RESEARCH ARTICLE Predicting determinants of the intention to use digital currency in the UAE: An empirical study Nasser A. Saif Almuraqab Dubai Business School, University of Dubai, Dubai, UAE Correspondence Nasser A. Saif Almuraqab, Dubai Business School, University of Dubai, Dubai, UAE. Email: Abstract Digital currencies represent a new, innovative means of exchange, utilizing the inter- net to simplify and enhance online transactions, which brings a revolution to the economy. This research investigated the acceptance predictors of digital currency among the people of the United Arab Emirates (UAE). To achieve this goal, I surveyed a sample of 181 UAE residents, aiming to predict the intention to use digital cur- rency. The researcher argues that perceived usefulness, perceived trust, social influ- ence, and perceived ease of use are significant determinants of citizens intention to use digital currency, and perceived ease of use and perceived usefulness mediate the relationship between awareness and intention to use. Also, this paper reveals the fac- tors affecting digital currency acceptance in the UAE, representing differences and similarities regarding global acceptance. The results extend existing models of digital currency acceptance, providing governments, policymakers, and IT professionals with understanding of how acceptance is developed in this context. KEYWORDS Aber, adoption, digital currencies, digital finance, technology acceptance, UAE 1 | INTRODUCTION The growth of virtual worlds within the digital era augments the importance of digital currencies. In fact, currency is a critical item in transactions and in conducting reliable agreements between citizens around the world. The modernization of financial systems and other high-tech progress has boosted the need for digital currencies to enable seamless and real-time financial transactions (Brezo Hoffman, 2012; Krause, 2016). Digital currencies are not widespread in developing countries due Received: 5 September 2019 Revised: 20 October 2019 Accepted: 21 October 2019 DOI: 10.1002/isd2.12125 E J Info Sys Dev Countries. 2019;e12125. 2019 John Wiley 86:e12125. 2019 John Wiley however, regardless of its fame, it remains unclear, and scholars have paid little attention to its acceptance among citizens, which has received the attention of scholars. In tandem with advancing smart government services, the UAE and Saudi Arabia are planning to introduce a joint digital currency or digital currency named “Aber” (Abbas, 2019; Rahman, 2017). Hence, it is important to study and predict the factors that will influence people to adopt this technology, in order to facilitate the adoption process and the implementation process (Almuraqab, 2017). To the best of the authors knowl- edge, there has not been any empirical research in Gulf Cooperation Council (GCC) countries, and especially in the context of the UAE, addressing and focusing on the acceptance of digital currencies. Thus, the purpose of this research is to examine and predict factors that affect individual preferences for using digital currency as a new innovation in the UAE. The findings of this paper demonstrate that perceived usefulness, perceived trust, perceived ease of use, and social influence significantly influence the intention to use digital currency, while perceived ease of use and per- ceived usefulness mediate the association between awareness and citizens intention to use digital currency. First, this paper presents core knowledge based on relevant research and hypotheses, to answer the research question. Then a description of the research methodology is provided, followed by research results, analysis of the results, and then discussion. Finally, it presents important implications and guidelines for academics and professionals. 2 | LITERATURE REVIEW Acceptance of digital currency relies on its features; however, the existing adoption status of this phenomenon is not steady (Shahzad et al., 2018), which negatively affects acceptance (Michael, 2013). In fact, the main cause of the hampering of digital currency adoption was its develop- ment stage and low number of user-friendly aspects (Evan-Pughe, 2012). On the other hand, Neroth (2013) described how the adoption rate of digital currencies, eg, Bitcoin, was constantly rising, because of the participation of traders for the payment of duties directly or through service providers. However, Stark (2013) argued reasons behind the low adoption of Bitcoin included challenging issues for merchants such as the unformalized marketplace, a deficiency of liquidity, a lack of market principles and governmental control, and high-security risks. The same issues had a similar effect on citizens and prevented them from being a part of the new digital economy (Stark, 2013). The importance of digital curren- cies cannot be underestimated, but they are not generally recognized as a type of currency or as a pillar of the financial system. Such new curren- cies must be supported by government as well as financial institutions; scholars (Bhme, Christin, Edelman, Elwell, Murphy, Seitzinger, Wu however, it is less apparent in the context of digital cur- rencies, which have comparable features to authorized currencies or commodities such as gold (Sas Ndubisi, 2006; Pan, Gunter, Sivo, Venkatesh Sultana, Ahlan, Lallmahomed, Lallmahomed, Almuraqab, 2017; Almuraqab, 2019) have used combinations of models and/or additional variables consis- tent with the field and conditions, to examine the adoption rate of particular technologies. In line with these studies, and to answer the research question, this paper based its model on the TAM as a fundamental model, with the addition of three critical factors of acceptance drawn from relevant studies: perceived trust, awareness, and social influence, to explore the acceptance of digital currency as a method of exchange in the UAE. The anticipated relationships and research tool are displayed in Figure 1. The proposed variables definitions and questionnaire items are described in Appendix A. 2.2.1 | Awareness Awareness (AW) of technology (its existence, benefits, and usage) is core to technology diffusion (Hall it provides information about it (Aloudat, Michael, Chen, Shareef, Kumar, Kumar, Jasimuddine et al., 2017). However, Shahzad et al. (2018) reported a direct influence of awareness on inten- tion to use. This paper tries to measure citizens awareness of the use and benefits of digital currency as an alternative currency. So it is hypothesized that H1. Awareness of digital currency will have a positive influence on UAE citizens perceived ease of use of digital currency. FIGURE 1 Research framework SAIF ALMURAQAB 3of12 SAIF ALMURAQAB 3 of 12 H2. Awareness of digital currency will have a positive influence on UAE citizens intention to use digital currency. H3. Awareness of digital currency will have a positive influence on UAE citizens perceived usefulness of digital currency. Perceived ease of use Perceived ease of use (PEOU) is the degree to which a person believes in the ease of use of a technology or system (Davis, 1989). Correspond- ingly, a study by Shahzad et al. (2018) argued that PEOU is a significant factor in adopting Bitcoin. Therefore, this research argues that citizens decisions on adoption of digital currency technology depend on its effortlessness and quick access. Consequently, it is hypothesized that H4. Perceived ease of use of digital currency will have a positive relationship with UAE citizens intention to use digital currency. 2.2.2 | Perceived usefulness Davis (1989) argued that perceived usefulness (PU) is the extent to which an individual believes that the use of a technology or system could be valuable for them and may improve their performance. Several studies (Adams, Nelson, Almuraqab, 2017; Shahzad et al., 2018; Venkatesh, Morris, Davis, Emad and Haider, 2015). The use of digital currencies will also be affected by PU, which will even- tually affect the acceptance of this phenomenon in the UAE. Hence, to add to the Shahzad et al. (2018), it is hypothesized that H5. Perceived usefulness will have a positive influence on UAE citizens intention to use digital currency. 2.2.3 | Social influence In terms of social influence (SI), individuals so far unfamiliar with the digital currency would attempt to try it on the recommendation of a family member/friend. A study by Baur et al. (2015) on Bitcoin adoption argued for the importance of social influence, and how word-of-mouth enhanced and triggered the adoption of digital currency (Bitcoin) as a fiat currency. This is in line with studies (Almuraqab, 2017; Jasimuddin, Mis- hra, Wang, 2008). Lowering of PT will negatively affect acceptance and satisfaction; in other words, it will reduce citizens intention to use a system or technology (Venkatesh, Thong, Chan, Hu, Almuraqab, 2017; Shahzad et al., 2018). This enhances the validity of the instrument and improves the reli- ability. This research extended the TAM model by adding PEOU and PU to the anticipated constructs from the literature, which are awareness, social influence, and trustworthiness. The items distributions are displayed in Table 2, which contains an estimate of items standard deviations and means. This procedure enhanced the content strength and the consistency of the tool. 3.2 | Sampling process The electronic questionnaire was established as follows: First, the researcher used an electronic survey distributed to a limited number of appli- cants (50) to collect a testing set. A reliability check utilizing SmartPLS3 was conducted using Cronbachs alpha (Table 2) reliability coefficient, which confirms the reliability of the questionnaire constructs. Regarding the sample size, the author prearranged to distribute the questionnaire using electronic targeting to collect a minimum of 50 responses if possible, to match the rules and conditions of SmartPLS analysis, which is 10 times the largest number of formative indicators used to measure a single construct, or 10 times the largest number of structural paths directed at a particular construct in the structural model (Hair, Hult, Ringle, moreover, social network messaging was used to distribute the survey to a different segment of citizens. The use of electronic distribution channels makes it impossible to calculate the response rate. After data collec- tion, a primary regression test was used to eliminate outliers that surpassed limits on specific measures or were extremely odd. The final sample size used for research analysis is 181, which satisfies the generalization ability terms and the suitability of examination. Table 1 explains the sam- ple demographics. 4 | ANALYSIS OF RESULTS An analysis was conducted, based on structural equation modelling (SEM) using a partial least squares (SmartPLS) application (Ringle, Wende, and Will, 2005). SEM assesses the measurement and structural models simultaneously, consequently doing factor analysis and hypotheses testing con- currently (Gefen, Straub, and Boudreau, 2000). Instead of covariance-based SEM, PLS was used since it is primarily appropriate for exploratory study (Gefen, Karahanna, these values are dis- played in Table 2. It is essential to test the correlation matrix, before responding to the research question. To start with variable discriminant validity: Fornell and Larcker (1981) specified that the square root of AVE must be higher for every construct than its correlation with other constructs. The inter- construct correlation matrix is presented in Table 3 with the square root of AVE displayed in bold. TABLE 2 Variables of internal validity figures, means, standard deviations, and item measurement loadings Item Mean SD Measurement Item Outer Loadings ITUAW PEOU PU SI PT AW1 3.52 0.91 0.82 0.22 0.30 0.24 0.38 0.35 AW2 3.72 0.82 0.85 0.25 0.43 0.28 0.37 0.35 AW3 2.82 0.96 0.73 0.30 0.39 0.35 0.42 0.42 AW4 3.41 0.99 0.86 0.27 0.42 0.23 0.36 0.35 PEOU1 3.22 0.86 0.24 0.84 0.54 0.31 0.55 0.56 PEOU2 3.46 0.78 0.33 0.90 0.61 0.28 0.56 0.58 PEOU3 3.38 0.77 0.17 0.86 0.50 0.25 0.48 0.47 PEOU4 3.33 0.82 0.33 0.86 0.60 0.23 0.45 0.53 PU1 3.43 0.81 0.49 0.50 0.81 0.35 0.58 0.62 PU2 3.37 0.88 0.38 0.58 0.82 0.34 0.53 0.60 PU3 3.14 0.79 0.28 0.44 0.74 0.41 0.47 0.52 PU4 3.56 0.78 0.35 0.57 0.81 0.29 0.45 0.58 SI1 2.91 0.84 0.33 0.27 0.38 0.90 0.42 0.48 SI2 2.85 0.85 0.27 0.30 0.41 0.91 0.36 0.45 SI3 3.04 0.85 0.33 0.27 0.38 0.90 0.40 0.52 PT1 3.25 0.86 0.41 0.58 0.64 0.44 0.91 0.69 PT2 3.27 0.78 0.40 0.51 0.50 0.28 0.72 0.55 PT3 4.13 0.91 0.42 0.49 0.56 0.37 0.92 0.66 PT4 4.16 0.82 0.43 0.50 0.52 0.38 0.92 0.65 PT5 3.05 0.95 0.38 0.48 0.56 0.40 0.85 0.59 ITU1 3.45 0.85 0.40 0.54 0.66 0.45 0.63 0.90 ITU2 3.18 0.89 0.45 0.58 0.67 0.53 0.68 0.92 ITU3 3.21 0.93 0.38 0.54 0.68 0.54 0.69 0.92 ITU4 3.55 0.87 0.39 0.58 0.61 0.40 0.61 0.85 Cronbachs alpha 0.83 0.89 0.81 0.89 0.92 0.92 Composite reliability 0.89 0.92 0.87 0.93 0.94 0.94 AVE 0.67 0.75 0.63 0.82 0.75 0.81 TABLE 3 Discriminant validity AW ITU PEOU PT PU SI AW 0.82 ITU 0.45 0.90 PEOU 0.32 0.62 0.87 PT 0.47 0.73 0.59 0.87 PU 0.48 0.73 0.66 0.64 0.80 SI 0.34 0.54 0.31 0.43 0.43 0.90 6of12 SAIF ALMURAQAB 6of12 SAIF ALMURAQAB In addition to the outcomes shown, and to respond to the research question, an SEM test using the SmartPLS3 application and its algorithms was engaged. In fact, PLS-SEM depends on a nonparametric bootstrapping method to test the research model; in other words, many smaller sub- samples are inherited from the main sample and verified to reach the best fit of the model. The SmartPLS3 tool is licensed for use by the Univer- sity of Dubai and easily estimates the items loadings and the relationships. The findings are displayed in Table 3. A SmartPLS tool computes both t and P values for each hypothesis to show the significance of results, as displayed in Table 4. The bootstrapping tool is conducted to calculate the t values of the model. The verified model expressively clarifies ITU with an R 2 value equal to 0.69. Also, the beta values (standardized coefficients of the factors) and their “t” values and “P” values are presented in Figure 2. 5 | DISCUSSION This research aimed to predict, and to improve our understanding of, the factors influencing UAE citizens intention to use the government- supported digital currency. The proposed model comprised six major constructs that are common in the technology adoption field. The first TABLE 4 t values and P values with hypothesis status Hypotheses Path Coefficient t Value P Value Hypothesis Status AW PEOU 0.32 4.62 .00 * Supported AW PU 0.48 6.71 .00 * Supported AW ITU 0.02 0.45 .66 Not supported PU ITU 0.33 4.34 .00 * Supported PEOU ITU 0.14 2.13 .03* Supported PT c ITU 0.35 4.88 .00 * Supported SI ITU 0.20 3.54 .00 * Supported * P .05. * P .01. FIGURE 2 Hypotheses b and P values SAIF ALMURAQAB 7of12 SAIF ALMURAQAB 7of12 construct, ITU, is a well-investigated variable that is used to explain tec