 2016) , the summary of our PLS-SEM (see Table 2 ) indicates direct path analysis with control variables as well as the estimated coefficient of determination (R 2 ). Concerning the hypothesis testing, we found that hypotheses: (HI and H3) proved significant at p-value < 0.05 whilst the remaining direct hypotheses; (H2, H4, H5) were insignificant. Again, in testing the control variables: gender, age, and experience of the study's respondents, it was found that 'Gender' and 'Age' of respondents played a significant control variable regarding the use of the e-banking app. Further, the value of the coefficient of determination (R 2 ) (see Table 2 ) indicates that our endogenous variables (e.g. Performance expectancy, hedonic motivation etc.) explain significant variance in the exogenous variable (intent to adopt banking app). This suggests that the model satisfied beyond the appreciable threshold of 50% (see Table 2 and Fig. 1 ). 

We took inspiration from the modified version of the unified theory of acceptance and use of technology (UTAUT2) by Venkatesh (2012) to investigate thoroughly and to know which issues could influence customers' adoption behaviour of the use of mobile banking app in a less digitized country as earlier unveiled. Hence, the key objective of the present study was to address the factors or issues, which could affect the customer's adoption of the mobile banking app in Ghana. The research model based on the UTAUT2 was tested with a sample of the population in Ghana using an intercept approach via a cross-sectional study. The measurement and structural models recorded good model fit metrics, and the study measures possessed sufficient validity and reliability.

RQ1 investigated which issues could influence customers' adoption of mobile banking app in Ghana. For this research question, it is worth noting that the research results revealed that potential users (or users) of the mobile banking app within the preview of the less digital economy are much more passionate about the performance expectancy (PE) and the hedonic motivation (HM) attached to the use of the mobile banking app. This assertion further corroborates with the recent works of Abdul-Hamid et al. (2019) and Boateng et al. (2016) who examined the adoption of mobile money apps, and Internet banking adoption in a developing country respectively. Similarly, it is important to note that the current empirical findings revealed 'age and gender' variations as key control variables, which significantly predict the user's intention to adopt the banking app.

RQ2 examined the relationship between UTAUT2 factors and the intentions to continue using the mobile banking app in Ghana. In answering this question, the current analysis confirmed PE and HM but not EE as the two major UTAUT2 factors that have a positive influence towards the intention to adopt the said app amongst potential users in the less digitalized economy (Alalwan et al. 2017) . Again, in evaluating the impact of Privacy/information concern (PIC) on the intent to adopt the banking app, the former showed an inverse relation towards the latter Owusu et al. 2020) . Perceived transaction cost (PTC) has no significant impact per the responses evaluated. This further suggests that any additional charges that may arise in an attempt to use the said app would not deter the potential users from using the service (i.e. mobile banking app). Hence the result is consistent with the research of (Oliveira et al. 2014; Owusu et al. 2020) .

Theoretically, the study's findings advance the literature on UTUAT in two ways: First, we utilized an abreast version of the technology adoption model to examine the influence of UTUAT2 on the intentions to adopt mobile banking app, while introducing additional two constructs to measure the intention and adoption of the app. This enabled a valued contribution to the evolving literature on the relationship between adoption/intentions, which has remained underexplored thus far. This is in line with the consistent recommendation of recent literature that has called for urgent steps to be taken to recognize the factors driving the adoption of online banking transactions in the developing world Nwaiwu et al. 2020) . Second, this research further informs scholars on the need to re-consider an extensive exploration of technology studies geared toward technology adoption/resistance in the less digitalized economies so far as digital divide continues to perpetuate between developed and developing worlds.

Practically, this study has two significant implications for stakeholders and shareholders in the financial sector in the developing country. First, the study findings could be of particular relevance to those emerging banks or lesser banks, which are yet to transition to the online banking system. Second, the study offers relevant information and knowledge to bank customers on the need to make use of banking technologies (apps) since the latter performs several functions to the user including enhancing service delivery, quick bank transaction, ability to use the app at the comfort of your home, and among others.

The present study has some limitations, which need to be kept in mind. First, the study participants were selected from a single country, namely Ghana. Therefore, the study could not have been generalised to reflect the entire economies that are described as less digitalized economies. However, despite these limitations, the study makes notable contributions to the existing literature. Hence, we recommend the following possible future directions for future studies. First, scholars could use our model to study customer behaviour in other contexts because the mobile banking app has come to stay. Lastly, scholars should explore the causal relationships of different theory as explained earlier with the adoption or resistance intention through longitudinal investigations.

Customers' perceived risk and trust in using mobile money services-an empirical study of Ghana

The theory of planned behavior

Factors influencing adoption of mobile banking by Jordanian bank customers: extending UTAUT2 with trust

Privacy and personalization in continued usage intention of mobile banking: an integrative perspective

The impact of subjective and objective experience on mobile banking usage: an analytical approach

Factors influencing an organisation's intention to adopt Cloud computing in Saudi Arabia

A comparative study of user data security and privacy in native and cross platform Android mobile banking applications

From attitudes to behaviour: basic and applied research on the theory of planned behaviour

Social Foundations of Thought and Action

Mobile banking: concept and potential

Assessing the determinants of Internet banking adoption intentions: a social cognitive theory perspective

Structural Equation Modeling with LISREL, PRELIS, and SIMPLIS: Basic Concepts, Applications, and Programming

m-commerce technology adoption

Perceived usefulness, perceived ease of use, and user acceptance of information technology

Extrinsic and intrinsic motivation to use computers in the workplace 1

Partial least squares structural equation modeling (PLS-SEM) an emerging tool in business research

Mobile-banking adoption: empirical evidence from the banking sector in Pakistan

Belief, Attitude, Intention, and Behavior

Evaluating structural equation models with unobservable variables and measurement error

User engagement for mobile payment service providers-introducing the social media engagement model

The usage and adoption of cloud computing by small and medium businesses

Mobile-banking adoption by Iranian bank clients

Issues and security measures of mobile banking Apps

Using PLS path modeling in new technology research: updated guidelines

A new criterion for assessing discriminant validity in variance-based structural equation modeling

Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives

Telecommunication/ICT market trends in Africa

Customers' perception of cybersecurity threats toward e-banking adoption and retention: a conceptual study

The role of the transaction assurance, perceived cost and the perceived innovation in the decision to continue using mobile money services among small business owners

Determinants of intention to use the mobile banking apps: an extension of the classic TAM model

Impact of security and trust as factors that influence the adoption and use of digital technologies that generate, collect and transmit user data

Psychological factors influencing customers' acceptance of smartphone diet apps when ordering food at restaurants

Extending the understanding of mobile banking adoption: when UTAUT meets TTF and ITM

Mobile banking adoption among the Ghanaian youth

Common method biases in behavioral research: a critical review of the literature and recommended remedies

Diffusion of Innovations

Consumer adoption of mobile banking services: an empirical examination of factors according to adoption stages

Quality in mobile payment service in India

Predicting the intention to use mobile banking in India

Selecting a trusted cloud service provider for your SaaS program

Understanding information technology usage: a test of competing models

Personal computing: toward a conceptual model of utilization

South African millennials' acceptance and use of retail mobile banking apps: an integrated perspective

User acceptance of information technology: toward a unified view

Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology

Mobile nudging: youth engagement with banking apps

Consumer experiences, attitude and behavioral intention toward online food delivery (OFD) services

Acknowledgement. This work was supported by the research project NPU I no. MSMT-7778/2019 RVO -Digital Transformation and its Impact on Customer Behaviour and Business Processes in Traditional and Online markets, and IGA/FAME/2020/002 and IGA/FAME/2019/ 008 of Tomas Bata University in Zlin.