Abstract Since smartphones are so common, digital wallets have developed swiftly to fulfill the demands of a contemporary culture that promotes mobility and streamlines transaction processes. This study’s objective is to clarify consumer behavior in Indonesia’s adoption of digital wallets. This study was carried out in Medan involving 360 respondents. This study analyzes using a structural equation model (SEM) with LISREL software tools. The findings of this study demonstrate the significance of perceived ease of use, trust, security, and intention in boosting actual usage of digital wallets in Indonesia.The practical implications of this study are useful for comparing perceived scores on various antecedents of digital wallet adoption. This study identifies potential differences in perceptions of the elements influencing the adoption of digital wallets in Medan, particularly among the younger generation, which makes up the bulk of respondents. The success or failure of digital wallets depends on a variety of ecosystem components as well as consumer-related factors.

 

Keywords: Perceived Ease of Use, Perceived Usefulness, Trust, Security, Attitude, Intention, Actual Usage, Digital Wallet Exchange rate;

INTRODUCTION

The Ministry of Communication and Information Technology of the Republic of Indonesia (Kemenkominfo) as the leader of digital transformation in Indonesia explained Indonesia’s success in accelerating the digital transformation process over the last five years. development in the last five years in Indonesia has focused on providing infrastructure to equalize connectivity and increase economic growth. Digital wallets are now a necessity for the community in carrying out their activities and fulfilling their needs [1], [2]. The Covid-19 pandemic has increased the usage of digital wallets for transactions, especially during the past two years [3]. The development of a digital financial service product needs to pay attention to the development of solutions that are relevant to the needs of users and the surrounding ecosystem. Seeing the characteristics and needs of society 4.0 which is increasingly complex, the financial technology industry certainly needs to provide solutions and follow the changes that occur [4]. The rapid development of adoption of financial technology in the country is directly proportional to the challenges faced, namely building public trust in data security. Electronic transactions using digital wallet applications or e-wallet are increasing. Besides being considered practical, safe, fast, and offering many benefits, transactions like this are also considered safer, during a pandemic like today. Because it is in accordance with health protocols that do not interact directly   [5].

By offering simple payment channels, mobile applications has revolutionized the worldwide banking and payments sector. Due to the availability of a sizable untapped customer base made up of unbanked clients, the impact is bigger for developing countries [6]. This possibility has drawn numerous new businesses, including start-ups, to Indonesia, where they are beginning with digital wallet services. As a result, firms in the telecom, banking, and online retail sectors have started to develop business plans for their own digital wallet services. Physical wallets and digital wallets both serve the same purposes. Bank card and credit card can be used to load money into digital wallets, which can then be used to make payments to individuals or businesses (Bagla & Sancheti, 2018). A variety of channels are possible for this transaction  (Shin, 2009).

Digital wallets are one of the current trends in banking transactions brought about by technological advancements. The capacity to use mobile phone features to conduct online financial transactions is commonly considered as a component of digital wallet in the context of electronic banking. Unsurprisingly, the existence of smartphones has facilitated the quick growth of digital wallets, which has enabled them to meet the needs of the modern society that encourages mobility and facilitates transaction services today. In other words, the popularity of digital wallets is increasing as a result of the flexibility and practicality of the transactions offered by these services. Liu et al. (2015) noted that technology advancements in mobile payment systems have played a significant role in launching and forming new inventions for more than 20 years [10]. Since 2012, Indonesia has seen a rise in the use of digital wallets.

The majority of studies concur that digital wallets are crucial for boosting mobile commerce and financial inclusion [11]–[13]. On the one hand, digital wallets enable mobile commerce transactions and provide convenience. Digital wallets enable mobile commerce transactions and offer convenience on the one hand. However, they lower the cost of financial services offered through mobile devices. [14]. The original proponent of the Technology Acceptance Model,  Davis (1989), includes both input (users’ attitudes toward the technology, its perceived utility, and how easy it is to use) and output (users’ actual activity) elements (ie, behavioral intentions, technology use). Perceived usefulness (PU) and perceived ease of use (PEU) are two of these factors that are thought to account for the outcomes in significant ways [16]. Variations in the perceived usefulness and ease of use are typically explained by these factors, as well as a number of others: It was shown that TAM (Technology Acceptance Model) was significantly associated to a number of other variables, including subjective norms (SN), self-efficacy (CSE), and facilitation conditions (FC).

The term “digital wallet” refers to a digital payment system that can be used in place of traditional currency using a smartphone and various technical media, such as QR codes, near-field communication (NFC), and one-time passwords (OTPs). For users to use a digital wallet, they must first possess digital currency (e-wallet). There are a plethora of international mobile payment options such as PayPal, Google Wallet, Master Paypass Card, ZipPay, and many others [17]. Meanwhile in Indonesia, the results of a study by an independent research institute under the auspices of the Financial Times, FT Confidential Research Digital wallet show the Top 5 digital wallets [18]. The results of this research have conducted an online survey during September-December 2020, covering all of Indonesia, with 1,000 respondents selected. The most popular digital wallet company is listed as ShopeePay (50%), with a significant advantage over the other four competitors Ovo (23%), Gopay (12%), Dana (12%), and LinkAja (3%). . Consumers currently use these five brands the most frequently when making digital payments. They are highly known among consumers and have a wide range of application features since they are diligent in advertising and collaborating with different merchants.[19].

Customer use digital wallets because payments are simpler [20], [21]. Carrying a traditional wallet would be cumbersome and inefficient [22]–[24]. If customer already use a digital wallet, they are no longer need to carry the wallet [25]. One only needs to top up the digital wallet balance on a particular application on their smartphone [8]. In addition, the payment transaction process using a digital wallet is also fairly easy and fast [1], [26]. Digital wallets also help customers reduce the worry of having to carry large amounts of money [2]. Now digital wallets also have a good security system and are safe from burglary, so customers don’t have to worry [27]. Another advantage of using a digital wallet is having a variety of promotions on offer [2]. Usually, digital wallets will work with many merchants. The promotions offered also vary, ranging from rebates/discounts to cashback. So, it is not surprising that using a digital wallet like this is in great demand by the public.  Moreover, if the promo given is very tempting [28]. These various promos are often held on special days such as national celebrations, beautiful dates, or celebration days.

Customers are skeptical of digital wallet technology and mobile service providers in general (Dinh et al., 2020). They worry about identity theft, fraud involving their bank accounts and credit card numbers, and clerical errors during financial transactions. That’s why they’re wary about making purchases via smartphone app (Ramadan & Aita, 2018). Consumers have also noted that due to the small number of establishments that accept digital wallet services, the likelihood of their regular use is low. Also, customers are wary since they don’t know what they’re doing and service providers haven’t been explicit (Chris, 2017). The use of cash for purchases also seems to be a deeply ingrained habit for many customers., so this phenomenon will slow down the acceptance rate of innovative payment methods such as digital wallets. In the realm of theory, differences in the use of variables and differences in models in the modification of the TAM model are one of the research gaps that require further research on the most appropriate TAM modification model to examine digital wallet adoption. Based on this gap, it is necessary to conduct research based on the latest findings which found the phenomenon that there is a shift in consumer behavior in terms of digital wallet transactions today, it is necessary to do further research to find out how big the shift in behavior is in an academic study.

Based on the identification of the problems above, given the existing capabilities and limitations, the study will only discuss digital wallet users in Medan and analyze several important factors based on theory and previous research, Additional analysis which is also one of the novelties in this study, especially in the context in Medan and the theme of the discussion of digital wallets is the impact of “intention” on “actual usage”. This research is contextual in nature, namely conducting research on the context of users in Medan. In the explanation above, there is a research gap in this research, Therefore, this study aims to identify the crucial elements in the still-sluggish adoption of digital wallets. The goal of this study is to analyze the variables that affect attitudes and behavioral intentions towards the use of digital wallets, as well as the causes of such use and the opinions of those who already use them. as a new form of consumer behavior that begins with the desire to use digital wallets for consumer groups with a high level of potential use and actualization of digital wallet usage [29]. The research model using the usage actualization dependent variable is still little used, especially in digital wallet adoption research [30].

METHODS

The time used by researchers for this research was carried out from April 2021 to December 2021. According to Hair et al. (2014) and Malhotra et al. (2006) these criteria are intended so that this research is measurable and clear when entering the analysis stage to be used as data analysis. The sample criteria in this study are as follows:

1. Residents of Medan
2. Age 18-35 Years
3. Have a digital wallet app

Due to the aforementioned fact that the study’s target population is unclear, the study’s sample size was calculated using the view of Malhotra et al. (2006), who state that the number of samples as responders needs to be modified depending on the total number of questions in the questionnaire. Assuming values between 5 and 10, the observed variable is n. (indicator). For this study, we have 35 indicators, which, when multiplied by 10, yields the total of 350.

n = Item Indicators x 10

n = 35 x 10

n = 350

This study employed 360 samples to raise research precision and decrease research error standards from errors, however the minimum required for such a determination is just 350. The collected surveys’ responses will be randomly picked in order to keep the study’s objectivity intact, as random sampling is being used.

This study uses the Structural Equation Model (SEM). According to Hair et al., 2014), SEM analysis basically aims to obtain a structural model. The model obtained can be used for prediction or model proof. In addition, SEM can also be used to see the size of the influence, either directly, indirectly, or the total effect between the independent variable (exogenous variable) and the dependent variable (endogenous). According to [33], Because this study combines factor analysis, regression analysis, and route analysis, SEM is also a statistical method that is frequently employed in customer behavior analysis. SEM is the development of linear model generation (GLM) with multiple regression as its main part. SEM tends to be more accurate, illustrative and robust than the regression technique when modeling interactions, non-linearity, error measurement, error terms correlation and correlation between latent variables as measured by multiple indicators. SEM can also be used as an alternative to path analysis and time series data analysis based on covariance. The random sampling mechanism in this study uses the SPSS 25 software tool on the Random Number Generators menu. Questionnaires will be distributed as much as possible to the population of Medan after exceeding the minimum limit according to the minimum proportional sample, the researcher will select the data using the menu in this SPSS. The reason for using this tool is to make it easier for researchers to choose homogeneous samples with the principles of fairness and transparency. The software will select a random sample automatically so that the study can meet the random principle in this study. The use of manual sampling based on numbers such as odd-even or multiples often results in manipulation of the data so that researchers use SPSS as a sample determination tool.

RESULTS AND DISCUSSION

The characteristics of the respondents profiled in detail in this chapter, in particular characteristics such as gender, age, employment status, and digital wallet used, provide the basis for the socio-demographic analysis of the results and subsequent discussion. Respondents have the ability to determine whether they wish to participate in the study. In the end, the responses they provided produced the data needed for this research study (see table 1).

 

TABLE 1. Profile Respondents

Category	Total	Percentage
Gender
Men	172	47.7%
Woman	188	52.3%
Age
15-25	207	57.5%
26-35	75	20.8%
36-45	41	11.4%
45>	37	10.3%
Status
High School	46	12.7%
Bachelor	145	40.2%
Employee	151	41.9%
Housewife	18	5.2%
Digital Wallet
OVO	59	16.3%
GoPay	68	18.8%
ShopeePay	106	29.4%
LinkAja	51	14.4%
Dana	76	21.1%
Total	360	100%

 

To determine the extent to which a set of variables faithfully represents a set of constructs, researchers often turn to a multivariate statistical technique called confirmatory factor analysis [34]. Although comparable, exploratory factor analysis (EFA) focuses solely on exploring the data and providing information about the required number of factors to reflect the data, whereas confirmatory factor analysis (CFA) also provides this information. To determine the relationships between latent variables, exploratory factor analysis uses all of the available data. In contrast, with confirmatory factor analysis (CFA), the researcher can establish the necessary number of factors in the data and identify the measurable variables that are linked to the desired latent variables. You can use confirmatory factor analysis (CFA) to back up or disprove a measurement theory [31]. The following is a CFA model for each construct:

 

Figure 1. CFA

According to Molenaar (2019), CFA (see figure 1) is a form of factor analysis to test whether the measure of a variable is consistent with the researcher’s understanding of the nature of the variable. Basically, the measurement of validity and reliability in the CFA is determined from a value called the Loading Factor (FL) and can be symbolized by (λ). An indicator can be said to pass the validity test, if the FL value > 0.60. Meanwhile, an indicator can be said to not pass the validity test, if the value of FL < 0.60. If the value of FL < 0.50, it is recommended to remove the indicator because it is considered weak in measuring the related variables [36]. To measure validity, it can also be explained by the value (t), where the value must be equal to or more than 1.96 (t 1.96).

 

FIGURE 2. GOF

Based on figure 2 the existing GOF criteria, where the p-value limit is > 0.05 and RMSEA < 0.08, the GOF has not been met, it is concluded that the model does not fit the data, because p-value = 0.00000 < 0.05 and RMSEA = 0.153 > 0.08. LISREL provides recommendations for connecting some residuals from several indicators so that the model fits the data, as stated in the menu The Modification Indices Suggest to Add an Error Covariance (see figure 3).

Figure 3. The Modification

Path analysis is intended to test whether a variable has a significant effect on other variables. The results of path analysis using lisrel software can be summarized in figure 4 below:

 

 

Figure 4. The Modification

 

LISREL output (see table 2) of this study explain that PEOU has a significant effect on TR, ATT, and PU. PU has no significant effect on TR, ATT, ITT. TR has a significant effect on ATT and not significant on ITT. SEC has a significant effect on TR and not significant on ATT and ITT. ITT has a significant effect on AT. The practical implications of this study are useful for comparing perceived scores on various antecedents of digital wallet adoption. This study helps identify possible variations in perceptions of the factors that influence digital wallet adoption in Medan, especially among the younger generation who make up the majority of respondents. Digital wallets’ success or failure is influenced by a variety of ecosystem components as well as consumer-related factors. This comprises companies that offer services for digital wallets, internet companies, governmental organizations, and financial institutions. There are a number of variables that significantly influence the model that appears to effect how users view the adoption of digital wallets, such as the perceived ease of use variable, which significantly influences practically all path analyses. These findings suggest that Medan consumers consider digital wallet usability to be the most critical element in the decision to use this technology. You might also take security and trust-related factors into consideration. Trust and security can be viewed from the views of numerous stakeholders, including the community of mobile wallet users, information system security, service providers for mobile wallets, and regulators.

CONCLUSIONS

This study contributes to the formulation of a new model in explaining the use intention and actual use of consumer behavior towards digital wallets. This research can be used as the basis for developing TAM theory in determining digital wallet adoption and what variables have a significant influence on the model. This study uses a model that has never been done before in other studies, combining two elements of the model, namely the intention to adopt a digital wallet and the actual use of a digital wallet. This study also has a contribution to confirm the theory used in the adoption of digital wallets in Medan. It is difficult to determine whether the model or research results will apply to other technologies such as internet banking, mobile banking or e-commerce. To validate the explanatory power of the model, it may be tested on different technologies used in other industries

References

[1]          W. Rachbini, T. Evi, Ign. S. Pramesworo, and E. Rahmawati, “Effect of E-Wallet Experience on E-Satisfaction and E-Word of Mouth (Study on E-Wallet Applications in Indonesia),” CEMJ, vol. 30, no. 4, pp. 302–316, Sep. 2022, doi: 10.57030/23364890.cemj.30.4.25.

[2]          S. N. Shetu, Md. M. Islam, and S. I. Promi, “An Empirical Investigation of the Continued Usage Intention of Digital Wallets: The Moderating Role of Perceived Technological Innovativeness,” Future Business Journal, vol. 8, no. 1, p. 43, Oct. 2022, doi: 10.1186/s43093-022-00158-0.

[3]          M. Fahlevi, “Online consumer behaviour and its relationship to website atmospheric induced flow: Insights into online travel agencies in Indonesia,” presented at the IOP Conference Series: Earth and Environmental Science, 2021, vol. 729, no. 1, p. 012114.

[4]          N. S. Alharbi, A. S. AlGhanmi, and M. Fahlevi, “Adoption of Health Mobile Apps during the COVID-19 Lockdown: A Health Belief Model Approach,” International Journal of Environmental Research and Public Health, vol. 19, no. 7, p. 4179, 2022.

[5]          D. Afshari, M. Nourollahi-darabad, and N. Chinisaz, “Demographic predictors of resilience among nurses during the COVID-19 pandemic,” Work, vol. 68, no. 2, pp. 297–303, Jan. 2021, doi: 10.3233/WOR-203376.

[6]          D. Chawla and H. Joshi, “The moderating role of gender and age in the adoption of mobile wallet,” Foresight, vol. 22, no. 4, pp. 483–504, 2020, doi: 10.1108/FS-11-2019-0094.

[7]          R. K. Bagla and V. Sancheti, “Gaps in customer satisfaction with digital wallets: challenge for sustainability,” Journal of Management Development, vol. 37, no. 6, pp. 442–451, Jul. 2018, doi: 10.1108/JMD-04-2017-0144.

[8]          D. H. Shin, “Towards an understanding of the consumer acceptance of mobile wallet,” Computers in Human Behavior, vol. 25, no. 6, pp. 1343–1354, Nov. 2009, doi: 10.1016/j.chb.2009.06.001.

[9]          J. Liu, R. J. Kauffman, and D. Ma, “Competition, cooperation, and regulation: Understanding the evolution of the mobile payments technology ecosystem,” Electronic Commerce Research and Applications, vol. 14, no. 5, pp. 372–391, Sep. 2015, doi: 10.1016/j.elerap.2015.03.003.

[10]        E. D. M. Hatzakis, S. K. Nair, and M. Pinedo, “Operations in Financial Services-An Overview,” Production and Operations Management, vol. 19, no. 6, pp. 633–664, Nov. 2010, doi: 10.1111/j.1937-5956.2010.01163.x.

[11]        A. Kumar, A. Adlakaha, and K. Mukherjee, “The effect of perceived security and grievance redressal on continuance intention to use M-wallets in a developing country,” International Journal of Bank Marketing, vol. 36, no. 7, pp. 1170–1189, Oct. 2018, doi: 10.1108/IJBM-04-2017-0077.

[12]        K. Madan and R. Yadav, “Behavioural intention to adopt mobile wallet: a developing country perspective,” Journal of Indian Business Research, vol. 8, no. 3, pp. 227–244, Aug. 2016, doi: 10.1108/JIBR-10-2015-0112.

[13]        N. Singh, S. Srivastava, and N. Sinha, “Consumer preference and satisfaction of M-wallets: a study on North Indian consumers,” International Journal of Bank Marketing, vol. 35, no. 6, pp. 944–965, 2017, doi: 10.1108/IJBM-06-2016-0086.

[14]        V. S. Dinh, H. V. Nguyen, and T. N. Nguyen, “Cash or cashless ?,” vol. 34, no. 1, pp. 1–4, 2020, doi: 10.1108/SD-08-2017-0126.

[15]        F. D. Davis, “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology,” MIS Quarterly, vol. 13, no. 3, pp. 319–340, Jul. 1989, doi: 10.2307/249008.

[16]        N. Marangunić and A. Granić, “Technology acceptance model: a literature review from 1986 to 2013,” Universal Access in the Information Society, vol. 14, no. 1, pp. 81–95, Feb. 2015, doi: 10.1007/s10209-014-0348-1.

[17]        We Are Social & Hootsuite, “Digital 2020: Global Digital Overview,” 2020. doi: https://datareportal.com/reports/digital-2020-global-digital-overview.

[18]        K. C. Laudon and C. G. Traver, E-commerce 2016: business. technology. society. 2016.

[19]        G. Mediatama, “Snapcart survei penggunaan e-wallet oleh konsumen, ini dia urutan penguasanya,” kontan.co.id, Dec. 07, 2020. https://keuangan.kontan.co.id/news/snapcart-survei-penggunaan-e-wallet-oleh-konsumen-ini-dia-urutan-penguasanya (accessed Jun. 12, 2021).

[20]        S. H. Kazmi, R. R. Ahmed, K. A. Soomro, A. R. Hashem E, H. Akhtar, and V. Parmar, “Role of Augmented Reality in Changing Consumer Behavior and Decision Making: Case of Pakistan,” Sustainability, vol. 13, no. 24, 2021, doi: 10.3390/su132414064.

[21]        S. Purnama, C. S. Bangun, and S. A. Faaroek, “The Effect of Transaction Experience Using Digital Wallets on User Satisfaction in Millennial Generation,” ATM, vol. 5, no. 2, pp. 161–168, Apr. 2021, doi: 10.33050/atm.v5i2.1593.

[22]        P. Kaur, A. Dhir, R. Bodhi, T. Singh, and M. Almotairi, “Why do people use and recommend m-wallets?,” Journal of Retailing and Consumer Services, vol. 56, p. 102091, 2020, doi: https://doi.org/10.1016/j.jretconser.2020.102091.

[23]        S. Nookhao and S. Chaveesuk, “The Consumer Trust Influencing Intention to Use Electronic Wallet in Thailand,” in 2019 11th International Conference on Information Technology and Electrical Engineering (ICITEE), Oct. 2019, pp. 1–6. doi: 10.1109/ICITEED.2019.8929973.

[24]        S. Teng and K. W. Khong, “Examining Actual Consumer Usage of E-Wallet: A Case Study of Big Data Analytics,” Comput. Hum. Behav., vol. 121, no. C, Aug. 2021, doi: 10.1016/j.chb.2021.106778.

[25]        A. Daragmeh, J. Sági, and Z. Zéman, “Continuous Intention to Use E-Wallet in the Context of the COVID-19 Pandemic: Integrating the Health Belief Model (HBM) and Technology Continuous Theory (TCT),” Journal of Open Innovation: Technology, Market, and Complexity, vol. 7, no. 2, 2021, doi: 10.3390/joitmc7020132.

[26]        Y. Yong Lee, C. Lay Gan, and T. Wei Liew, “Impulse Buying’s Antecedents and Consequences: Malaysian E-Wallet Users Perceptions,” in 2021 5th International Conference on Software and E-Business (ICSEB), New York, NY, USA, 2021, pp. 45–50. doi: 10.1145/3507485.3507493.

[27]        D. Handayani, D. Setyadi, R. Suharto, S. Mintarti, T. Hidayati, and M. Setini, “Traffic Police Management Information System Performance in Digitalization Era at Samarinda Indonesia,” Productivity Management, vol. 25, pp. 126–140, Jan. 2020.

[28]        M. Fahlevi, “Mobile applications for health management in Indonesia,” presented at the Proceedings of 2020 International Conference on Information Management and Technology, ICIMTech 2020, 2020. doi: 10.1109/ICIMTech50083.2020.9211243.

[29]        J. Choudrie, C.-O. Junior, B. McKenna, and S. Richter, “Understanding and conceptualising the adoption, use and diffusion of mobile banking in older adults: A research agenda and conceptual framework,” Journal of Business Research, vol. 88, pp. 449–465, 2018.

[30]        R. Sarmah, N. Dhiman, and H. Kanojia, “Understanding intentions and actual use of mobile wallets by millennial: An extended TAM model perspective,” Journal of Indian Business Research, 2021.

[31]        J. F. Hair, M. Gabriel, and V. Patel, “AMOS Covariance-Based Structural Equation Modeling (CB-SEM): Guidelines on Its Application as a Marketing Research Tool by Joseph F. Hair, Marcelo Gabriel, Vijay Patel :: SSRN,” Brazilian Journal of Marketing, vol. 13, no. 2, Mar. 2014, [Online]. Available: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2676480#references-widget

[32]        N. Malhotra, J. Hall, M. Shaw, and P. Oppenheim, Marketing research: An applied orientation. Pearson Education Australia, 2006.

[33]        J. F. Hair, G. T. M. Hult, C. M. Ringle, and M. Sarstedt, A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), 2nd ed. California: Sage, 2017.

[34]        M. R. R. Adam, Practical guide of the integrated structural equation modeling (SEM) with LISREL and AMOS for marketing & social sciences thesis. Deepublish, 2018.

[35]        P. C. Molenaar, “Review of Multivariate Analysis With LISREL,” 2019.

[36]        I. B. De Groot, M. M. Favejee, M. Reijman, J. A. Verhaar, and C. B. Terwee, “The Dutch version of the Knee Injury and Osteoarthritis Outcome Score: a validation study,” Health and quality of life outcomes, vol. 6, no. 1, pp. 1–11, 2008.