The objective of this paper is to investigate the factors influencing the intention to use mobile payment methods. To achieve this, a literature review was conducted to identify the significant factors affecting behavioural intentions (BI). The proposed set of hypotheses based on the research model is discussed below. Effort expectancy refers to the perceived ease of use associated with a particular technology [5] . When an individual perceives using a technology as easy, they are more likely to use it [24] . Prior studies have demonstrated that customers’ belief that using mobile payment methods is effortless strengthens their intentions to use them [4] .

Therefore, the following hypothesis is proposed:

H1: Effort expectancy (EE) has a positive effect on customers’ intentions

(BI) to use mobile payment methods. (1)

The concept of Performance Expectancy refers to an individual’s perception of the degree to which a technology can enhance their performance [5] . If a user

perceives that a technology will improve their performance, they are more likely to use it [25] . The existing literature on mobile payment acceptance suggests that users are more likely to adopt mobile payment if they believe it will be helpful [4] . Therefore, the following hypothesis is proposed:

H2: Performance Expectancy (PE) positively influences customers’ intentions

(BI) to use mobile payment. (2)

Social Influence is the pressure exerted on an individual by their social network to use or not to use an innovation [5] . This factor has been shown to be a significant predictor of technology acceptance in various contexts. In the area of mobile payment acceptance, research confirms that social pressure affects a user’s willingness to use mobile payment systems [4] . Thus, the following hypothesis is proposed:

H3: Social Influence (SI) positively influences customers’ intentions

(BI) to use mobile payment. (3)

Trust in the online environment refers to a customer’s willingness to be vulnerable to a vendor after evaluating the vendor’s characteristics, who is expected to provide an agreed-upon service [26] . Trust influences a customer’s likelihood of accepting a given technology [27] . In the mobile payment context, customers’ trust in the security and reliability of mobile payment systems is expected to drive their acceptance of it [28] [29] . Thus, the following hypothesis is proposed:

H4: Trust positively influences customers’ intentions

(BI) to use mobile payment. (4)

Network externalities occur when the benefits of using a product increase as more people use it [30] . The network externality effect, as identified by Van Hove, is a significant factor affecting payment systems. In the context of mobile payments, customers’ adoption decisions are influenced by the number of users of a technology [31] , with a large customer base considered a necessary condition for adoption. Moreover, the availability of mobile payment services at a greater number of merchants also increases customers’ willingness to adopt them [17] . Additionally, the more merchants that offer mobile payment services, the more willing customers are to adopt them [32] . Thus, the following hypothesis is proposed:

H5: The number of mobile payment users (merchants and customers)

positively influences customers’ intentions to use mobile payment. (5)

The model depicted in Figure 1 was utilized to employ the UTAUT model to investigate the acceptance of mobile payment technology. This includes the impact of age and gender on the four fundamental factors of the UTAUT model, including effort expectancy, performance expectancy, facilitating conditions, and social influence, and found that all four factors significantly influenced the acceptance of mobile payment technology in the specific region. This shows the importance of considering demographic factors such as age and gender when evaluating the acceptance and usage of new technologies across different countries [33] .

This study utilized a questionnaire to measure the proposed constructs in the developed model. To ensure the questionnaire’s validity, the items were adapted from existing literature and customized to fit the context of mobile payment technology. The questionnaire utilized a 5-point Likert scale ranging from strongly disagrees to strongly agree. As the original questionnaire was in the Arabic language (refer to Appendix Table A1), it was necessary to translate it into English for use in Nigeria. To ensure the translation’s accuracy, back translation was employed [34] . Furthermore, to guarantee that the translated questions were culturally appropriate and easily understood by the specific population being studied, the questionnaire underwent back translation [35] . The final translated questionnaire was then distributed to the survey participants.

Recruiting participants for a study in Nigeria can be challenging and is influenced by various factors such as research design and target population. Similarly, this study adopted a meticulous approach to recruit participants and distribute questionnaires to ensure a representative sample that accurately reflects the research objectives. Recruitment methods used include personal distribution, contact persons, or online survey platforms depending on specific factors.

As in previous studies, participation was voluntary, and incentives were used to encourage participation. The target sample size was 300, with most surveys administered in person. To aid in data collection, six contact individuals consisting of secondary school teachers, university undergraduates, graduate students, and researchers themselves were utilized. The research team randomly selected participants in natural settings such as schools and assessed their familiarity with mobile payments before administering the survey. The survey was distributed between July 4^th and 15^th, 2022, with no incentives given to participants. Although the target sample size was not achieved, 258 out of 280 surveys were retrieved, and 253 were considered usable after a preliminary visual assessment.

Despite some deviation from distribution percentages, the study participants were considered representative of the Nigerian population, including individuals from various age groups, educational levels, and average income levels. Ensuring a representative sample and the validity and reliability of survey instruments are essential in producing trustworthy and generalizable research findings, despite any differences between previous studies and research in Nigeria.

1) Effort expectancy is a crucial variable in UTAT as it influences the likelihood of technology adoption and use. Designers and developers should prioritize making the technology easy to use and minimizing the perceived effort required to encourage adoption.

2) Social influence plays a significant role in UTAT, as it can positively or negatively affect technology adoption and use. Incorporating social elements into the technology can encourage positive social influence and increase the likelihood of adoption by leveraging the perception of others using and benefiting from the technology.

3) Performance expectancy is another important variable in UTAT, as individuals need to believe that using the technology will lead to improved performance. Designers and developers should focus on demonstrating the benefits and value of the technology to enhance performance expectancy and increase the likelihood of adoption.

4) Trust is a critical factor in technology adoption and use. Building trust in the technology by ensuring its security, reliability, and providing transparent information is essential to encourage individuals to adopt and use the technology.

5) Network externalities, which stem from the benefits derived from the number of people using the technology, are influential in UTAT. Increasing the network externalities of the technology can enhance its value and subsequently drive adoption and use. Promoting the technology through various channels, such as social media, can increase its visibility and attract more users, thereby increasing network externalities.

To assess the potential impact of outlier data on the results, the study conducted an initial multiple regression test on 253 responses, after excluding two cases based on case-wise diagnostics Table 1 displays the demographic characteristics of the sample, including information on gender, age, education, and income. The sample comprises more females, middle-aged individuals, bachelor’s degree holders, and middle-income subjects. The second preliminary analysis examined the correlations between independent variables, which can affect the magnitude and direction of betas in the regression equation.

The analysis involved two tests, inspection of the correlation matrix and collinearity statistics. To test for collinearity, the study computed tolerance and variance inflation factor (VIF), as presented in Table 2. Tolerance values measure how much a variable influences other independent variable, while VIF values measure how much the variance of each regression coefficient increases when independent variables are correlated. The study concluded that there were no issues with multicollinearity, as all tolerance and VIF values in Table 2 were within acceptable limits. The internal consistency between the survey items measuring each independent variable and the dependent variable was assessed using Cronbach’s α. In social sciences, alpha values greater than 0.6 are considered acceptable for reliability. Table 3 reports the reliability measures, which indicate that the survey items used to measure each variable were highly consistent.

A confirmatory factor analysis was conducted to comprehend how the variables measure the factors they signify. The outcomes of this analysis, presented in Table 4, include the factor loadings, construct reliability, average variance extracted, maximum shared variance, and average shared variance. Factor loadings exceeding 0.5 are deemed significant, and except for Q14, all factors exhibit significant loadings.

The construct reliability (CR) and average variance extracted (AVE) measures suggest that all the model constructs possess reliable and convergent features. Model constructs with CR values above 0.7 are deemed reliable, and all the model constructs meet this standard. Furthermore, constructs with an AVE greater than 0.5 indicate that the construct’s variance is more significant than the variance caused by error, and all the model constructs satisfy this requirement. Additionally, the results reveal that the factors within each construct are not

¹Item #14 has been deleted because of the result analysis.

excessively correlated, indicating a good level of convergent validity. Discriminant validity occurs when the factors measuring each construct have more correlations with each other than with other factors. To establish discriminant validity, the average shared variance (ASE) and maximum shared variance (MSV) must be lower than the AVE. In this study, the measurement model fulfils these criteria.

Before testing the hypotheses, it is crucial to examine the rationale behind the variables. Table 5 presents the correlation matrix showing the significant relationships between the independent variables, namely performance expectancy, effort expectancy, social influence, trust, and network externalities, and the dependent variable, behavioural intentions. Performance expectancy and network externalities have the most considerable bivariate impact on behavioral intentions among these variables. All the correlation values in Table 5 are acceptable, indicating a significant correlation between behavioral intentions and each independent variable.

To assess the suitability of the research model, we utilized AMOS 20 and evaluated the values of chi-square, degree of freedom, CFI, and RMSEA, which are shown in Table 6 along with their acceptable thresholds. All reported values fall within the acceptable range, indicating a good fit for the research model. The model is capable of predicting behavioral intentions, with the exception of effort expectancy, and has an explanatory power of 58%. Table 7 shows the standardized

**At a level of 0.01 denoted as (2-tailed) the Correlation is significant.

beta coefficients for the estimated relationships, as well as the results of hypothesis testing. The prediction equation is provided below:

BI = 0.35 PE + 0.13 SI + 0.22 T + 0.34 NE + error (6)

The results of the research model were largely positive, with the exception of effort expectancy (H1), which did not have a significant impact on customers’ behavioral intentions towards mobile payment, despite a significant correlation in the bivariate relationship (as indicated in Table 5). This implies that while ease of use is somewhat important to Nigerian customers, its importance is diminished by the high penetration and daily use of mobile phones for this relatively new mobile-based technology. This finding is consistent with previous studies but differs from some other research [36] [37] .

In contrast, customers’ acceptance of mobile payment was significantly predicted by both performance expectancy (H2) and social influence (H3). The findings suggest that Nigerian customers value the potential benefits of mobile payment and are influenced by the opinions of others in their social circle, which is consistent with prior studies. Trust (H4) and network externalities (H5), two other variables included in the research model, were also found to be significant predictors of mobile payment acceptance. For customers to conduct financial transactions via mobile payment, they must have confidence in the technology and the service provider, which is where trust (H4) comes in. Network externalities (H5) had the most significant impact on customers’ behavioral intentions towards mobile payment, accounting for 23.7% of the variance [37] .

This result indicates that customers are more likely to use mobile payment when more merchants accept this payment method, and they believe that the more people using it, the lower the cost. As previous research has shown, creating a critical mass is essential for driving customers’ acceptance of mobile payment [17] . Table 7 displays the estimated relationships and their corresponding standardized beta, along with the results of hypothesis testing. The prediction equation is presented below, and the findings are summarized in the table.

This paper discusses several implications for future research in the field of mobile payment acceptance. The following recommendations should be considered:

・ The influence of effort expectancy (ease of use) on mobile payment acceptance should be studied further through path analysis, as it did not have a direct influence in the model but may have an indirect influence through performance expectancy.

・ Network externalities should be included as a major predictor in future mobile payment acceptance studies, as it is an important factor for technology acceptance but not commonly included in technology acceptance models.

・ Performance expectancy remains an essential predictor of technology acceptance and should be included in any model, as it is a dominant construct in technology acceptance research. Merchants, banks, and other businesses should monitor the factors influencing the adoption of mobile payment technology and design their marketing policies accordingly. Trust is also a key factor for customers’ willingness to accept mobile payment services, and managing the organization’s image and creating a trustworthy brand name should precede offering mobile payment services.

・ System developers should prioritize performance expectancy when designing mobile payment systems, as it is the second most influential predictor of acceptance. To improve uptake, developers should aim to maximize the number of payment types supported, the ability to handle different currencies, and processing speed.

Governments across Africa have been promoting the use of mobile payments to combat the spread of the pandemic, with some even waiving transaction fees. Kenya, where mobile money was first developed in Africa, has been particularly successful in this regard. According to the Economic and Financial Affairs Council, Kenya recorded a record-high of $55.1 billion in mobile transactions in 2021, which is almost a 20% increase from the previous year. The Global Consumer Survey conducted by Statista also revealed that 84% of Kenyan internet users used mobile phones for payments, which is higher than in Europe. Although only a quarter of the Kenyan population has internet access, there is still room for growth in mobile payments. In Nigeria, where internet penetration is around 34%, 60% of internet users made mobile payments in 2021 [38] [39] [40] .

In African countries like Kenya and Nigeria, mobile payments play a crucial role in promoting financial inclusion, especially for those without access to traditional banking services. Safaricom’s M-Pesa, launched in Kenya in 2007, is the leading mobile wallet in Africa, with over 50 million active monthly users.

Statista’s Global Consumer Survey found that 84% of Kenyan internet users made mobile payments in 2021, while in Nigeria, 60% of internet users utilized mobile payments, despite low internet penetration rates (Figure 2). The ease of use and accessibility of mobile payments, as well as reduced transaction times, are some of the benefits they offer. However, challenges such as inadequate infrastructure and a need for trustworthy agents can hinder their adoption. Researchers have identified factors such as relative advantage, perceived ease of use, compatibility, trust, and security as significant determinants of intention to use mobile payments. Consumers who are already comfortable with internet and mobile banking are more likely to use mobile payments and encourage others to do so. A secure and reliable service is essential for the successful transition to a cashless society in Nigeria [4] .