Abstract
Purpose
This study aims to construct an appropriate framework by incorporating essential components from the most renowned theories to investigate the variables that impact behavioural intentions towards embracing cashless transactions (CLT).
Design/methodology/approach
A survey was conducted to ascertain the users’ intention to adopt CLT in Chennai, Tamil Nadu, India. Further, this study used a “partial least squares-based structural equation modelling” technique to analyse the relationships between latent factors.
Findings
The results of the proposed model revealed that 11 independent variables together explain the intention to use CLT with a 60.5% explanatory power. Further, perceived usefulness is the most influential factor in predicting users’ willingness to adopt CLT, followed by social influence, perceived costs, attitude, trust and device barriers. Finally, the findings of moderator effects indicate that income and experience interact positively and strongly with behavioural intention to adopt CLT. It indicates that high-income, experienced users are more likely to convert their intentions into actions.
Originality/value
This study integrated critical elements from the major theories, such as Theory of Reasoned Action, Technology Acceptance Model, Decomposed Theory of Planned Behaviour, the unified theory of acceptance and use of technology (UTAUT) model and UTAUT2, to investigate the adoption of CLT. As a result, 11 crucial factors were identified from the existing literature that impacts CLT adoption without overlapping. Consequently, the model presented in this study provides a more profound understanding than previous research regarding why individuals adopt CLT systems. Accordingly, these results could aid policymakers in addressing people’s concerns and facilitating a seamless transition to a cashless society.
Keywords
Citation
Raj L., V., Amilan, S. and Aparna, K. (2023), "Factors influencing the adoption of cashless transactions: toward a unified view", South Asian Journal of Marketing, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/SAJM-11-2022-0071
Publisher
:Emerald Publishing Limited
Copyright © 2023, Vimal Raj L., S. Amilan and K. Aparna
License
Published in South Asian Journal of Marketing. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
The prevention of crimes has been regarded as superior to the punishment of criminals after the crime has been committed. Despite the efforts of governments to enhance the resources available to the criminal justice system, the crime has continued to surge and negatively affect our daily lives (Warwick, 1993). Further, a substantial proportion of these crimes involve stealing cash or property to obtain cash, which is the primary source of revenue for illegal activities (Warwick, 1993; Armey et al., 2014). In addition, criminals often deposit their illicit profits in financial institutions using cash transactions, which can make it difficult for law enforcement to trace the money (Alba, 2003; Goel and Mehrotra, 2012). Therefore, countries with high cash volumes may be more susceptible to corruption among officials and bribe-takers.
Moreover, cash provides anonymity for terrorists and rebels to engage in undetected transactions (Lipow, 2010), and even a minimal quantity of physical cash in circulation is adequate to finance a broad range of illegitimate operations (Wright et al., 2017). Hence, governments must move towards a completely cashless economy to combat these crimes effectively (Warwick, 1993). Further, eliminating physical currency would force criminals to use cashless transactions (CLT), which are traceable, leading to increased transparency and economic security (Vimal Raj et al., 2023; Raj et al., 2021). The shift to cashless economies can also reduce government spending and promote employment. However, the full benefits of a cashless economy cannot be realised until a significant percentage of the population regularly engages in CLT (Vimal Raj et al., 2023). Furthermore, “cashless transactions (CLT)” were described as modes of payment that do not require the exchange of tangible currency. Such methods include “Real Time Gross Settlement (RTGS),” “National Electronic Fund Transfer (NEFT),” “Immediate Payment Service (IMPS),” “Unified Payments Interface (UPI),” “Near Field Communication (NFC)” technology, payment via mobile wallets, banking cards, “Quick Response (QR)” codes and other payment methods that do not involve cash (Raj et al., 2021).
In India, the CLT systems have become widely promoted after implementing the Digital India campaign in 2015. In addition, the “Reserve Bank of India (RBI),” the central bank of India responsible for monetary policy implementation, has undertaken multiple campaigns to promote the use of CLT among the general public (Vimal Raj et al., 2023). Consequently, according to the RBI’s annual reports, this has resulted in an enormous increase in the entire volume of CLT in India, which has gone from 7,046.6 million in the fiscal year 2015–2016 (Reserve Bank of India Annual Report, 2017) to 71,953.1 million in the fiscal year 2021–2022 (Reserve Bank of India Annual Report, 2021), marking a growth rate of 921.10%. However, despite this surge in the adoption of CLTs, the rate in India is still relatively low compared to the rates in developed countries. For example, according to “Bank for International Settlements (BIS)” Statistics Explorer (BIS Statistics Explorer, n.d.): Table CT5, India’s average CLT per capita in 2021 was just 47 units, whereas the figure for advanced economies was above 400 units. Additionally, according to the RBI Bulletin published in August 2019, the average value of CLT made in India is 3,910 United States Dollars per person. In contrast, the value of CLT in developed nations was above 500,000 (Reserve Bank of India Bulletin, 2019). Based on these findings, it seems that the implementation of CLT in India is still in its infancy and has substantial opportunities for growth. Consequently, it is essential to comprehend users’ acceptance of CLT and identify the factors influencing their intent to use it through an extensive study model.
2. Theoretical background
Previous studies have demonstrated that multiple theories have been widely employed to explore individuals’ adoption and usage of new “Information Technology (IT).” These models include the “Theory of Reasoned Action (TRA),” “Social Cognitive Theory (SCT),” “Technology Acceptance Model (TAM),” “Theory of Planned Behaviour (TPB),” “Model of PC Utilization (MPCU),” “Decomposed Theory of Planned Behaviour (DTPB),” “Innovation Diffusion Theory (IDT)” and “Unified Theory of Acceptance and Use of Technology (UTAUT).” Additionally, the TAM and the UTAUT are among the most commonly utilised theories to elucidate users’ behavioural intentions to engage in any technology.
Davis developed the TAM model in 1989 and derived it from TRA, which assesses an individual’s willingness to participate in a technological activity. The TAM was explicitly designed for the IT field and is utilised to forecast the adoption and usage of IT. The model is intention-based, estimating usage based on behavioural intention. The model has five key components: “perceived usefulness,” “perceived ease of use,” “attitude toward use,” “intention to use” and “actual use.” Further, the TAM model has been extensively used to investigate the acceptance of online banking, mobile banking and other CLT systems, owing to its effectiveness, conciseness and simplicity (Sarmah et al., 2021; Shin, 2009; Williams, 2021; Lisana, 2021; Alshurideh et al., 2021; Saha et al., 2022; Flavian et al., 2020; Chawla and Joshi, 2019; Priya et al., 2018; Bailey et al., 2017; Koksal, 2016; Bashir and Madhavaiah, 2015a, b).
Venkatesh et al. (2003) integrated major information system (IS) theories to develop UTAUT, which includes four fundamental constructs, namely “performance expectancy,” “effort expectancy,” “social influence,” and “facilitating conditions” that impact intention and use. Moreover, they employed “gender,” “age,” “experience,” and “voluntariness of use” as moderators in this model. Additionally, Venkatesh et al. (2012) expanded on UTAUT and introduced UTAUT2, which included three extra elements, namely “hedonic motivation,” “price value,” and “habit.” Furthermore, earlier researchers in this area have extensively utilised UTAUT2 in their investigations due to its capacity to provide comprehensive and explanatory information (Sivathanu, 2019; Raj et al., 2023; Gupta and Arora, 2020; Vimal Raj et al., 2023; Alalwan et al., 2017; Sripalawat et al., 2011). In addition, earlier research in the field of CLT has employed UTAUT2 due to its significant association with this domain (Abegao Neto and Figueiredo, 2022; Al-Saedi et al., 2020; Giovanis et al., 2019; Patil et al., 2020). Nevertheless, previous studies have not covered the full spectrum of adoption factors in CLT due to the following deficiencies:
The existing theoretical models, such as UTAUT, UTAUT2, and TAM, were developed to explore the acceptance of technological devices such as computers and mobile phones. Accordingly, these models do not take into account several additional factors that could impact the adoption of CLT. In addition, prior literature comprises numerous analogous constructs, leading to the redundancy of its components. Further, given the context of CLT acceptance, a number of the assumptions made in UTAUT2 are rendered meaningless. As an illustration, the UTAUT2 incorporates the significant construct of hedonic motivation, which gauges individuals' enjoyment while playing games or watching movies in the context of adopting mobile or computer technology. However, this construct is not applicable in the area of CLT adoption. Consequently, these theories did not provide a completely accurate explanation for adopting CLT. Therefore, to address these deficiencies, it is necessary to amalgamate all existing literature on CLT to formulate a complete model that incorporates all positive/negative factors and interaction variables. Accordingly, the subsequent sections describe how this research framework establishes a suitable model to elucidate the uptake of CLT.
3. Conceptual framework and hypothesis development
The present research began with a literature review to establish an appropriate model that comprehensively encompasses various factors influencing behavioural intentions to adopt CLT without overlapping constraints. The following is an overview of the processes that were included in the literature study that was carried out as a part of the current inquiry to get a glimpse of the antecedents that are relevant to the suggested research model:
In the initial stage of this research, over 110 previously published research articles were reviewed and analysed using several online databases, including “Science Direct,” “Emerald,” “Sage,” “Springer,” “Taylor & Francis,” “IEEE,” and “Google Scholar.” Accordingly, this research identified 63 factors that directly influence the behavioural intention towards using CLT. Further, the identified factors were classified based on similarity to prevent overlapping. Furthermore, the essential elements of the CLT adoption process were determined to be picked from each of the groups investigated in this research. Consequently, to develop a comprehensive model that explains the uptake of CLT, this study has identified 13 dimensions, including “Perceived Usefulness (PU),” “Perceived Ease of Use (PEoU),” “Attitude (ATT),” “Perceived Trust (PT),” “Personal Innovativeness (PI),” “Social Influence (SI),” “Self-Efficacy (SE),” “Perceived Risk (PR),” “Perceived Costs (PC),” “Anxiety (ANX),” “Device Barrier (DB),” “Behavioural Intention (BI)” and “Actual Use (AU)” (Figure 1 illustrates the research framework).
3.1 Perceived ease of use (PEoU)
“Perceived ease of use refers to the degree to which a person believes that using a particular system would be ‘free of effort’ (Davis, 1989).” This definition focuses on the user's impression of the effort required to utilise technology and is an essential component in forecasting the propensity to accept new kinds of technology (Williams, 2021; Lisana, 2021; Alshurideh et al., 2021). The significance of PEoU is particularly relevant in determining the acceptance of CLT systems, as it is the most important and frequently cited precursor (Alhassany and Faisal, 2018; Koksal, 2016; Bashir and Madhavaiah, 2015b). Several researchers have investigated the impact of PEoU on PU and ATT towards adopting CLT systems (Sarmah et al., 2021; Sripalawat et al., 2011; Priya et al., 2018). As such, we hypothesise:
PEoU positively influences individuals' ATT towards the adoption of CLT.
PEoU positively influences individuals' PU of CLT.
3.2 Perceived usefulness (PU)
“Perceived usefulness defines the degree to which a person believes that using a particular system would enhance his or her job performance (Davis, 1989).” As a result, users perceive increased value in adopting new technology (Lisana, 2021; Alshurideh et al., 2021). In the context of CLT, PU refers to how effective an individual believes using a CLT method is compared to using cash (Flavian et al., 2020; Chawla and Joshi, 2019). In addition, this can result in individuals developing positive ATT towards CLT (Alhassany and Faisal, 2018; Flavian et al., 2020; Koksal, 2016). These unique features of CLT are likely to reinforce these positive views. Further, in previous research, it has been shown that there is a substantial correlation between an individual's PU of CLT, their ATT towards it, and their BI to use it (Williams, 2021; Sripalawat et al., 2011; Chawla and Joshi, 2019, 2020; Priya et al., 2018; Bashir and Madhavaiah, 2015b). Therefore, we hypothesise:
PU positively influences individuals' ATT towards the use of CLT.
PU positively influences individuals' BI towards the adoption of CLT.
3.3 Attitude (ATT)
“Attitude” refers to a person's overall evaluation of a particular object or behaviour (Davis, 1989). This concept has been evaluated through various influential theories on the adoption of IT, such as the TRA and the TAM (Flavian et al., 2020). These theories propose that a person's BI to use an IT is determined by their ATT (Patil et al., 2020). Similarly, if an individual has a positive ATT towards CLT, they are more likely to have a higher BI to adopt it. On the other hand, if an individual has a negative ATT towards CLT, they are less likely to embrace it (Flavian et al., 2020). Besides, several studies on CLT have demonstrated a robust association between ATT and BI towards utilising CLT (Chawla and Joshi, 2019, 2020; Shin, 2009; Bailey et al., 2017; Bashir and Madhavaiah, 2015a). Therefore, we hypothesise:
ATT positively influences individuals' BI towards the adoption of CLT.
3.4 Perceived trust (PT)
The notion of “perceived trust” refers to a person's impression of the reliability of the institutional surroundings (Vimal Raj et al., 2023; Alalwan et al., 2017). This concept encompasses trust in service providers such as financial institutions and telecommunications companies (Sarmah et al., 2021). Further, this trust might result from pleasant previous experiences or a well-established reputation (Raj et al., 2023). Additionally, security, convenience, prestige, familiarity, and regulation build trust among consumers and encourage them to embrace CLT methods (Chawla and Joshi, 2020; Giovanis et al., 2019). Furthermore, prior studies provide evidence of a meaningful and positive relationship between PT and the BI to adopt CLT (Sarmah et al., 2021; Shin, 2009; Raj et al., 2023; Lisana, 2021; Alshurideh et al., 2021; Bashir and Madhavaiah, 2015a). Therefore, we hypothesise:
PT positively influences individuals' BI towards the adoption of CLT.
3.5 Personal innovativeness (PI)
“Personal innovativeness” pertains to an individual's readiness and openness to adopt and utilise novel technologies (Raj et al., 2023). Similarly, the CLT methods, such as mobile payments, e-wallets and contactless cards, are relatively new technologies that require users to be comfortable with and willing to use them (Williams, 2021; Saha et al., 2022). Therefore, individuals who are more innovative and open to using new technologies may have a higher BI to use CLT compared to those who are less innovative (Raj et al., 2023; Vimal Raj et al., 2023). Moreover, prior research found a positive link between PI and individuals’ BI towards adopting CLT (Williams, 2021; Vimal Raj et al., 2023; Raj et al., 2023). Therefore, we hypothesise the following:
PI positively influences individuals’ BI towards the adoption of CLT.
3.6 Social influence
“Social influence” refers to other people’s impact on an individual’s attitudes, beliefs and behaviours (Venkatesh et al., 2003). In the context of CLT, SI can come from various sources, such as family, friends, peers and even strangers. These individuals can influence their perception of CLT and their decision to use them (Sivathanu, 2019). Similarly, SI can also be amplified by social media and other digital platforms, which can spread information and opinions about CLT. The use of influencers and social media marketing by companies offering several CLT methods can also positively influence an individual’s decision to use them (Patil et al., 2020). Additionally, earlier research has shown that there is a favourable correlation between SI and BI in the adopting of CLT (Raj et al., 2023; Lisana, 2021; Patil et al., 2020; Vimal Raj et al., 2023; Giovanis et al., 2019). Therefore, we hypothesise the following:
SI has a positive effect on a person’s willingness to adopt CLT.
3.7 Self-efficacy
The term “self-efficacy” refers to an individual’s conviction in their capacity to successfully use CLT systems to conduct financial transactions (Lisana, 2021; Raj et al., 2023; Al-Saedi et al., 2020). Further, individuals with a strong sense of SE regarding CLT are usually self-assured in their capacity to utilise these payment methods proficiently, tackle potential challenges and feel comfortable with the technology used (Vimal Raj et al., 2023). On the other hand, individuals with low levels of self-efficacy regarding CLT may exhibit reluctance to employ these payment systems, lack faith in their abilities, and reject adopting them (Singh and Srivastava, 2018). Moreover, previous studies have shown that SE significantly predicts an individual's intention to adopt CLT methods (Lisana, 2021; Bashir and Madhavaiah, 2015b). Therefore, we hypothesise:
SE positively influences individuals' PEoU in CLT.
SE positively influences individuals' BI towards the adoption of CLT.
3.8 Device barrier (DB)
“Device barriers” can pose challenges to the adoption of CLT. Factors such as the user interface, ease of navigation and input, display clarity, and the quality of the mobile device can all impact the user's perception of the service quality and affect their willingness to use mobile payment methods (Laukkanen, 2008). Additionally, the small screen size of mobile devices can present obstacles to adopting CLT, particularly for services requiring complex or detailed inputs (Sripalawat et al., 2011). Hence, to encourage greater adoption of CLT, it is essential to address these DB and tailor mobile content to the limitations and preferences of the user's device. Moreover, prior research demonstrates significant associations between DB and users' BI to engage in cashless transactions (Laukkanen, 2008; Sripalawat et al., 2011). Therefore, we hypothesise:
DB negatively influences individuals' BI towards the adoption of CLT.
3.9 Anxiety (ANX)
Anxiety refers to the concerns and worries that users may have regarding the security and usability of CLT applications (Raj et al., 2023). Users could be worried about the safety of the CLT apps, which might result in ANX and a lack of faith in the system (Vimal Raj et al., 2023). In addition, forgetfulness can be a source of ANX, and users may fear that they will forget their usernames and password, resulting in the loss of account access (Bailey et al., 2017). Furthermore, users may be concerned about the physical security of their devices and the possibility of theft, loss, or misappropriation by third parties (Celik, 2016). Thus, reducing CLT's ANX may enhance users' opinions of the technology and increase their desire to use it (Patil et al., 2020). Besides, previous research on the acceptance of CLT has verified that ANX is a crucial factor in shaping people's technology adoption (Bailey et al., 2017; Patil et al., 2020; Celik, 2016). Therefore, we hypothesise:
ANX negatively influences individuals’ BI towards the adoption of CLT.
3.10 Perceived cost
The expression “perceived cost” pertains to an individual’s perception of the expenses linked to utilising CLT techniques (Abegao Neto and Figueiredo, 2022). These costs can encompass the expenses incurred in acquiring the required devices, Internet charges, transaction processing fees levied by banks or other financial institutions and any additional charges necessary to finalise the transaction (Vimal Raj et al., 2023). Further, a high PC may discourage individuals from adopting or using these methods, whereas a low PC may enhance their inclination to adopt and use them (Singh and Srivastava, 2018). Moreover, earlier studies have shown that an users’ comprehension of the PC has an adverse effect on their BI to adopt CLT methods (Abegao Neto and Figueiredo, 2022; Vimal Raj et al., 2023; Raj et al., 2023; Priya et al., 2018; Al-Saedi et al., 2020). Therefore, we hypothesise the following:
PC has an adverse effect on individuals' BI concerning adopting CLT methods.
3.11 Perceived risk
“Perceived risk” is the degree to which people believe CLT procedures will lead to adverse outcomes, including money loss, fraud or identity theft (Raj et al., 2023). Further, this certainty makes CLT approaches less appealing since it raises concerns about transaction security (Vimal Raj et al., 2023). Moreover, previous research has shown that the associated PR negatively impacts BI using CLT technologies (Abegao Neto and Figueiredo, 2022; Raj et al., 2023; Priya et al., 2018; Vimal Raj et al., 2023; Bashir and Madhavaiah, 2015b; Giovanis et al., 2019). Therefore, we hypothesise the following:
PR has an adverse impact on individuals’ BI concerning accepting CLT methods.
3.12 Behavioural intention
The term “BIs” refers to an individual’s subjective propensity or willingness to partake in a particular behaviour, such as using CLT (Raj et al., 2023). The concept was first introduced in the TPB and TRA technology adoption models and has since been combined into consequent adoption models (Alalwan et al., 2017). Moreover, earlier research has indicated a robust association between BI and actual behaviour (Sivathanu, 2019; Raj et al., 2023; Sarmah et al., 2021; Gupta and Arora, 2020; Alalwan et al., 2017; Sripalawat et al., 2011; Vimal Raj et al., 2023; Shin, 2009; Patil et al., 2020). Therefore, we hypothesise the following:
BI positively influences individuals’ adoption of CLT.
3.13 Moderators
The intention is a crucial element in decision-making that determines subsequent actions. However, no empirical evidence in the current literature supports the hypothesis that the effects of education, income and experience act as moderators between BIs and performance. Thus, this study investigates the factors that moderate the relationship between intentions and actions. Conversion of intentions into actions necessitates knowledge and experience, and education directly influences the knowledge acquired. Similarly, cash availability at the bank is essential for CLT since they cannot be conducted without money. As a result, higher-income individuals may have more opportunities to engage in CLT. Consequently, this research examines the interactive impact of education, income and experience with CLT on the relationship between behavioural intention and performance. Based on this, we propose the following hypothesis:
The influence of BI on the AU of CLT will be moderated by education.
The influence of BI on the AU of CLT will be moderated by income.
The influence of BI on the AU of CLT will be moderated by experience.
4. Research methodology
4.1 Data and sample
A poll was performed to determine the factors influencing BI in adopting CLT. Further, the respondents were CLT users in Chennai, India. As this population is unknown, this study collected 456 responses using convenience sampling (Alalwan et al., 2017; Vimal Raj et al., 2023; Al-Saedi et al., 2020). Further, 438 valid responses were taken for final analysis from the 456 respondents who replied. Additionally, the study utilised previously published instruments with slight modifications to measure agreement using a “seven-point Likert scale,” ranging from “strongly disagree” to “strongly agree.” Furthermore, the scales were initially pretested twice with 110 participants to ensure validity and appropriateness.
4.2 Data analysis technique
The study utilised a method called “partial least squares-based structural equation modelling (PLS-SEM)” to examine how different factors are correlated. Further, the research was conducted in two stages. The first stage used SEM to explore how observed and latent variables were connected in the measurement model. In the second stage, the study employed the PLS method in the structural model to investigate how the latent constructs were related to each other. Therefore, the research was split into two phases.
5. Data analysis and results
5.1 Measurement model
The measurement model’s reliability and validity are essential for accurate results. The analysis shows high Cronbachs Alpha readings ranging from 0.721 to 0.937 and “composite reliability” values ranging from 0.837 to 0.952 (Table 1), both surpassing the recommended thresholds of 0.7 (Hair et al., 2014). These outcomes indicate that the measurement model has dependable internal consistency. Further, “convergent validity (CV),” which examines the conceptual relationship between multiple items, was evaluated by analysing factor loadings and “average variance extracted (AVE)” values. All factor loading levels exceeded 0.7, and AVE values ranged from 0.632 to 0.823 (Table 1), surpassing the threshold of 0.5 (Hair et al., 2010). Thus, the convergence approach is valid. “Discriminant validity” was tested by comparing correlation coefficients with the square root of AVEs (Table 2). The square roots of AVEs were higher than the corresponding correlation coefficients, indicating high discriminant validity (Fornell and Larcker, 1981; Aparna et al., 2023).
5.2 Structural model
Researchers used R2 and Q2 values and path significance to assess a structural model’s quality. A high-quality model has an R2 value of 0.1 or higher for the dependent variable and all paths (Hair et al., 2014). Table 3 shows all R2 values are above 0.1, indicating excellent predictive ability. A Q2 value larger than 0 indicates the model is meaningful in its predictions (Hair et al., 2017), and the constructs studied have significant predictability. The model’s “Goodness-of-fit (GoF)” is evaluated with the standardized root mean square residual (SRMR) value, which compares the fit to a baseline. The SRMR value of 0.066 is below the acceptable threshold of 0.10 (Hair et al., 2014), indicating a satisfactory model fit.
Furthermore, in addition to assessing the GoF, hypotheses were examined to determine the significance of the relationships. The findings of this investigation, as presented in Table 4, revealed that PU, ATT, PT and SI have a statistically significant and positive impact on BI to engage in CLT. Hence, H2b (β = 0.423, t = 7.688), H3 (β = 0.133, t = 4.080), H4 (β = 0.121, t = 3.327) and H6 (β = 0.310, t = 3.968) are supported. However, while PI and SE positively affect BI to use CLT, they are not statistically significant. Therefore, H5 (β = 0.126, t = 1.729) and H7a (β = 0.056, t = 1.538) are not supported. Additionally, the effects of PEoU on ATT, PEoU on PU, PU on ATT, SE on PEoU and BI on AU are all significantly and positively correlated to one another. Hence, H1a (β = 0.258, t = 4.016), H1b (β = 0.489, t = 10.059), H2a (β = 0.179, t = 3.476), H7b (β = 0.305, t = 6.004), and H12 (β = 0.570, t = 11.90) are supported (Figure 2).
Moreover, the present study found that the DB and PC have a considerably negative impact on the BI to use CLT, while ANX and PR have an insignificant negative impact. As an outcome, the hypotheses H8 (β = −0.080, t = 2.151) and H10 (β = −0.182, t = 4.781) are supported, and H9 (β = −0.044, t = 1.225) and H11 (β = −0.056, t = 1.256) are not supported. Besides, the study examined the moderating effects of education, income and experience on the relationship between the independent and dependent variables. The findings suggest that education does not have a statistically significant effect on BI to use CLT. Therefore, H13 (β = 0.025, t = 0.649) is not supported. However, income and experience significantly and positively moderate the relationship between the BI and AU of CLT. Consequently, H14 (β = 0.200, t = 4.893) and H15 (β = 0.133, t = 2.416) are supported.
6. Discussion
This study proposes a comprehensive model that integrates all the existing research on CLT to explore its widespread adoption. Consequently, this research provides a more thorough comprehension of the roles played by various factors such as PU, PEoU, ATT, SE, PT, SI, PI, ANX, PR, DB, PC and BI in the CLT adoption process among users of such transactions.
According to the findings of this research, the PEoU has a substantial influence on users’ ATT towards the use of CLT. This finding aligns with previous studies (Bashir and Madhavaiah, 2015a; Shin, 2009). It recommends that individuals are more likely to have a positive ATT towards CLT if they find them easy to use. Additionally, users’ perception of the PEoU of CLT significantly affects their PU. This result aligns with previous studies (Sarmah et al., 2021; Williams, 2021; Lisana, 2021; Alshurideh et al., 2021). It indicates that, if users perceive CLT as easy to use, they are more expected to perceive them as useful. Therefore, banks and other service providers of CLT should promote their usage and prioritise designing user-friendly systems to enhance users’ perception of the usefulness of CLT. Furthermore, the study reveals that individuals’ positive ATT towards CLT are influenced by their PU. Moreover, the research confirms that this positive perception of use also affects the BI in adopting CLT, consistent with earlier studies (Chawla and Joshi, 2019, 2020; Flavian et al., 2020; Patil et al., 2020). This indicates that individuals who view CLT as beneficial are more inclined to have a favourable ATT towards their use than those who do not perceive their advantages. Consequently, developers of CLT systems should prioritise creating user-friendly applications that improve the efficacy of CLT for their users.
Likewise, the investigation discovered that PT substantially affected individuals’ BI to utilisese CLT. This outcome corresponds with findings from prior studies (Sarmah et al., 2021; Vimal Raj et al., 2023; Lisana, 2021; Raj et al., 2023; Alshurideh et al., 2021; Chawla and Joshi, 2020), which also found that the trustworthiness of the CLT system positively affected the BI to use it. Therefore, providers of CLT services must continuously enhance the trustworthiness of their payment applications to preserve optimistic BI among CLT users. The study also revealed that SI significantly and positively impacted individuals’ BI towards using CLT. This finding aligns with prior research (Vimal Raj et al., 2023; Giovanis et al., 2019). Further, previous studies have demonstrated a positive relationship between SI and BI in the context of CLT (Sivathanu, 2019; Raj et al., 2023; Lisana, 2021; Giovanis et al., 2019; Patil et al., 2020), which found that social pressure and the opinions of essential peers significantly influenced the BI to use CLT. Therefore, to encourage the adoption of CLT systems, service providers must consider the influence of social factors and develop strategies to positively influence individuals’ perspectives towards their use.
The study also found that perceived SE did not directly affect BI but rather an indirect effect through PEoU and ATT. This result differs from previous studies (Singh and Srivastava, 2018). This suggests that most users believe they can use CLT and are willing to use them based on their ability. Furthermore, PI had a positive but not yet significant association with BI, contrasting with earlier studies (Williams, 2021; Vimal Raj et al., 2023; Giovanis et al., 2019) that found users were generally eager to adopt new CLT technology, regardless of their level of adoption.
In terms of impediments, devices still have limitations such as screen size and processing power, which can hinder the use of CLT. Therefore, the device constraint is one of the reasons why individuals may resist using CLT methods. Hence, service providers should consider developing applications optimised for devices with small screens and limited processing power to ensure users can easily and quickly complete transactions. Additionally, the research indicates that PR has a small and negative impact on the BI to adopt CLT. These outcome contrasts previous studies (Abegao Neto and Figueiredo, 2022; Raj et al., 2023; Priya et al., 2018; Vimal Raj et al., 2023; Alhassany and Faisal, 2018), which suggest that higher PR is associated with lower BI to use CLT systems. Therefore, users of CLT should be informed about the potential risks associated with these systems. Further, this study is consistent with previous research that suggests PC significantly negatively impacts the BI to use CLT (Raj et al., 2023; Vimal Raj et al., 2023; Saha et al., 2022). This could be because users are less likely to use CLT systems when the cost of the device and the transaction fees are high. Therefore, service providers of CLT systems should focus on developing applications that can operate smoothly on various device types, ensuring that all CLT users have equal access to services regardless of their device’s cost. This study found that ANX has a minor negative impact on the BI to use CLT. It recommends that users of CLT understand how the system operates while using the technology.
Furthermore, the study’s results suggest that income and experience significantly positively impact the relationship between BI and the use of CLT. This indicates that users with higher incomes and more experience are likelier to act on their intentions to use these transactions. However, while education positively influences the BI in adopting CLT, it is not statistically significant. This suggests that most respondents, including both low and high users of CLT, have a good education.
7. Conclusions
7.1 Theoretical implications
The findings of this research provide several significant contributions to the theoretical frameworks. First, to explore the factors influencing the adoption of CLT, this study incorporated relevant components from renowned theories like TAM, TRA, DTPB, UTAUT and UTAUT2. As a result, the study identified 11 essential components from the existing body of research that impacts the adoption of CLT without overlapping one another. These factors are PU, PEoU, ATT, PI, PT, SI, SE, PR, PC, ANX and DB. Consequently, the proposed model provides an explanation for why individuals obtain CLT methods that is more comprehensive than any previous study. Furthermore, this investigation effectively analysed the impact of education, income and experience as moderators on BI and performance, which had not been explored previously. The results of this study indicate that income and experience play a significant role in moderating users’ BI towards their actual behaviour.
7.2 Practical implications
In terms of practical implications, this study first quantifies the factors influencing the adoption of CLT. Accordingly, this would improve the comprehension of policymakers and bankers concerning the influence of these factors on the growth of CLT infrastructure. Additionally, the presented findings would enlarge the knowledge of CLT application developers by prompting them to consider these issues when designing such applications. Specifically, application developers should focus on creating CLT applications that are compatible with multiple devices, efficient in task completion, improve payment transaction performance, user-friendly, easy to use and socially responsible to sustain the positive intention of using CLT systems. Moreover, bank decision-makers should organise awareness programmes to educate CLT users on the risks of improper use of CLT systems. Furthermore, CLT service providers should ensure that their channels for CLT conduct financial transactions securely and efficiently, regardless of the location or time. As a result, this study would lead to the creation of user-friendly and secure CLT methods that match the preferences and demands of users.
7.3 Limitations and future research perspectives
Although this study contributes to the existing knowledge, particularly in the Indian context, it has several limitations. First, this inquiry focuses on Chennai, where CLT is prevalent, and the residents are relatively more educated than in other parts of the nation. In addition, people with higher levels of education tend to be more aware and adaptive towards novel technologies. Thus, this study recommends exploring the situation in rural areas of the country, which require more investigation. In addition, users’ viewpoints are the only ones considered in this study. Yet, while conducting this research, it would have been prudent to consider the perspectives of a broader range of stakeholders, including non-users, merchants, retailers and business owners. Finally, despite the present research having enough participants, it is possible that the findings cannot be generalised to all Indian customers.
Figures
Results of the measurement model
| Latent factors | Factor loadings | VIF | Cronbach’s alpha | Composite reliability | AVE |
|---|---|---|---|---|---|
| Perceived usefulness (PU) | 0.908 | 0.929 | 0.686 | ||
| PU1 | 0.836 | 2.682 | |||
| PU2 | 0.832 | 2.515 | |||
| PU3 | 0.771 | 1.893 | |||
| PU4 | 0.827 | 2.712 | |||
| PU5 | 0.833 | 2.848 | |||
| PU6 | 0.866 | 3.252 | |||
| Perceived ease of use (PEoU) | 0.929 | 0.942 | 0.700 | ||
| PEoU1 | 0.879 | 3.397 | |||
| PEoU2 | 0.893 | 3.885 | |||
| PEoU3 | 0.867 | 3.436 | |||
| PEoU4 | 0.835 | 2.639 | |||
| PEoU5 | 0.818 | 2.342 | |||
| PEoU6 | 0.846 | 2.904 | |||
| PEoU7 | 0.704 | 1.905 | |||
| Attitude (ATT) | 0.928 | 0.949 | 0.823 | ||
| ATT1 | 0.934 | 4.154 | |||
| ATT2 | 0.854 | 2.54 | |||
| ATT3 | 0.906 | 3.401 | |||
| ATT4 | 0.932 | 4.331 | |||
| Perceived trust (PT) | 0.879 | 0.910 | 0.670 | ||
| TR1 | 0.876 | 2.563 | |||
| TR2 | 0.787 | 2.067 | |||
| TR3 | 0.740 | 2.079 | |||
| TR4 | 0.796 | 2.339 | |||
| TR5 | 0.885 | 3.169 | |||
| Personal innovativeness (PI) | 0.905 | 0.920 | 0.744 | ||
| PI1 | 0.961 | 3.15 | |||
| PI2 | 0.763 | 2.863 | |||
| PI3 | 0.823 | 2.433 | |||
| PI4 | 0.891 | 3.814 | |||
| Social influence (SI) | 0.885 | 0.920 | 0.743 | ||
| SI1 | 0.884 | 2.412 | |||
| SI2 | 0.868 | 2.335 | |||
| SI3 | 0.845 | 2.321 | |||
| SI4 | 0.850 | 2.365 | |||
| Self-efficacy (SE) | 0.885 | 0.920 | 0.743 | ||
| SE1 | 0.872 | 2.040 | |||
| SE2 | 0.912 | 2.940 | |||
| SE3 | 0.815 | 2.310 | |||
| SE4 | 0.808 | 2.268 | |||
| Perceived risk (PR) | 0.931 | 0.941 | 0.667 | ||
| PR1 | 0.714 | 2.390 | |||
| PR2 | 0.802 | 3.993 | |||
| PR3 | 0.841 | 3.207 | |||
| PR4 | 0.851 | 2.962 | |||
| PR5 | 0.885 | 3.031 | |||
| PR6 | 0.741 | 2.893 | |||
| PR7 | 0.802 | 3.556 | |||
| PR8 | 0.882 | 3.922 | |||
| Perceived costs (PC) | 0.906 | 0.928 | 0.722 | ||
| PC1 | 0.800 | 3.030 | |||
| PC2 | 0.880 | 3.289 | |||
| PC3 | 0.892 | 2.574 | |||
| PC4 | 0.876 | 3.549 | |||
| PC5 | 0.796 | 3.153 | |||
| Anxiety (ANX) | 0.937 | 0.952 | 0.798 | ||
| ANX1 | 0.866 | 2.810 | |||
| ANX2 | 0.905 | 3.763 | |||
| ANX3 | 0.872 | 3.358 | |||
| ANX4 | 0.915 | 4.191 | |||
| ANX5 | 0.908 | 3.722 | |||
| Device barrier (DB) | 0.850 | 0.892 | 0.675 | ||
| DB1 | 0.825 | 1.418 | |||
| DB2 | 0.800 | 2.343 | |||
| DB3 | 0.821 | 2.311 | |||
| DB4 | 0.838 | 2.377 | |||
| Behavioural intention (BI) | 0.894 | 0.927 | 0.76 | ||
| BI1 | 0.893 | 2.751 | |||
| BI2 | 0.850 | 2.173 | |||
| BI3 | 0.883 | 2.725 | |||
| BI4 | 0.860 | 2.288 | |||
| Use behaviour (UB) | 0.721 | 0.837 | 0.632 | ||
| UB1 | 0.804 | 1.611 | |||
| UB2 | 0.724 | 1.433 | |||
| UB3 | 0.852 | 1.335 | |||
Note(s): All factor loadings are statistically significant at a significance level of 0.05; AVE, “average variance extracted; ” VIF, “variance inflation factor; ”
Source(s): Authors’ own creation
Discriminant validity
| ANX | ATT | BI | DB | PC | PEoU | PI | PR | PT | PU | SE | SI | UB | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ANX | 0.893 | ||||||||||||
| ATT | 0.069 | 0.907 | |||||||||||
| BI | 0.286 | 0.292 | 0.872 | ||||||||||
| DB | 0.328 | −0.147 | 0.216 | 0.821 | |||||||||
| PC | −0.125 | 0.220 | −0.299 | −0.300 | 0.850 | ||||||||
| PEoU | 0.133 | 0.345 | 0.278 | 0.089 | −0.074 | 0.836 | |||||||
| PI | 0.246 | 0.278 | 0.117 | −0.077 | 0.411 | 0.054 | 0.863 | ||||||
| PR | −0.123 | 0.307 | −0.149 | −0.363 | 0.444 | 0.009 | 0.417 | 0.817 | |||||
| PT | 0.239 | 0.046 | 0.334 | 0.101 | −0.016 | 0.207 | 0.051 | 0.018 | 0.819 | ||||
| PU | 0.345 | 0.304 | 0.690 | 0.285 | −0.244 | 0.489 | 0.087 | −0.105 | 0.328 | 0.828 | |||
| SE | 0.292 | 0.038 | 0.313 | 0.192 | −0.170 | 0.305 | −0.035 | −0.300 | 0.186 | 0.363 | 0.853 | ||
| SI | 0.363 | 0.124 | 0.603 | 0.417 | −0.288 | 0.181 | 0.061 | −0.261 | 0.186 | 0.539 | 0.239 | 0.862 | |
| UB | 0.294 | 0.122 | 0.569 | 0.167 | −0.238 | 0.399 | 0.052 | −0.198 | 0.237 | 0.443 | 0.375 | 0.381 | 0.795 |
Source(s): Authors’ own creation
R2, adjusted R2 and Q2
| R2 | Adjusted R2 | Q2 (=1-SSE/SSO) | |
|---|---|---|---|
| ATT | 0.143 | 0.139 | 0.113 |
| BI | 0.605 | 0.586 | 0.438 |
| PEoU | 0.138 | 0.091 | 0.056 |
| PU | 0.239 | 0.237 | 0.161 |
| UB | 0.324 | 0.323 | 0.189 |
Note(s): SRMR = 0.066; SSE = sum of squared errors; SSO = sum of squares of off-diagonal elements
Source(s): Authors’ own creation
Path analysis
| H | Relationships | β | STDEV | t-value | p-value | Results |
|---|---|---|---|---|---|---|
| Direct effects | ||||||
| H1a | PEoU→ATT | 0.258 | 0.064 | 4.016 | 0.000 | Supported |
| H1b | PEoU→PU | 0.489 | 0.049 | 10.059 | 0.000 | Supported |
| H2a | PU→ATT | 0.179 | 0.051 | 3.476 | 0.001 | Supported |
| H2b | PU→BI | 0.423 | 0.055 | 7.688 | 0.000 | Supported |
| H3 | ATT→BI | 0.133 | 0.033 | 4.080 | 0.000 | Supported |
| H4 | PT→BI | 0.121 | 0.036 | 3.327 | 0.001 | Supported |
| H5 | PI→BI | 0.126 | 0.073 | 1.729 | 0.084 | Not supported |
| H6 | SI→BI | 0.310 | 0.078 | 3.968 | 0.000 | Supported |
| H7a | SE→BI | 0.056 | 0.037 | 1.538 | 0.125 | Not supported |
| H7b | SE→PEoU | 0.305 | 0.051 | 6.004 | 0.000 | Supported |
| H8 | DB→BI | −0.080 | 0.037 | 2.151 | 0.032 | Supported |
| H9 | ANX→BI | −0.044 | 0.036 | 1.225 | 0.221 | Not supported |
| H10 | PC→BI | −0.182 | 0.038 | 4.781 | 0.000 | Supported |
| H11 | PR→BI | −0.056 | 0.044 | 1.256 | 0.210 | Not supported |
| H12 | BI→AU | 0.570 | 0.048 | 11.90 | 0.000 | Supported |
| Interaction Effects | ||||||
| H13 | BI × Education→AU | 0.025 | 0.039 | 0.649 | 0.517 | Not supported |
| H14 | BI × Income→AU | 0.200 | 0.041 | 4.893 | 0.000 | Supported |
| H15 | BI × Experience→AU | 0.133 | 0.055 | 2.416 | 0.016 | Supported |
Note(s): p < 0.05, Supported; p > 0.05, Not supported
Source(s): Authors’ own creation
References
Abegao Neto, F.L. and Figueiredo, J.C.B.d. (2022), “Effects of age and income moderation on adoption of mobile payments in Brazil”, Innovation and Management Review, Vol. ahead-of-print No. ahead-of-print, doi: 10.1108/INMR-06-2021-0109.
Al-Saedi, K., Al-Emran, M., Ramayah, T. and Abusham, E. (2020), “Developing a general extended UTAUT model for M-payment adoption”, Technology in Society, Vol. 62, 101293, doi: 10.1016/j.techsoc.2020.101293.
Alalwan, A.A., Dwivedi, Y.K. and Rana, N.P. (2017), “Factors influencing adoption of mobile banking by Jordanian bank customers: extending UTAUT2 with trust”, International Journal of Information Management, Vol. 37 No. 3, pp. 99-110, doi: 10.1016/j.ijinfomgt.2017.01.002.
Alba, R.M. (2003), “Evolution of methods of money laundering in Latin America”, Journal of Financial Crime, Vol. 10 No. 2, pp. 137-140.
Alhassany, H. and Faisal, F. (2018), “Factors influencing the internet banking adoption decision in North Cyprus: an evidence from the partial least square approach of the structural equation modeling”, Financial Innovation, Vol. 4 No. 1, pp. 1-21.
Alshurideh, M.T., Al Kurdi, B., Masa’deh, R. and Salloum, S.A. (2021), “The moderation effect of gender on accepting electronic payment technology: a study on United Arab Emirates consumers”, Review of International Business and Strategy, Vol. 31 No. 3, pp. 375-396, doi: 10.1108/RIBS-08-2020-0102.
Aparna, K., Amilan, S. and Vimal Raj, L. (2023), “Customers' response to mandatory corporate social responsibility in India: an empirical evidence”, Social Responsibility Journal, Vol. 19 No. 3, pp. 429-445, doi: 10.1108/SRJ-04-2021-0174.
Armey, L.E., Lipow, J. and Webb, N.J. (2014), “The impact of electronic financial payments on crime”, Information Economics and Policy, Vol. 29, pp. 46-57, doi: 10.1016/j.infoecopol.2014.10.002.
Bailey, A.A., Pentina, I., Mishra, A.S. and Ben Mimoun, M.S. (2017), “Mobile payments adoption by US consumers: an extended TAM”, International Journal of Retail & Distribution Management, Vol. 45 No. 6, pp. 626-640, doi: 10.1108/IJRDM-08-2016-0144.
Bashir, I. and Madhavaiah, C. (2015a), “Consumer attitude and behavioural intention towards Internet banking adoption in India”, Journal of Indian Business Research, Vol. 7 No. 1, pp. 67-102, doi: 10.1108/JIBR-02-2014-0013.
Bashir, I. and Madhavaiah, C. (2015b), “Trust, social influence, self‐efficacy, perceived risk and internet banking acceptance: an extension of technology acceptance model in Indian context”, Metamorphosis, Vol. 14 No. 1, pp. 25-38, doi: 10.1177/0972622520150105.
BIS Statistics Explorer (n.d.), “CT5: use of payment services/instruments: volume of cashless payments”, available at: https://stats.bis.org/statx/srs/table/CT5
Celik, H. (2016), “Customer online shopping anxiety within the unified theory of acceptance and use technology (UTAUT) framework”, Asia Pacific Journal of Marketing and Logistics, Vol. 28 No. 2, pp. 278-307.
Chawla, D. and Joshi, H. (2019), “Consumer attitude and intention to adopt mobile wallet in India – an empirical study”, International Journal of Bank Marketing, Vol. 37 No. 7, pp. 1590-1618, doi: 10.1108/IJBM-09-2018-0256.
Chawla, D. and Joshi, H. (2020), “The moderating role of gender and age in the adoption of mobile wallet”, Foresight, Vol. 22 No. 4, pp. 483-504, doi: 10.1108/FS-11-2019-0094.
Davis, F.D. (1989), “Perceived usefulness, perceived ease of use, and user acceptance of information technology”, MIS Quarterly, Vol. 13 No. 3, pp. 319-340, doi: 10.2307/249008.
Flavian, C., Guinaliu, M. and Lu, Y. (2020), “Mobile payments adoption–introducing mindfulness to better understand consumer behavior”, International Journal of Bank Marketing, Vol. 38 No. 7, pp. 1575-1599, doi: 10.1108/IJBM-01-2020-00393.
Fornell, C. and Larcker, D.F. (1981), “Evaluating structural equation models with unobservable variables and measurement error”, Journal of Marketing Research, Vol. 18 No. 1, pp. 39-50.
Giovanis, A., Assimakopoulos, C. and Sarmaniotis, C. (2019), “Adoption of mobile self-service retail banking technologies: the role of technology, social, channel and personal factors”, International Journal of Retail and Distribution Management, Vol. 47 No. 9, pp. 894-914, doi: 10.1108/IJRDM-05-2018-00894.
Goel, R.K. and Mehrotra, A.N. (2012), “Financial payment instruments and corruption”, Applied Financial Economics, Vol. 22 No. 11, pp. 877-886, doi: 10.1080/09603107.2011.628295.
Gupta, K. and Arora, N. (2020), “Investigating consumer intention to accept mobile payment systems through unified theory of acceptance model: an Indian perspective”, South Asian Journal of Business Studies, Vol. 9 No. 1, pp. 88-114, doi: 10.1108/SAJBS-03-2019-0037.
Hair, J.F., Black, W.C., Babin, B.J. and Anderson, R.E. (2010), Multivariate Data Analysis: A Global Perspective, 7th ed., Pearson Prentice Hall, Uppersaddle River, NJ.
Hair, J.F. Jr, Hult, G.T.M., Ringle, C.M. and Sarstedt, M. (2014), A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), SAGE Publications, London.
Hair, J.F., Hult, G.T.M., Ringle, C.M. and Sarstedt, M. (2017), A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), Sage, Thousand Oaks, CA.
Koksal, M.H. (2016), “The intentions of Lebanese consumers to adopt mobile banking”, International Journal of Bank Marketing, Vol. 34 No. 3, pp. 327-346, doi: 10.1108/IJBM-03-2015-0025.
Laukkanen, T. (2008), “Determinants of mobile banking resistance: a preliminary model”, InProceedings of ANZMAC (Australian and New Zealand Marketing Academy) Conference, Sydney, Australia, December, pp. 1-3.
Lipow, J. (2010), Turn in Your Bin Ladens, The New York Times, New York, NY, Vol. 18, p. A19.
Lisana, L. (2021), “Factors influencing the adoption of mobile payment systems in Indonesia”, International Journal of Web Information Systems, Vol. 17 No. 3, p. 10, doi: 10.1108/IJWIS-01-2021-0004.
Patil, P., Tamilmani, K., Rana, N.P. and Raghavan, V. (2020), “Understanding consumer adoption of mobile payment in India: extending Meta-UTAUT model with personal innovativeness, anxiety, trust, and grievance redressal”, International Journal of Information Management, Vol. 54, 102144, doi: 10.1016/j.ijinfomgt.2020.102144.
Priya, R., Gandhi, A.V. and Shaikh, A. (2018), “Mobile banking adoption in an emerging economy”, Benchmarking: An International Journal, Vol. 25 No. 2, pp. 743-762, doi: 10.1108/BIJ-01-2016-0009.
Raj, L.V., Amilan, S. and Aparna, K. (2021), “Role of perceived countries' advantages of cashless economy in behavioral intentions of using cashless transactions: an empirical analysis”, Journal of Indian Business Research, Vol. 13 No. 3, pp. 413-433, doi: 10.1108/JIBR-06-2020-0186.
Raj, L.V., Amilan, S., Aparna, K. and Swaminathan, K. (2023), “Factors influencing the adoption of cashless transactions during COVID-19: an extension of enhanced UTAUT with pandemic precautionary measures”, Journal of Financial Services Marketing. doi: 10.1057/s41264-023-00218-8.
Reserve Bank of India Annual Report (2017), “Payment and settlement systems and information technology”, p. 156, available at: https://rbidocs.rbi.org.in/rdocs/AnnualReport/PDFs/0ANREPORT201718077745EC9A874DB38C991F580ED14242.PDF
Reserve Bank of India Annual Report (2021), “Payment and settlement systems and information technology”, p. 176, available at: https://rbidocs.rbi.org.in/rdocs/AnnualReport/PDFs/0RBIAR2021226AD1119FF6674A13865C988DF70B4E1A.PDF
Reserve Bank of India Bulletin (2019), “Drivers of digital payments: a cross country study”, p. 27, available at: https://www.rbi.org.in/Scripts/BS_ViewBulletin.aspx?Id=18409
Saha, T., Dey, T. and Hoque, M.R. (2022), “Initial trust and usage intention: a study on mobile payment adoption in Bangladesh”, Global Business Review, 09721509221120805, doi: 10.1177/09721509221120805.
Sarmah, R., Dhiman, N. and Kanojia, H. (2021), “Understanding intentions and actual use of mobile wallets by millennial: an extended TAM model perspective”, Journal of Indian Business Research, Vol. 13 No. 3, pp. 361-381, doi: 10.1108/JIBR-06-2020-0214.
Shin, D.H. (2009), “Towards an understanding of the consumer acceptance of mobile wallet”, Computers in Human Behavior, Vol. 25 No. 6, pp. 1343-1354, doi: 10.1016/j.chb.2009.06.001.
Singh, S. and Srivastava, R.K. (2018), “Predicting the intention to use mobile banking in India”, International Journal of Bank Marketing, Vol. 36 No. 2, pp. 357-378, doi: 10.1108/IJBM-12-2016-0186.
Sivathanu, B. (2019), “Adoption of digital payment systems in the era of demonetization in India: an empirical study”, Journal of Science and Technology Policy Management, Vol. 10 No. 1, pp. 143-171, doi: 10.1108/JSTPM-07-2017-0033.
Sripalawat, J., Thongmak, M. and Ngramyarn, A. (2011), “M-banking in metropolitan Bangkok and a comparison with other countries”, Journal of Computer Information Systems, Vol. 51 No. 3, pp. 67-76, doi: 10.1080/08874417.2011.11645487.
Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, F.D. (2003), “User acceptance of information technology: toward a unified view”, MIS Quarterly, Vol. 27 No. 3, pp. 425-478, doi: 10.2307/30036540.
Venkatesh, V., Thong, J.Y. and Xu, X. (2012), “Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology”, MIS Quarterly, Vol. 36 No. 1, pp. 157-178, doi: 10.2307/41410412.
Vimal Raj, L., Amilan, S., Aparna, K., et al., (2023), “Developing and validating a cashless transaction adoption model (CTAM)”, 25 January 2023, PREPRINT (Version 1) available at: Research Square. doi: 10.21203/rs.3.rs-2510615/v1.
Warwick, D.R. (1993), Reducing Crime by Eliminating Cash, National Council on Crime and Delinquency, San Francisco, CA, available at: https://eric.ed.gov/?id=ED377114
Williams, M.D. (2021), “Social commerce and the mobile platform: payment and security perceptions of potential users”, Computers in Human Behaviour, Vol. 115, 105557, doi: 10.1016/j.chb.2018.06.005.
Wright, R., Tekin, E., Topalli, V., McClellan, C., Dickinson, T. and Rosenfeld, R. (2017), “Less cash, less crime: evidence from the electronic benefit transfer program”, The Journal of Law and Economics, Vol. 60 No. 2, pp. 361-383, doi: 10.1086/693745.