Factors Motivating Behavioural Intention Towards Digital Banking Among Women in Rural Kerala: Evidence from A Structural Equation Modelling Framework

Digital banking, also known as online banking or internet banking process of managing bank accounts and performing money transactions.  Through these platforms, Clients have simple access to traditional banking services such as fund transfers, account management, fund deposits, debt management, investment activities, and other related services. These rapid technological advancements encourage the use of technology to boost productivity and business competitiveness, and the creation of a variety of new tools and applications to meet customer needs (Mehdiabadi et al.,2020). So all banking institutions are forced to develop innovative utility services called digital banking (Nguyen, 2020). Digital banking is a banking method brought about by the quick development of mobile technology and the growing number of smartphone users (Suhaimi & Hassan, 2018). According to the RBI Payment Systems Report, 2025, India’s financial payment mechanism has shown a drastic growth in recent years, with internet-based payments accounting for 99.8% of all payment transactions in the first 6 months of 2025. 56.8 % of people in Kerala perform online banking transactions using a computer or mobile device. According to “The Ministry of Statistics and Programme Implementation’s Comprehensive Modular Survey – Telecom, 2025”, 30.0 percent of rural Indian women can perform online banking transactions. Compared to 17.1% in the previous year, this is a notable growth. In Kerala, 49.2% rural women in the 15 years and above age category can successfully perform digital financial transactions effectively www.esankhyiki.mospi.gov.in.  In this view, the present study concentrates on identifying the main factors that motivate rural women towards digital banking.  Identifying the motivating factors is helpful to authorities, policymakers, and financial services providers, who can promote financial inclusion in our country through new technology initiatives. This study enhances digital banking platforms and provides advice on how to successfully promote new digital systems to potential clients by determining the elements that promote and prevent the use of digital banking.  Additionally, this study significantly contributes to women's empowerment, which is essential to our nation’s equitable and sustainable economic development. The study focuses on the following research questions:

RQ1: Is PEOU a motivating factor to use digital banking for their financial transactions?

RQ2: Does the intention to utilise digital banking depend on PU?

RQ3: Does TS motivate consumers’ willingness to access online banking facilities?

RQ4: Does the consumer’s intention to use digital banking services change due to the effect of CA?

RQ5: Does FSE shape consumers' decisions to utilise digital banking services?

The following way this research study is organised. The research article is introduced in Part A, where the literature review and hypothesis development are presented. Part C includes the study's methodology, and Part D contains the Data Analysis and outcomes of the study. The conclusion and recommendations are presented in Part E.

B. Literature Review, Theoretical Background, and Hypothesis Formulation

Several studies have been conducted globally to identify the critical elements that contribute to the broad adoption of digital banking. Research in the advanced nations has been concentrated on the variables that motivate individuals to use internet-based banking channels (Ananda et al., 2020; Fonchamnyo, 2012; Muñoz-Leiva et al., 2017; Vuković, 2025). Based on the “Theory of Reasoned Action” (TRA), Ajzen & Fishbein, 1980, TAM (Davis, 1989) is believed to be the most often used model among information systems researchers. An ETAM was utilized in most of the research to pinpoint the major variables impacting the adoption of various kinds of digital banking applications. Here, we studied the rural women's acceptance of digital banking using ETAM. In addition to the traditional TAM categories of PEOU and PU, the model incorporates TS, FSE, and CA as external variables impacting behavioural intention.

5. Perceived Ease of Use (PEOU)

 PEOU was suggested by Davis (1989) as "the degree to which a person believes that using a particular system would be free of effort”. Potential consumers' views about difficulties of learning and utilizing technologically based apps are collected by PEOU (Gounaris & Koritos, 2008). It refers to the way people understand and utilize electronic banking, or how much they thought it would be effortless (Davis et al., 1989). Banks can better connect with their customers by making their products easy to use (Jeong & Yoon, 2013). According to previous studies, results have a strong influence on usage intention (Davis, 1989; Ali et al., 2021; Gounaris & Koritos, 2008; Utomo et al., 2025. In the light of the aforementioned findings we propose the following hypothesis.

H1: “PEOU has a strong positive influence on Intention to use Digital Banking.”

5. Perceived usefulness (PU)

According to Gounaris & Koritos,2008, PU captures responses from potential clients regarding the advantages of using a technology-based application. According to Davis et al. (1989), the most often used independent variable in earlier studies to measure the readiness of individuals to adopt digital banking is perceived utility. Customers are more likely to stay with digital payment methods if they find them effective, even though they had negative experiences (Ali et al., 2021; Linh & Huyen, 2025: Al-Fahim, 2012; Ryu, H.-S, 2018). Hence, we developed the hypothesis as follows.

H2: “PU has a great influence on Intention to use Digital Banking.” 

5. Convenience and accessibility (CA)

Convenience is defined as “the time and effort saved in performing a task.” (Cheney, 2008).  An individual's capacity to access information and services on the internet is known as accessibility, and it depends on several factors, including hardware, software, infrastructure, and weather conditions, etc. (Hackett and Parmanto, 2009). E-banking allows users to access financial services 24/7. Digital banking allows clients to make online payments, transfer money between account holders, and monitor their bank account statements. Therefore, based on the above literature support, we developed a hypothesis as follows.

H3: “CA has a beneficial impact on Intention to use Digital Banking.” 

5. Trust and security (TS)

According to Kolsaker and Payne (2002), security means opinion about the various safe payment options and the mechanisms for information storage and transmission systems. Casaló et al. (2007) state that customer trust and commitment in financial services organizations' online transactions are directly and significantly impacted by website security, privacy, and reputation. The behavioural intention to accept mobile applications will be positively impacted by the security of the system (Sharma et al. 2018; Hamlet and Strube 2000; Sathye 1999; Damghanian et al., 2016). As a result, the following hypothesis has been developed based on the research mentioned earlier:

H4 - TS has a significant influence on Digital Banking financial services.

5. Financial self-efficacy (FSE)

The belief in one’s own skills, knowledge, and ability to perform a task is called self-efficacy. (Luarn & Lin, 2005). Furthermore, self-efficacy is the confidence that one can use a computer to obtain the information and knowledge one needs. According to Compeau and Higgins 1995, Self-efficacy is the belief in one's own abilities, which generally leads to the adoption of new technology. In this research, FSE refers to a person's level of confidence in her/his ability to make smart financial decisions (Netemeyer et al., 2017). The objective of our study is to determine whether respondents' opinions about their own knowledge, ability, and talent affect digital banking. Therefore, from the previous studies, we can conclude that one’s ability and confidence to utilise digital technologies influence online banking transactions Noor et al., 2020; Hedau, A., 2025; Aisjah, 2024). Thus, this study framed the following hypothesis:

H5: “FSE has a great impact on Intention to use Digital Banking.”

C. Research methodology

The influence of digital banking on rural women was investigated in this study using a descriptive research Methodology. Descriptive research can be useful for studies that investigate an event, situation, or population (Siedlecki, 2020). This study makes an effort to understand how digital banking platforms promote financial inclusion foe range of demographic groups by applying this methodology (Manlapaz & Quendangan, 2024). The primary data was collected from a sample size of 240 women respondents using a structured questionnaire. The data was collected between November 2025 and December 2026. The sample was chosen from the population via judgment sampling. It is a non-probability sampling technique, where samples are chosen based on certain attributes that help to achieve the study’s goals. Etikan, Musa, & Alkassim, 2016). The questionnaire was designed based on an exhaustive literature review to ensure that the questions are properly included according to the theme of the study. The demographic details of the respondents is included in the first part of the questionnaire. The second section of the survey instrument includes the indicators of each construct, as shown in Table 1. A 5-point Likert scale, with 1 denoting "strongly disagree" and 5 denoting "strongly agree," was used to measure each indicator in this study.

Table 1.  Measurement scales.

Figure1: Conceptual Framework of Digital Banking Adoption

RESULTS AND DISCUSSION

Descriptive findings

According to the researchers’ results, demographic characteristics play an important role in using digital financial services. Table 2 includes the demographic classification of rural women respondents under various categories such as age, marital status, education, occupation, and monthly income. This table demonstrates that out of 240 respondents, the largest number of digital financial service users, i.e., 48.8%, lies in the age group of 20-30 years, and the least number of users, i.e., 4.1%, lies in the age group of above 61 years. It indicates that rural women in the 20-30-year age group are more interested in using digital banking services effectively. About marital status most of the respondents are married (66.7 %), and 32.5% lies in the category of single. Most of the respondents (54.2%) are post-graduates, followed by undergraduates (25.8%). The maximum percent of respondents, i.e., 72.9%, were salaried women.  53% of rural women earn below 25000 rupees per month, and 31.3% get a monthly income between 25001 and 50000 rupees.

Table 2: Demographic data of Respondents.

Source: Author's calculation (2026)

Results of PLS-SEM analysis

This study employed SEM as the key analytical technique to assess the proposed research model. SEM is particularly beneficial in fields like marketing, the social sciences, and the adoption of digital technology, where research models incorporate mediating variables and latent constructs (Iacobucci, 2009). To perform the SEM analysis, Smart PLS 4.0 software was used.

Measurement model assessment

Table 3 shows the item reliability, which ranges from 0.710 to 0.931, which has an acceptable level of reliability. Following the item reliability assessment, our study used ‘Cronbach's alpha’, ‘Composite reliability’(CR), and ‘Average Variance extracted’ (AVE) to check the construct validity. According to Nunnally and Bernstein (1994), the range of the ‘Cronbach’s a lpha coefficients’ is 0.784 to 0.877, and all the measurements in this study are greater than 0.7.  Additionally, CR ratings from 0.876 to 0.916, which is above 0.7 and acceptable (Hair et al., 2009). Furthermore, the AVE coefficient, which ranges from 0.672 to 0.758, is likewise greater than 0.5 in terms of convergent validity. (Hair Jr et al., 2020). Consequently, these findings show strong overall reliability.

Table 3: Reliability Results

Source: Author's calculation (2026)

Table:4 Fornell- Larcker Criterion (Discriminant Validity results)

Source: Author's calculation (2026)

Table 4 shows the Discriminant Validity Analysis using Fornell-Larcker (1981) criterion.  To have appropriate discriminant validity, the square root of AVE needs to be greater than all (Chin,1998).

Table 5: HTMT (Heterotrait-Monotrait Ratios)

Source: Author's calculation (2026)

According to the HTMT criterion, HTMT should not exceed the threshold of 0.90 for conceptually distinct indicators (Hair et al. 2019).  All HTMT ratios (Table 5) are less than the threshold.

Structural Model Results

 The gathered data was analysed using “Partial Least Squares Structural Equation Modelling” (PLS-SEM) using Smart PLS 4 software.

Table 6: Direct effect Results

Source: Author's calculation (2026)

Hypothesis test.

The path coefficients, S.D., T values, and P values were used to explain the structural model. The PEOU level of respondents has a positive and statistically significant effect on intention to use digital banking (β = 0.093, t = 2.293, p = 0.022). PU (β = 0.162, t = 2.995, p = 0.003) has a significant influence on digital banking. CA (β = 0.123, t = 2.217, p = 0.027) has a great influence on the use of digital banking facilities. TS (β = 0.211, t = 3.092, p = 0.002) emerges as the second most influential factor. FSE (β = 0.447, t = 6.516, p = 0.000) is the strongest predictor of behavioural intention toward digital banking, which indicates that rural women’s confidence in their ability to use digital financial services plays a decisive role in adoption. Table 6 shows the specific results of the five hypotheses.

Figure 2: Measurement Model of Digital Banking Adoption

Quality Analysis Model

Model performance can be measured using R-squared and adjusted R-squared. The regression model's quality of fit is frequently assessed using the R² (Hair et al. 2019). R², a measure of the variance explained in each internal construct in PLS-SEM studies, indicates the explanatory strength of the model.  Normally R-square value of 0.75 is considered to be good, 0.50 is considered to be acceptable, and 0.25 is considered to be weak (Hair et al. 2019). Table 7 shows the R-Square value of 0.673, which means that the independent variables, namely PEOU, PU, CA, TS, and FSE, collectively have 67.3% influence over behavioural intention on digital banking. The adjusted R-squared value of 0.666 means that after taking into consideration several predictors, 66.6% of the variance in the dependent variable remains explained. The model is well-specified and free of unnecessary variables, as evidenced by the modest difference between R2 and adjusted R2 (0.007).

Table 7. Results of Hypothesis Testing

Source: Author's calculation (2026)

Table 8 shows the status of a hypothesis tested in the study

Table 8: Hypothesis Assessment Summary