Exploring the Determinants of FinTech Adoption Among University Students: A Second-Order Construct Analysis

Abstract

How individuals and organizations interface with the digital economy has been largely influenced by transformations ushered in on the global financial map by the rapidly expanding Financial Technology (FinTech). This paper seeks to shed light on the successes of FinTech, namely on how it contributed to sustainability through financial inclusion, reduction in reliance on cash and the promotion of an innovation-driven economy known for being paperless. Based on contributions from students at Taif University in Saudi Arabia, determinants of FinTech adoption intentions are analyzed using data from n = 544. Our study focuses on evaluating the effects of financial, technical and external factors on adoption behavior by using a two-prong approach: first, we use the DeLone and McLean IS Success Model; then we employ a Second-Order Construct using Structural Equation Modelling (SEM). The results indicated that the strongest effects on attitudes stem from technical factors—information, system and service quality. Additionally, they also show that adoption intention is considerably shaped by financial as well as external dimensions. The Saudi Vision 2030 has set national goals of digital transformation, financial inclusion and human capital empowerment. This study provides a modest contribution to those goals by fostering FinTech adoption among the youth. Furthermore, its findings also offer educators, policymakers and Fintech providers a platform to enhance literacy, strengthen trust and develop sustainable digital finance ecosystems in line with the Kingdom’s Vision 2030 objectives.

1. Introduction

The advent of FinTech has contributed to a global digital finance revolution that shapes how governments, institutions and individuals engage with financial services. An instance of this is illustrated by FinTech’s ushering of an environment where inclusion, transparency and environmentally responsible cash-free transactions are the norm [1]. The FinTech industry is particularly influential in fostering financial inclusion by providing accessible, convenient, and personalized services. As technology use expands across various demographics, understanding factors influencing the adoption of FinTech among different user segments has become a focal point of academic and industry research [2]. Financial technology is reshaping the global financial services landscape in multiple ways, including enhancing efficiency and innovation, advancing sustainability goals such as promoting financial inclusion and narrowing the digital divide, and fostering the transition toward a more equitable and resilient digital economy [3]. Students are the litmus test in this issue because they represent a critical demographic in shaping a sustainable financial and social future. In Saudi Arabia, FinTech is emerging as a critical sector, driven by Vision 2030, a national initiative aimed at diversifying the economy and promoting digital transformation across all sectors [4]. This initiative promotes the development of digital solutions in finance, aiming to modernize financial services, enhance accessibility, and support small and medium-sized enterprises. With the youth population representing a significant segment of the country’s demographics [5], university students are a vital group for the adoption of FinTech. University students born in the digital age are more likely to adopt financial innovations; therefore, it is of paramount importance to understand their motivations as well as perceived obstacles in order to implement successful FinTech and financial literacy initiatives [6,7]. However, understanding their specific motivations, concerns, and potential barriers is crucial to promoting effective adoption. Within that context, this study seeks to fill a research gap created by prior research that lacked the foresight to study the hierarchical relationships between determinants but focused instead on treating determinants independently. Such prior research includes those that applied the Technology Acceptance Model (TAM) and Unified Theory of Acceptance and Use of Technology (UTAUT) [8,9]. Additionally, this research coalesces the financial, technical and external aspects of SOFA by extending the DeLone and McLean Information Systems Success Model. Further, in Saudi Arabia’s conservative cultural context, factors such as perceived security, religious considerations, and societal norms may also shape FinTech adoption intentions [10]. This study explored these factors in detail, focusing on Saudi university students to understand their unique perceptions and behavioral intentions regarding FinTech services.

It is against the background of Saudi Vision 2030 that this study aims to shed light on the adoption of youth-driven FinTech. This approach should impact educators, FinTech providers and regulators such as the Saudi Central Bank (SAMA) and the Capital Market Authority (CMA). Ultimately, we aim to offer valuable recommendations for FinTech providers, educators, and policymakers to promote adoption, enhance financial literacy, and align FinTech solutions with the preferences and needs of this demographic group. Although the FinTech adoption has been extensively discussed in the literature by incorporating the TAM and the UTAUT [8,9], there is a lack of literature on the topic in the Saudi higher education sector. Instead, prior research has focused on the general population and customers. Consequently, there is a clear need for investigations into motivations, adoption approaches, and awareness of barriers faced by university students regarding FinTech services.

The approach that has recently been followed by researchers in the area of shaping digital financial behavior has been to underscore the importance of FinTech adoption. However, such an approach has often been developed in the context of isolated technological or behavioral lenses [11]. In line with contemporary research, and in the context of emerging economies, we propose integrated models capable of fusing multidimensional factors that influence adoption [12,13]. As mentioned earlier, previous studies have failed to look into users’ behavioral intentions from an integrative perspective. Thus, the issue of how financial, technical and external factors form an integral part of the equation is often ignored, thereby creating a research gap that we hope to fill by adopting a Second-Order Construct framework based on the DeLone and McLean IS Success Model. Our approach, we hope, will provide a more holistic understanding of FinTech adoption among Saudi university students. In the end, we hope that our research combines FinTech adoption, digital transformation, financial inclusion and human capital empowerment into one symbiotic entity that aligns perfectly with the Saudi Vision 2030. Thus, this study enhances the evaluation of FinTech adoption by offering a deeper and more complete framework for understanding the concept.

The purpose of this research is to investigate the key factors that influence the intention of business students at Taif University to utilize FinTech services in Saudi Arabia. Specifically, we sought to:

• Evaluate the roles of financial elements, such as financial understanding, fiscal teaching, and fiscal risk, in deciding how and when to utilize different FinTech products.
• Understand components of technical factors such as information quality, system quality, and service quality, and how these factors affect adoption intentions.
• Examine the influence of social factors, facilitating conditions, and brand trust on the attitude of users to the FinTech services.
• Apply Second-Order Construct in (SEM) to provide a deeper and more intensive, more appropriate approach than previous research to study Fintech adoption.

In achieving these objectives, this research enhances FinTech adoption models with the Second-Order Construct approach to advance existing literature on FinTech. Our conclusions offer valuable implications for addressing issues of financial illiteracy, advancing our understanding of financial technologies, and developing permissive legislation for these technologies.

2. Literature Review

The post-pandemic era has put the FinTech industry at the heart of sustainable finance, digital inclusion and entrepreneurship. This section presents a preliminary review of the prior literature on FinTech, focusing on several key determinants (e.g., financial literacy, trust, security, perceived usefulness, and social influence). In addition, we explain Second-Order Constructs and the DeLone and McLean IS Success Model in the context of FinTech adoption.

2.1. Determinants of FinTech Adoption

2.1.1. Financial Factors in FinTech Adoption

Knowledge about finance and money matters is a determining factor in the users’ confidence and decision-making when it comes to obtaining financial services through technology. Ref. [14] found that customers willing to adopt FinTech services are typically financially literate individuals who can assess the risks and benefits associated with FinTech. Likewise, ref. [15] mentioned that financial education enables university students to safely adopt FinTech by establishing structured programs in financial education and digitally enabled financial activities. Although there are numerous advantages to developing financial literacy, the risks are often perceived as factors that prevent people from adopting FinTech services. According to ref. [16] Potential users are slow to adopt FinTech because of the difficulties and risks associated with fraud, secure financial transactions, and lost funds. This hesitation is especially noticeable in emerging countries, where the rules, norms, and regulations governing data protection are weaker compared to those in more advanced countries [17].

2.1.2. Technical Factors: System, Information, and Service Quality

The technological quality of FinTech services refers to users’ confidence in and satisfaction with the services offered. From the literature, ref. [14] observed that several factors influence the adoption of FinTech, one of which is system quality (i.e., usability, efficiency, and reliability). Similarly, ref. [12] suggested that information quality (specifically, the accuracy, relevance, and completeness of financial data) is conducive to improving user confidence in FinTech applications. Service quality, including the timeliness of customer response and the solving of customer issues, is another important factor in determining Fintech adoption. Ref. [18] also emphasized that service quality enhances the level of confidence and satisfaction among users and consequently improves adoption decisions. These observations strongly suggest that, for FinTech platforms to succeed, they must be secure, efficient, and user-friendly.

2.1.3. Other Factors: Social Influence and Trust in FinTech

Several behavior-shaping factors are at play in the adoption of FinTech among students; these include social endorsement, cultural norms, perceived trust, peer and family influence, on the one hand [19,20]. On the other hand, brand credibility and platform security assure long-term trust [21]. This suggests that user confidence in sustainable digital finance ecosystems is multidimensional in nature. Furthermore, this user confidence is also determined within the GCC societies by such factors as social validation [9], ethical standards and Shariah-compliant design [10,22].

2.1.4. Cultural and Regulatory Factors in Saudi Arabia’s FinTech Adoption

Saudi Arabia has various cultural and regulatory conditions that impact the use of FinTech. According to ref. [23], the two most dominant factors influencing Saudi consumers in adopting FinTech services are perceived usefulness and perceived social influence. Still, issues related to do with financial ethical standards, data privacy, and Shariah compliance are some of the challenges that are still notable [14]. This is because, in the use of FinTech services, trust is attributable to the reputation of the brand and the security of the system in place [2,22]. Indeed, Odoom and Kosiba [20] argued that perceived trust is directly related to consumers’ confidence in digital transactions and found that consumers are more likely to engage with firms with good policy approaches to data protection. Similar trends in the adoption of FinTech have been seen in other GCC countries, like the UAE and Bahrain. These countries’ digital transformation efforts and policies for financial inclusion have sped up financial innovation [24,25].

2.2. Second-Order Constructs in FinTech Research

Prior research in the FinTech context has primarily employed first-order constructs to measure the factors influencing the adoption of FinTech. However, a new second-order construct has recently been proposed regarding the adoption factor, offering a more hierarchical and structural perspective. Ref. [26] has noted that Second-Order Constructs make it possible to aggregate some related variables (financial literacy, financial education, and financial risk) into the higher-order factor. It is beneficial to enhance the overall quality of the structural models and provide clearer insights regarding the underlying variables. In the same vein, ref. [27] also supported this analysis by establishing that Technical Factors, which include information quality, system quality, and service quality, should be treated as a Second-Order Construct because they predict users’ trust and perceived usability.

2.3. DeLone and McLean’s Information Systems (IS) Success Model

Frameworks are employed to analyze the FinTech adoption intentions of university students. This methodology was developed to assess the efficacy of information systems and is currently favored by scholars researching FinTech [28]. It delineates variables such as perceived system quality, information quality, and perceived service quality. These variables, in turn, directly influence user satisfaction and usage. This study seeks to enhance the current DeLone and McLean Model by incorporating two additional and potentially relevant variables. In particular, we examine the relationship of financial and external factors to the acceptability of FinTech. This framework facilitates comprehension of user behavior by incorporating both technical and socio-economic elements. Thus, this study utilizes financial literacy, system quality, trust, and social impact as key financial variables that influence the adoption of FinTech. Previous research has focused on first-order constructs [29,30,31] and, as a consequence, may inadequately represent the various relationships discussed above. By contrast, this study employs the DeLone and McLean Success Model, along with Second-Order Construct analyses, to address this literature gap. In addition, more rigorously than prior work, we examine FinTech uptake among Saudi university students by employing the DeLone and McLean Success Model with Second-Order Construct analysis.

A holistic, eclectic approach to explaining FinTech adoption has become the norm—a norm that combines technological, behavioral and contextual determinants. Seminal work by ref. [27] brings together core drivers, technology/system qualities, user attitudes and trust, social influencers and financial literacy/risk. Our approach in this study was to set a DeLone & McLean IS Success Paths background against which we can align technical determinants: information, system and quality. This justifies their inclusion as second-order facets [12]. While behavioral determinants such as attitude and trust remain central in the post-pandemic era [13], external/social determinants, including social influence and brand trust, are more prevalent among the youth as well as the influencer-rich settings. Finally, it is also important to note that financial determinants, including financial literacy/education and perceived risk, are often associated with adoption or market-level growth of FinTech [32]. This fact lends credence to our second-order specification in the sense that it coalesces highly correlated first-order indicators into broader dimensions that are better suited to demonstrate the hierarchical structure of adoption determinants. Furthermore, this fact also strongly suggests demographic nuancing (e.g., gender) as a boundary condition for effects [33].

3. Theoretical Framework and Hypotheses Development

3.1. Theoretical Framework

3.1.1. DeLone and McLean IS Success Model

The DeLone & McLean IS Success Model (The DeLone and McLean Information Systems Success Model defines IS success as a multidimensional concept that includes system quality, information quality, service quality, system use, user satisfaction, and net benefits. The model suggests that high-quality systems produce high-quality information and services, which in turn lead to greater user satisfaction and use, ultimately resulting in positive individual and organizational impacts. This integrated framework provides a comprehensive way to evaluate the effectiveness and value of information systems within an organization.) has been identified as the most widely employed instrument for measuring the results of IS and the impact these have on users [28,34]. The model has three core elements that consider systems and their success factors:

• System Quality: The usability, reliability, adaptability, and response time.
• Information Quality: The accuracy, relevance, and timeliness of financial information.
• Service Quality: The care given to the clients, staff, and the extent to which the system attends to its clientele promptly.

These dimensions, therefore, affect user satisfaction, trust, and behavioral intention, and are ideal for investigating FinTech adoption [35]. Because most IS are in the digital environment, the model incorporates new aspects and factors, including financial and external factors, into existing models. In doing so, the model offers a more comprehensive systems view of the adoption process than previous models.

3.1.2. The Role of Second-Order Constructs

Most prior research on themes of FinTech adoption analyzed the influence of the factors as first-order constructs [26,29]. However, a more hierarchical structure is needed to show the interdependency among these factors. This paper uses Second-Order Constructs (In SEM, first-order factors are latent variables directly measured by observed indicators, representing specific dimensions of a construct. In contrast, second-order factors are higher-level latent constructs inferred from multiple first-order factors. They capture a broader, more abstract concept by explaining the relationships among the first-order factors. This hierarchical modeling enables researchers to represent complex, multidimensional constructs more effectively [36].) to refine determinants into three categories of factors as shown below:

• Financial Factors include three factors: financial literacy, financial education, and financial risk.
• Technical factors have three factors, namely information quality, system quality, and service quality.
• Other factors include social influence, facilitating conditions, and brand and service trust.

Using these determinants as second-order constructs enhances model fitness and advances our empirical understanding of the relationships among the above factors [14]. Therefore, to direct the subsequent investigation of our research questions, this study conceptualizes these factors into three Second-Order Constructs as shown in Figure 1. These factors, with attitudes as a moderator variable, are expected to positively influence adoption intention. Therefore, to direct our investigation of the research questions, these factors are conceptualized as the three Second-Order Constructs, and attitudes are proposed to serve as moderators.

Figure 1. Conceptual Framework.

3.2. Hypotheses Development

Based on Second-Order Constructs, this research incorporates the DeLone and McLean IS Success Model to establish a FinTech usage model for university students. For this reason, structuring the financial, technical, and other factors into higher-order constructs makes the models of adoption more accurate and provides a more systematic portrait of user behavior than in prior studies. The key determinants for the proposed hypotheses will be tested via SEM. This technique enables inferences about the relationships among variables and the moderating role of adoption intentions.

3.2.1. Technical Factors as a Second-Order Factor

The level of financial literacy and education helps to determine the level of interaction of users in digital financial services [14]. Specifically, users’ financial literacy significantly influences their usage and adoption of FinTech solutions [15]. Perceived financial risk may serve as a variable that constrains the likelihood of FinTech adoption [16]. Thus, this study utilized three factors to present the influence of financial factors: financial education (FE), financial literacy (FL), and financial risk (FR). From these, the following hypotheses were developed.

H1. 

The three distinct, but related, sub-dimensions of financial factors can be accounted for by a common underlying higher-order financial factor model, which is significantly better than a first-order financial factor model.

3.2.2. Financial Factors as a Second-Order Factor

If a system has a high level of security, especially in transactions, users will likely trust it and recommend it to others. Based on the DeLone and MacLean IS Success model, the technical factors of FinTech services are expected to significantly influence user satisfaction and usage, including (1) system quality (STQ), (2) information quality (IFQ), and (3) service quality (SVQ) [29,37]. So, the following hypothesis is developed.

H2. 

The three distinct, but related, sub-dimensions of technical factors can be accounted for by a common underlying higher-order technical factor model, which is significantly better than a first-order technical factor model.

3.2.3. Other Factors as a Second-Order Factor

Psychological factors, such as user trust and perceived facilitating conditions, also influence adoption behavior [19,20]. These findings demonstrate that the intention of FinTech users to adopt FinTech services is significantly influenced by perceptions of social influences, including the perceived credibility of available brands, primarily in the financial sector, and the perceived availability of FinTech resources. As a consequence, we suggest three social-perceptual variables to present the influence of other factors, which are social influence (SI), facilitating conditions (FC), and brand & service trust (BS), so the following hypotheses are developed.

H3. 

The three distinct, but related, sub-dimensions of other factors can be accounted for by a common underlying higher-order other factor model, which is significantly better than a first-order factor model.

3.2.4. Financial Factors and Users’ Attitude

Extant research shows that financial factors, such as affordability, cost transparency, and financial incentives users’ attitudes toward adopting digital services [17,29]. This prior research suggested the following hypothesis:

H4. 

There are positive relationships between financial factors and users’ attitudes.

3.2.5. Technical Factors and Users’ Attitude

Technical factors are considered important influences on users’ attitudes and their intention to adopt digital services. According to ref. [11], a well-designed, easy-to-use interface encourages the use of digital environments, leading to the successful adoption of digital services. In addition, ref. [37] highlights positive relationships between system reliability and users’ attitudes toward the digital environment. We expected these variables to be positively correlated and form a second-order technical factor. This led to the following hypothesis: This study, thus, highlights the positive influence of technical factors, as a second-order factor, on users’ attitude, so the following hypothesis is developed.

H5. 

Technical factors have a positive influence on users’ attitudes.

3.2.6. Other Factors and FinTech Adoption Intention

Other social-perceptual variables, such as social influence—specifically, whether others believe services are secure, whether others have adopted them, and the extent to which others endorse those services—and trust may also directly affect adoption decisions. The probability that these are positively related variables is supported by some prior findings (e.g., [21,38]). From this, we derived the following hypothesis:

H6. 

Social variables affect users’ intentions of adopting FinTech services.

3.2.7. Attitude and FinTech Adoption Intention

This paper also addresses the issue of whether user attitudes towards FinTech moderate the variable that enhances understanding of the relationships among external factors, financial factors, and technical factors in relation to intended adoption. High perceived levels of information technology have been found to enhance the usage of financial services, as illustrated by refs. [9,29].

H7. 

Attitude has a positive impact on the intention to adopt FinTech.

4. Research Methodology

4.1. Measures

Surveys are the most prevalent quantitative research methodology for data collection and are commonly used in social science research. This study utilized a five-point Likert scale questionnaire as a quantitative data gathering instrument [39]. Each item is a statement that requires the respondent to indicate their level of agreement or disagreement on a scale from 1 to 5 (e.g., 1 = strongly agree, 2 = agree, 3 = neutral, 4 = disagree, 5 = strongly disagree). Three main independent components were analyzed—Financial factors: (Financial education (FE), Financial literacy (FL), Financial Risk (FR)), Technical factors: (Information quality (IFQ), System quality (STQ), Service quality (SVQ)), Other factors: (Social influence (SI), Facilitating conditions (FC), Brand & service trust (BS))—alongside two dependent constructs: Attitude and FinTech Adoption Intention. Each concept was assessed using three items.

To determine the internal consistency (i.e., Cronbach’s alpha) of each scale, reliability tests were conducted to assess whether the questionnaire meets the study’s goals. The research was employed for the reliability assessment. The threshold for Cronbach’s alpha is 0.70. Consequently, the surveys were deemed reliable if Cronbach’s alpha was >0.70.

4.2. Sample

A probability sampling involves selecting individuals to be representative of the population. The method utilized most often to produce findings that are trustworthy and robust is probability sampling. According to ref. [40], larger samples provide a more accurate representation of the population because they increase the generalizability and reliability of the findings. Fully completed surveys were received from 544 Business students at Taif University (Saudi Arabia).

4.3. Data Collection

The questionnaire was made available to respondents through Google Forms, email, and Moodle. After the questionnaire was distributed, respondents ‘ personal information was removed to maintain anonymity. The questionnaire constructs and their associated items are presented in Table 1 and Appendix A.

Table 1. Questionnaire.

Variables	ID	Items Description	References
Financial factors:
Financial education	(FE)	It is essential for university students, as the ability to manage and utilize financial resources is crucial for the effective operation of any organization, and students are seen as future entrepreneurs.	[41]
Financial literacy	(FL)	Can be described as “measuring how well an individual can understand and use personal finance-related information” to make a decision.	[42]
Financial Risk	(FR)	It refers to the potential financial loss in the financial transactions of Fintech.	[43]
Technical factors:
Information quality	(IFQ)	It pertains to an individual’s assessment of service providers’ capacity to fulfill the user’s requirements.	[37]
System quality	(STQ)	It pertains to impressions formed by the whole operation of IT systems.	[37]
Service quality	(SVQ)	It is characterized as an individual’s assessment of the extent of assistance provided by an Information Systems department and its IT support framework.	[44]
Other factors:
Social influence	(SI)	The effect of a person’s family and friends on their decision to employ a specific system is what this term alludes to “the degree to which individuals recognize the need to use a system, since they notice that other relevant people are already using it”.	[19,20]
Facilitating conditions	(FC)	The available resources a person has to perform any particular behavior. “the degree to which an individual notices that a system can be easily controlled if supported by the organization and its technical infrastructure”.	[19,20]
Brand and service trust	(BS)	“The degree of effect on consumer trust toward enterprises that provide Fintech Service by brand and system security”.	[21,22,45]
Attitude	(ATT)	Attitude refers to the user’s subjective judgments and personal tendencies related to something.	[46]

In order to test a theoretically grounded hierarchical structure that linked technical qualities (namely, information, system and service quality), financial factors, external/social influences, attitudes and adoption intention, our research resorted to a covariance-based structural equation modeling (CB-SEM) with a second-order (higher-order) specification. This choice of modeling was based on our conviction that it was better-suited for the purposes of theory testing. This made possible such things as model-fit evaluation and explicit treatment of latent-variable measurement error [36]. In addition, a second-order CFA/SEM framework also captures the shared variance among correlated first-order dimensions, providing a parsimonious representation of broader constructs while reducing multicollinearity. On the one hand, in comparison to multiple regression, SEM can be used to simultaneously estimate the measurement and structural models; on the other hand, compared with EFA, it is confirmatory. Therefore, although PLS-SEM is particularly suited for prediction-oriented research, the evidence shows that for confirmatory, theory-driven investigations as well as for comprehensive global fit assessment, CB-SEM is better equipped [47]. Finally, based on recent methodological recommendations, our research investigates robustness considerations and potential alternative representations (e.g., bifactor models).

5. Results

This study utilizes AMOS 21 (SEM) software to analyze the proposed model through three successive subsections about reliability and validity, (SOFA) and SEM. However, prior to the SEM analysis, our research tested the data set with a view to confirming compliance with key statistical assumptions. The results were as follows: all variables exhibited acceptable levels of univariate normality, with skewness values ranging between −1.84 and +1.73 and kurtosis values between −2.35 and +2.67. Those results fell within the commonly accepted thresholds of ±2 for skewness and ±3 for kurtosis [36]. In addition, we conducted a multicollinearity diagnostics using the Variance Inflation Factor (VIF), where all constructs recorded VIF values below 3.0. These results confirmed the absence of significant multicollinearity issues among predictors. The overall results collectively support the adequacy of the dataset for subsequent SEM analysis while ensuring the robustness of parameter estimation.

5.1. Reliability and Validity Tests

The model reliability was evaluated through the Composite Reliability assessment. SEM utilizes Composite reliability (CR) to determine the degree to which latent construct indicators are internally consistent (i.e., measurement reliability). The CR value must exceed 0.80 to establish reliability according to accepted standards, with higher values indicating stronger reliability. With the exception of financial risk (0.757), the model constructs exhibit excellent reliability, as shown in Table 2.

Table 2. Composite Reliability and Convergent Validity.

Construct	CR	AVE
Financial Education (FE)	0.808	0.517
Financial Literacy (FL)	0.803	0.507
Financial Risk (FR)	0.757	0.510
Information Quality (IFQ)	0.834	0.558
System Quality (STQ)	0.851	0.594
Service Quality (SVQ)	0.910	0.716
Social Influence (SI)	0.834	0.558
Facilitation Conditions (FC)	0.805	0.508
Brand and Service Trust (BS)	0.863	0.678
Attitude (ATT)	0.843	0.643
Adoption Intention of FinTech (AI)	0.920	0.793

This research employs two validity evaluation methods: convergent validity and discriminant validity to assess the model design. Convergent validity demonstrates that a set of variables representing a construct accurately measures its corresponding core concept. Convergent validity is examined through Average Variance Extracted (AVE). Specifically, good convergent validity is suggested when a value exceeds 0.50; thus, the construct explains better than half of the variance among its indicators. As shown in Table 3, Most variables presented demonstrated good to excellent convergent validity. Although the Financial Literacy construct demonstrates almost acceptable validity, researchers can still refine it with further studies.

Table 3. Discriminate Validity.

Title 1	FE	FL	FR	IFQ	STQ	SVQ	SI	FC	BS	ATT	AI
FE	0.719
FL	0.602	0.712
FR	0.400	0.638	0.714
IFQ	0.450	0.333	0.284	0.747
STQ	0.630	0.503	0.420	0.605	0.771
SVQ	0.548	0.451	0.368	0.678	0.735	0.846
SI	0.570	0.446	0.401	0.551	0.683	0.604	0.747
FC	0.525	0.419	0.393	0.608	0.618	0.597	0.668	0.713
BS	0.508	0.392	0.274	0.560	0.536	0.628	0.556	0.657	0.823
ATT	0.645	0.584	0.446	0.456	0.720	0.642	0.720	0.689	0.640	0.802
AI	0.620	0.489	0.331	0.439	0.686	0.597	0.647	0.602	0.602	0.800	0.891

Discriminant validity determines whether a construct is unique from other constructs in the model. We used the Fornell-Larcker criterion to contrast the correlations between constructs and the square root of the Average Variance Extracted (AVE) for each construct. If the square root of the AVE for each construct is higher than the correlations between that construct and other constructs, then that concept exhibits discriminant validity. As shown in Table 3, the discriminant validity analysis indicated that every construct satisfied the Fornell-Larcker criterion and was unique from the other constructs.

5.2. Second Order Factor Analysis (SOFA)

First-order factor analysis is the basis for (SOFA). The first-order factor analysis identifies the variable underlying the variance in the observed data. These first-order factors are groups of related variables. By contrast, SOFA entails conducting a factor analysis on the constructs discovered in the first-order analysis part of the investigation. SOFA is intended to identify hidden, more fundamental components that may explain the correlations between the first-order factors. For the sake of simplicity, SOFA facilitates the discovery of more intricate structural patterns within a dataset.

Following identification of first-order factors, a (SOFA) is performed to ascertain whether or not the original factors can be classified into more generic and all-encompassing (second-order factors) dimensions [36]. Three steps must be taken to test a Second-Order Factor Model. To begin, it is necessary to specify both the first-order factor model and the second-order factor model. The first step is to specify the first-order factors by using the observed variables and the Second-Order Factor Model, which explains the relationships between the first-order factors. Second, the regression weights for both models are examined. These weights are included in the first-order factor loadings to articulate the connections between the observed variables and the first-order factors; high (usually greater than 0.4) values should be the case with these. Significantly high weights in the second-order regression weights describe the extent to which each first-order factor contributes to the second-order factor. Therefore, it investigates these regression weights to evaluate the importance and strength of the second-order correlations. Third, Compare the First- and Second-Order Models (Optional but Recommended).

The overall fit of the model needs to be checked after the model and regression weights have been estimated. Fit indices help determine whether the data are consistent with the second-order factor structure. The second-order factor analysis examines three composite variables: financial considerations, technical factors, and other factors. The following subsections provide an in-depth discussion of the outcomes for each one. The model reliability was evaluated through the Composite Reliability assessment. SEM utilizes Composite reliability (CR) to determine the degree to which latent construct indicators are internally consistent (i.e., measurement reliability). The CR value must exceed 0.80 to establish reliability according to accepted standards, with higher values indicating stronger reliability. With one exception (financial risk, 0.757), the model constructs exhibit excellent reliability, as shown in Table 2.

5.2.1. Financial Factor (FF)/(SOFA)

The first three order factors are FE, FL, and FR. Table 4 displays the factor loadings for (a) first-order relationships between latent constructs and their observed manifest variables, and (b) second-order relationships between the same latent constraints. However, the former is normally higher than the latter when taking into account the relations between Latent Constructs. To set identification equations, FE was regressed on Financial Factors (FF), with the estimate for FE fixed at 1.000. Hypothesis 2 reveals a positive relationship between FL and FF, with an estimated coefficient of 1.119. This means that if FF goes up by 1 unit, FL is expected to increase by 1.119 units. There are positive and significant coefficients between FR and FF with an estimate of = 0.951, which means that with an increase in FF by one unit, FR will also increase by 0.951 units. These results seem to indicate that there is a correlation between increases in the overarching financial factor and proportional increases in literacy and risk awareness.

Table 4. Regression Weights for (FF): (Group number 1—Default model).

			Estimate	S.E.	C.R.	p
FE	←	FF	1.000
FL	←	FF	1.119	0.150	7.450	***
FR	←	FF	0.951	0.107	8.882	***
FE1	←	FE	1.000
FE2	←	FE	0.925	0.054	17.007	***
FE3	←	FE	0.806	0.060	13.390	***
FE4	←	FE	0.797	0.055	14.387	***
FL1	←	FL	1.000
FL2	←	FL	1.067	0.082	13.000	***
FL3	←	FL	1.291	0.092	14.040	***
FL4	←	FL	1.059	0.082	12.836	***
FR1	←	FR	1.000
FR2	←	FR	1.026	0.076	13.566	***
FR3	←	FR	0.786	0.062	12.705	***

* significant at 10% level, ** significant at 5% level, *** significant at 1% level.

The strong positive loading of Financial Literacy on FF correlates with cross-country evidence that higher literacy supports greater FinTech engagement and market expansion. Studies by ref. [11]; ref. [18] show that literacy and education are central to adoption intentions by Gen Y and students. In contrast, the positive correlation between Financial Risk and FF in our measurement model is vastly different from previous studies that report a negative structural path from perceived risk to intention, where greater perceived risk suppresses adoption. This difference is perhaps due to the hierarchical modeling used here, whereby at the measurement level, risk is an integral part of the broader financial awareness domain, although it later acts as an inhibitory determinant in behavioral models.

Table 4 shows that each relationship between the latent constructs and their indicators (based on the first-order factor analysis) is significant (p < 0.001), indicating that the indicators are valid measures of their respective latent constructs. The estimates provide insight into the strength of these relationships, with higher values indicating stronger relationships (Figure 2).

Figure 2. Financial Factor as Second-Order Factor.

According to ref. [46], a combination of CFI > 0.95 and SRMR < 0.08 indicates a good model fit. Table 5 indicates that most indices are excellent fits for the model. RMSEA is slightly above the threshold for an excellent fit but still within the acceptable range. Overall, the model fit is acceptable and effectively represents the underlying data structure.

Table 5. Model fit of the Financial Factors Model.

CMIN/DF	3.000	Excellent
CFI	0.963	Excellent
SRMR	0.048	Excellent
RMSEA	0.061	Acceptable
PClose	0.075	Excellent

5.2.2. Technical Factor (TF)/(SOFA)

Table 6 is an illustration of the unstandardized estimates within the second-order technical domain with the first-order factors (represented by IFQ, STQ, and SVQ) and second-order relations. What was expected here is for the first-order loadings to reflect direct links with indicators, whereas second-order loadings capture the higher-level associations among latent constructs. Insofar as model identification is concerned, TF → IFQ was fixed at 1.000. Thus, the higher-order Technical Factor (TF) shows strong positive effects on STQ (β = 0.981) and SVQ (β = 1.299). This result seems to indicate that increases in the overarching technical dimension corroborate gains in system and service quality. In line with Figure 3, the absolute effect on SVQ is larger than on STQ.

Table 6. Regression Weights for (TF): (Group number 1—Default model).

			Estimate	S.E.	C.R.	p
IFQ	←	TF	1.000
STQ	←	TF	0.981	0.090	10.869	***
SVQ	←	TF	1.299	0.105	12.381	***
IFQ1	←	IFQ	1.000
IFQ2	←	IFQ	0.955	0.057	16.769	***
IFQ3	←	IFQ	0.876	0.060	14.477	***
IFQ4	←	IFQ	1.006	0.056	17.999	***
STQ1	←	STQ	1.000
STQ2	←	STQ	0.950	0.059	16.111	***
STQ3	←	STQ	1.274	0.076	16.754	***
STQ4	←	STQ	1.234	0.074	16.708	***
SVQ1	←	SVQ	1.000
SVQ2	←	SVQ	1.000	0.040	24.808	***
SVQ3	←	SVQ	1.029	0.044	23.529	***
SVQ4	←	SVQ	1.034	0.042	24.783	***

* significant at 10% level, ** significant at 5% level, *** significant at 1% level.

Figure 3. Technical Factors as Second-Order Factors.

Our findings show that the technical dimension (information/system/service quality) strongly drives FinTech adoption, and this is coherent when analyzed against the background of post-pandemic evidence, which shows that system/information quality remains a primary determinant of FinTech usage among young users [36]. Furthermore, they also align with recent syntheses that position IS Success qualities (information, system, service) as core antecedents of satisfaction/intent in FinTech contexts [27]. At the same time, some studies suggest different dominance patterns—for example, the fact that social influence/brand cues can outweigh technical quality in influencer-rich youth settings. Similarly, research in emerging markets sometimes highlights usefulness/ease-of-use routes (proximal to system/information quality) alongside financial capability factors, and this has the effect of producing mixed magnitudes across samples [11].

Considering the relationship between the three first-order factors and their indicators, the analysis results reveal the following: First, for the Information Quality, the indicators IFQ2, IFQ3, and IFQ4 are positively related to IFQ. IFQ, IFQ2, IFQ3, and IFQ4 increased by approximately 0.955, 0.876, and 1.006 units, respectively. IFQ4 has the strongest impact on Information Quality. Second, regarding System Quality, the indicators STQ2, STQ3, and STQ4 exhibit significant and positive relationships with System Quality (STQ). For each unit increase in STQ, the indicators STQ2, STQ3, and STQ4 increase by approximately 0.950, 1.274, and 1.234 units, respectively. STQ3 has the strongest impact on System Quality, with the highest estimate. Third, for Service Quality, the indicators SVQ2, SVQ3, and SVQ4 have significant, positive relationships with the Service Quality (SVQ) construct. For each unit increase in SVQ, the indicators SVQ2, SVQ3, and SVQ4 increase by approximately 1.000, 1.029, and 1.034 units, respectively. SVQ2 has the strongest impact on Service Quality.

Finally, when the model was evaluated for technical factors, analysis revealed that the majority of fit indices suggested an acceptable model fit, and some indices indicated an excellent fit. According to ref. [46] criteria, a combination of CFI > 0.95 and SRMR < 0.08 indicates a good model fit, which is met by the model, as shown in Table 7. The RMSEA is slightly above the threshold for an excellent fit but is nonetheless within the acceptable range.

Table 7. Model fit of the Technical Factors Model.

CMIN/DF	4.587	Acceptable
CFI	0.956	Excellent
SRMR	0.051	Excellent
RMSEA	0.061	Acceptable
PClose	0.046	Acceptable

5.2.3. Other Factor (OT)/(SOFA)

OT represents the third second-order factor in this study. Table 8 presents the Confirmatory Factor Analysis on the second-order factor, which presents the unstandardized estimates of the analysis. The antecedent variable was defined as Other Factor (OT). OT, in turn, affected the first-order variables: SI, FC, and BS. When exploring the relationship between Second-Order Factor and First-Order Factors, the following points emerge: First, the Other Factor (OT) is related to SI, whereby the known value of the estimate is set at 1.000 for identification of the model. Second, FC is positively correlated with OT, with an estimated probability of 1.207. Thus, FC is positively influenced by OT, whereby a 1-unit increase in OT results in a 1.207-unit rise in FC. Third, BS and OT are positively correlated, with an estimated correlation coefficient of 1.115. Thus, one unit of OT corresponds to one unit of BS is 1.115 units.

Table 8. Regression Weights for (OT): (Group number 1—Default model).

			Estimate	S.E.	C.R.	p
SI	←	OT	1.000
FC	←	OT	1.207	0.117	10.334	***
BS	←	OT	1.115	0.099	11.222	***
SI1	←	SI	1.000
SI2	←	SI	0.956	0.053	17.881	***
SI3	←	SI	0.875	0.052	16.951	***
SI4	←	SI	0.821	0.051	16.075	***
FC1	←	FC	1.000
FC2	←	FC	0.931	0.066	14.063	***
FC3	←	FC	0.953	0.067	14.116	***
FC4	←	FC	0.982	0.050	19.564	***
BS1	←	BS	1.000
BS2	←	BS	0.956	0.046	20.588	***
BS3	←	BS	0.910	0.043	21.004	***

* significant at 10% level, ** significant at 5% level, *** significant at 1% level.

Recent research on the social-external dimension of FinTech adoption is compatible with results showing that social influence, facilitating conditions and brand trust load strongly on the higher-order Other Factor (OT). In that context, studies ref. [11] confirms that social cues, trust, and supportive conditions significantly enhance adoption among young users. On the other hand, ref. [27] identify these determinants as core across FinTech contexts, while ref. [12] suggest, conversely, that in digitally mature environments, technological and financial factors may outweigh social influence. This seems to suggest that the strength of social determinants is based on user experience and digital ecosystem maturity.

In summary, the results indicate that the Other Factor (OT) has a significant influence on both Facilitation Conditions (FC) and Brand and Service Trust (BS); specifically, Facilitation Conditions are more strongly affected by the Other Factor than Brand and Service Trust (see Figure 4).

Figure 4. Other Factors as Second-Order Factors.

Table 9 highlights the overall model fit. The majority of the model’s fit indices were acceptable; some indices were in the extremely excellent range. The RMSEA is just beyond the twice value, but it is still satisfactory to some extent. Ref. [46] suggest that satisfactory model fit can be defined by the combination of CFI value > 0.95 and SRMR values < 0.08. The table indicates that the proposed model fits this criterion.

Table 9. Model fit of the Other Factors Model.

CMIN/DF	3.764	Acceptable
CFI	0.962	Excellent
SRMR	0.041	Excellent
RMSEA	0.071	Acceptable

5.3. Structural Equation Model (Path Analysis)

The third stage of analytical procedures, path analysis, identifies both direct and indirect pathways between elements of a system. SEM consists of path analysis as its essential component. Table 10 displays the non-standardized results of the four steps of our path analysis. The regression demonstrates a positive effect of 0.425. This suggests that Financial Factors (FF) substantially influence attitudes (ATT). When Financial Factors increase at a higher level, it becomes more likely that individuals develop positive attitudes toward FinTech usage. Technical Factors demonstrated a significant positive relationship with Attitude, with an estimated value of 0.617. Thus, technical factor improvement through user-friendly and reliable FinTech services appears to generate better user attitudes.

Table 10. Path Analyses.

Path	Estimate	S.E.	C.R.	p	Null Hypothesis	Decision	Conclusion
ATT ← FF	0.425	0.081	5.238	***	No impact of FF on ATT	Reject	Financial Factors positively impact Attitude towards FinTech. Improving FF enhances ATT.
ATT ← TF	0.617	0.110	5.620	***	No impact of TF on ATT	Reject	Technical Factors positively impact Attitude towards FinTech. Improving TF enhances ATT.
AI ← ATT	0.639	0.086	7.438	***	No impact of ATT on AI	Reject	Attitude significantly impacts Adoption Intention of FinTech. Improving ATT increases AI.
AI ← OT	0.503	0.100	5.034	***	No impact of OT on AI	Reject	Other Factors positively impact Adoption Intention of FinTech. Identifying OT factors enhances AI.

* significant at 10% level, ** significant at 5% level, *** significant at 1% level.

The analysis also indicates that Attitude (ATT) has a positive impact on the Adoption Intention of FinTech (AI), specifically with a significant estimate of 0.639. A positive attitude toward FinTech services will drive higher customer intentions to adopt these services. The fourth path analysis reveals a positive relationship between Other Factors (OT) and FinTech Adoption Intention (AI), with an estimated coefficient of 0.503. Additional external elements, unrelated to financial resources, technical proficiencies, and emotional perspectives, also affect people’s choices to adopt FinTech systems. The results of the path analyses are displayed in Figure 5 and Table 10.

Figure 5. Structural Equation Model.

Table 11 highlights the results of the model fit of the SEM model. While the CFI is slightly below Hu and Bentler’s criterion of 0.95, the SRMR and RMSEA are in an excellent range. Overall, the model fit is acceptable given that most indicators suggest a good to excellent fit. Thus, the model demonstrates an acceptable fit with the data. That is, the hypothesized model is both well-specified and an effective representation of the underlying data structure.

Table 11. Model fit of the SEM model.

CMIN/DF	2.421	Excellent
CFI	0.922	Acceptable
SRMR	0.052	Excellent
RMSEA	0.051	Excellent
PClose	0.263	Excellent

6. Discussion

This paper focuses on determining the factors that affect the intention to adopt FinTech among business students. To achieve this goal, we used Second-Order Construct analysis, as prescribed by the DeLone and McLean IS Success Model [34]. The results contribute to the comprehension of six established and three emerging financial, technical, and external potential antecedents that may influence FinTech adoption. Consequently, both financial literacy and financial education are real factors that may enhance financial service users’ confidence in FinTech [17,29]. Students who better understood financial matters also believed that FinTech is more helpful and riskier than their peers. And this appeared more willing to embrace it. However, financial risk may hinder adoption intention because of fear of being fraud, low transaction security, and poor compliance with regulatory frameworks [16]. Based on these observations, a clear need exists for enhancing financial literacy and users’ abilities to manage risk when using digital financial services.

We also found that technical variables have the most powerful effects on attitudes toward adopting FinTech. Each of our three primary hypotheses was supported. Specifically, for all three main factors—system quality, information quality, and service quality—user trust increased, as did ease of use. These findings compare favorably to those of refs. [11,27]. Technical factors were also significant predictors of system quality. Recent studies (e.g., [32,33]) illustrated the importance of students’ perceptions of usability, reliability, and transaction security in FinTech platforms. This supports the call for FinTech providers to leverage the usability, security, and efficiency of the services they offer. Perceived social influences also had positive relationships with FinTech adoption. Specifically, the more students trusted a brand (i.e., the provider’s reputation), the more likely they were to engage with FinTech services. Others (e.g., [19,20]) have similarly argued that social influences, such as peer pressure, family support/coercion, have important effects on users’ adoption habits. Thus, our results suggest that, to improve customers’ beliefs in brand security. FinTech providers should (a) focus on brand credibility, (b) enhance customers’ engagement, and (c) ensure that their policies are transparent and comprehensible to potential users.

The findings also demonstrate that attitudes toward FinTech adoption moderate the association between the financial, technical, and external factors with adoption intention. This finding is consistent with the theories of refs. [9,29] concerning the perceived usefulness, perceived ease, and social influence of FinTech based on adoption intention. Further, external variables bore direct relationships with Fintech adoption intention. Thus, social influences and brand trust have direct relationships to adoption behavior; that is, these effects do not necessarily pass through attitudes to affect adoption intentions. Ref. [21] argued that users may adopt FinTech because of the driving force or because of the perceived credibility of FinTech’s producers, even if their attitudes toward FinTech are neutral.

The results extend prior research on FinTech adoption by adding the value of Second-Order Construct analyses to classify the key factors. Whereas previous research on the FinTech adoption examined financial, technical, and external factors separately, we established a hierarchy of constructs that encompassed three categories of factors [12]. The assertion is further supported by the relationship between technical factors and the financial and external dimensions. The influence of social pressure and switching to brand trust extends the work of ref. [20] and implies that external variables may enhance the adoption of FinTech. However, the negative role of financial risk in FinTech adoption contrasts with ref. [16]. This relationship will therefore require additional research before adequate policies can be recommended for KSA’s limited but growing digital financial environment. However, differences between studies of FinTech adoption need to be resolved before the findings can be incorporated and synthesized into broader conceptualizations. Second-Order Constructs offers a statistical method that attempts to generate more accurate models of adoption behaviors.

This study shows that beyond issues of technology acceptance and behavioral intention, FinTech adoption among university students carries profound implications for sustainable development. From an economic perspective, FinTech enhances financial inclusion and contributes to resilient and inclusive economic growth by lowering transaction costs and enabling more efficient financial practices [48]. From a social standpoint, the adoption of FinTech reduces inequalities by providing accessible digital financial services for youth and underserved groups, promoting equity and trust within financial ecosystems [49]. From an educational viewpoint, students’ engagement with FinTech reflects the role of higher education in shaping financial literacy, digital competencies, and innovation-oriented mindsets, ultimately fostering education for sustainable development [50,51]. The findings are compatible with the Saudi Vision 2030 goals of digital transformation and financial inclusion. In order to build user trust, our study suggests that regulators (SAMA, CMA) to strengthen digital governance and consumer protection. Universities could contribute to this effort by including financial literacy and FinTech awareness in their curricula. Additionally, FinTech startups are encouraged to design secure, user-centered platforms that promote inclusion. Limitations of this study include the single-institution, cross-sectional design, and future research should explore longitudinal and cross-country comparisons.

7. Conclusions

Using the IS Success Model of DeLone and McLean, this study investigated the determinants of intentions to adopt FinTech among business students, Taif University, Saudi Arabia. We showed that Fintech adoption intentions are motivated by financial, technical, and external factors. This research also generated knowledge of young Saudi adults’ digital attitudes towards FinTech and their perceptions of issues concerning and prospects for adopting FinTech. To establish this connection, the study used financial literacy, perceived financial education, and the level of financial risk that students are willing to embrace in making decisions to adopt FinTech. These findings support the well-established knowledge of financial issues, which increases confidence and trust in embracing digital services for managing financial assets. Furthermore, regarding the website’s qualities, perceived information quality, perceived system quality, and perceived service quality may have a considerable impact on users’ trust and perceived ease of use. Additionally, the results revealed that social factors, brand and service trust, and facilitating conditions all influenced adoption decisions. The results support the D&M IS Success model by illustrating the underlying effects of social influences and facilitating conditions on adoption behaviors. However, more research is needed to further increase our understanding of FinTech services. Moving forward, the Saudi Arabian Vision 2030 initiative among students will be explored.

The study has both practical and theoretical applied implications. Theoretically, this work extends the DeLone and McLean IS Success Model to a novel demographic and cultural context, thereby supporting the model’s generalizability. The findings also provide reasons for further studying financial and technical predictors to help explain adoption intentions. For FinTech enhancement, these insights can be used to understand user preferences, ensuring that FinTech providers develop services that are secure, user-friendly, and reliable for young people. Universities and colleges should incorporate the concepts of finance and financial technologies into their teaching curricula for their learners, enabling them to navigate the future financial world. To increase people’s confidence and reduce risks associated with finances, policymakers should strengthen digital regulations and security. This study is of practical pertinence to university students because it links key determinants of FinTech adoption with specific actions. The assumption is that enhancing financial literacy and education can empower students to make informed digital financial decisions on the one hand. On the other hand, improving technical competence helps them evaluate system quality and security. After all, it is possible to promote responsible adoption by building trust and social awareness through peer collaboration and FinTech workshops. Further research should look into cross-disciplinary perspectives to expand our understanding of FinTech dissemination among university students. One reason for that is that the data for this study were collected from business students, limiting generalization. Students from other disciplines—though less familiar with financial concepts—often display strong engagement with digital tools, and this may influence FinTech adoption differently.

However, the current study has several limitations: First, the data were collected from university business students and therefore may not generalize to other educational levels and demographics. Research should be conducted with large populations of students from various faculties and geographical locations. Additionally, students from different fields may not be as familiar with financial concepts, but they may be very engaged with technology and able to adapt to new technologies, which could also affect how they adopt FinTech. Second, rapid technological change may render the findings obsolete if they fail to incorporate periodic adjustments and capture emerging trends and innovations in FinTech services. Further research should be conducted to understand whether financial literacy interventions have a long-term impact on FinTech adoption. Furthermore, examining current government policies and incentives that promote the expansion of digital finance may offer additional policy insights. Thus, comparing the outcomes of the FinTech system in Saudi Arabia with those of other GCC countries can reveal some peculiarities of the regional context and the practices that would accelerate the formation of an efficient FinTech system for Saudi Arabia.