Clients Seeking Redress and its Influence on Relationship Quality Dimensions: A South African Banking Fraud Perspective

ABSTRACT Clients Seeking Redress and its Influence on Relationship Quality Dimensions: A South African Banking Fraud Perspective Kholiswa Mathiyase* Department of Business Management, Faculty of Business and Economic Sciences, Nelson Mandela University, Gqeberha, South Africa [email protected] * Corresponding author Dr. Xolile Antoni Department of Management, Faculty of Commerce, Rhodes University, Makanda, South Africa [email protected] Prof. Chantal Rootman Department of Business Management, Faculty of Business and Economic Sciences, Nelson Mandela University, Gqeberha, South Africa [email protected] The Retail and Marketing Review Volume 21, Issue 2, November 2025, Pages 51-68 Doi: https://doi.org/10.5281/zenodo.17341018 Despite the rise in digital banking fraud incidents in South Africa and its influence on banking client relationships, there are limited studies conducted in South Africa that consider how seeking redress influences relationship quality between banks and clients who have encountered banking fraud. Seeking redress allows clients to voice their complaints and improves the quality of their relationship with the bank. Clients who seek redress when they have encountered banking fraud provide banks with the opportunity to correct the mistake and prevent re-occurrences of fraudulent transactions in clients' bank accounts. This study investigates the influence of seeking redress on relationship quality dimensions, based on the perceptions of banking clients in South Africa. A quantitative research methodology was adopted, and an online questionnaire was used to sample 399 South African bank clients who had previously been affected by fraud. The study used statistical techniques such as confirmatory factor analysis (CFA) and structural equation modelling (SEM) to test the relationships between seeking redress and relationship quality dimensions. The results show that seeking redress has a significant impact on relationship quality dimensions, such as satisfaction and loyalty. Banks should thus encourage banking clients who have encountered banking fraud to seek redress to improve the satisfaction and loyalty of banking clients. The implications are that banks need a functioning complaint management system to receive fraud complaints and provide satisfactory responses to their clients. The results also showed that seeking redress and satisfaction does not directly influence loyalty. Keywords: banking fraud; clients, complaint behaviour; relationship quality, seeking redress. Clients Seeking Redress … A South African Banking Fraud Perspective 52 1. INTRODUCTION AND BACKGROUND Banking fraud continues to rise in South Africa despite all the banks’ efforts to prevent such incidents. Banking fraud escalated by 45% in 2023, with incidents related to banking applications accounting for 60% of all banking fraud reported incidents (SABRIC 2023). The rise in fraud incidents has also increased the number of complaints against various banks in South Africa. In 2023 alone, the Ombudsman for Banking Services received about 21.641 banking fraud complaint cases, an 11% increase from the previous year, 2022 (Ombudsman for Banking Services 2023). The impact of banking fraud extends beyond financial losses, affecting both clients and banks in various ways (Phiri et al. 2024:2). Therefore, banking clients are expected to complain about banking fraud so that banks can be informed about the incident. Client complaints are a valuable source of important market information that firms can use to correct the problem's root cause and improve their service delivery or product (Oluwasanmi et al. 2023:20). Istanbulluoglu et al. (2017) explain client complaint behaviour (CCB) as a complex and multidimensional concept. Iram and Iqbal (2023:1276) define CCB as an action undertaken by clients after they have experienced a negative element of a product or service delivery. Oluwasanmi et al. (2023:21) also explain CCB as a process that constitutes a subset of clients’ possible and distinct responses to perceived dissatisfaction around purchasing products or during service encounters and consuming products or services. Clients who encounter banking fraud may exhibit different types of complaint behaviour (Sofia et al. 2023:4). Several authors have outlined different types of complaint behaviour, including voicing a complaint (which refers to a client expressing their dissatisfaction with a firm); taking no action and remaining loyal; exiting or ending the relationship with the bank; spreading negative word of mouth; third-party complaints and seeking redress (Iram & Iqbal 2023; Istanbulluoglu et al. 2017; Oluwasanmi et al. 2023; Sofia et al. 2023). This study focuses on clients seeking redress. Ong et al. (2016:1) defines redress as the process of dealing with complaints emanating after a client has used a firm’s service or after purchasing products. Redress is a process where a firm resolves or atones for the problems that may have occurred after clients have used a firm’s service or after the purchase of products (Mayombo 2014:4). According to Frasquet et al. (2021:1643), redress also provides clients with an opportunity to express their dissatisfaction and address their complaints. Therefore, seeking redress concerns procedures through which individuals seek remedies or responses to their complaints (Iram & Iqbal 2023:1310). Client complaints are a valuable source of important market information which firms can use to correct the root cause of the problem and to improve their service delivery (Oluwasanmi et al. 2023:21). Although complaints are an inseparable part of the banking industry complaining clients represent clients who are willing to provide a bank with the opportunity to improve and maintain business transactions (Parikh & Dutt 2022). According to Pio et al. (2024:1050), banks may not be able to improve without valuable feedback from complaints and may lose their competitiveness in the marketplace. This means that complaints in the banking industry must be recognised and resolved to improve services and processes for growth and sustainability. Therefore, complaints are an important feedback mechanism for firms to monitor client satisfaction with a firm’s products and services (Oluwasanmi et al. 2023:21). Encouraging clients to seek redress when they encounter fraud may therefore improve the relationship quality between banks and clients. Relationship quality dimensions include satisfaction, trust and loyalty. Banks committed to delivering excellent client service should allow clients to complain (Pio et al. 2024:1050; van Deventer & Redda 2023:211). Frasquet et al. (2021:1643) also assert that “firms encouraging clients to seek redress through specially designated channels can increase client satisfaction and loyalty”, and thus possibly improve the relationship between banks and their clients. 53 The Retail and Marketing Review: Vol21 Issue 2 (2025) ISSN:2708-3209 2. RESEARCH PROBLEM AND PURPOSE Banking fraud may negatively impact clients’ earlier perceptions of feeling secure and protected by their bank. Clients who no longer feel secure and protected by their bank may ultimately switch to another (Puente-Cavazos et al. 2025:2). Moreover, gaining new clients is more costly than maintaining an existing clientele (Rudd et al. 2022:70). When firms do not know the sources of dissatisfaction among clients, these firms may lose clients and suffer financially (Phiri et al. 2024:2). Therefore, complaints, particularly seeking redress are an important feedback mechanism for firms to monitor client satisfaction with a firm’s products and service effectiveness (Frasquet et al. 2021:1643). Complaining clients should not be seen as a nuisance, but rather as those that are invested in providing banks with an opportunity to resolve failures and strengthen relationships (Tax et al.1998).Given this background, the purpose of this study is to investigate the influence of seeking redress on relationship quality dimensions (trust, satisfaction and loyalty) among banking fraud clients. While studies on complaint behaviour exist in relation to service failure, service recovery and dissatisfaction, very few studies have examined complaints, specifically after fraud incidents where both financial and emotional harm are heightened (Kassem, 2024; Alashwali et al. 2024; Phiri et al. 2024:2). There are also limited studies that have investigated the direct influence of seeking redress on relationship quality dimensions such as trust, satisfaction and loyalty. This study addresses that gap by examining whether clients seeking redress following a banking fraud incident strengthens or weakens the bank-client relationship in terms of satisfaction, trust and loyalty. This is particularly important in South Africa, where banking fraud is on the rise, with the Ombudsman for Banking Services reporting approximately 20,000 fraud-related complaints in 2023 alone. Therefore, this study focuses on banking fraud and complaints in the South African banking industry. The complaint behaviour processes, and bankclient relationships are relevant to this. On this basis, it is also concluded that the study is based on marketing literature, specifically relationship marketing and banking fraud. 3. A LITERATURE OVERVIEW OF COMPLAINT BEHAVIOUR, SEEKING REDRESS AND RELATIONSHIP QUALITY DIMENSIONS A few seminal client complaint models were considered when developing the hypotheses for the study. The Day and Landon’s (1977) model of client complaint behaviour represents the differences between those who act (take action) and those who do not act (non-action) when they have a complaint. The model further differentiates between taking public or private action, in other words, the different ways in which complaint action can be taken. In addition, the model specifies actions, such as seeking redress directly, taking legal action, complaining to a third-party or friends and boycotting a firm. Similarly, the other model that was considered was the Singh’s (1988) model of client complaint behaviour, where clients' responses are classified into three categories: voice responses, private responses and thirdparty responses. Voice responses are direct complaint behaviour to the service firm, including seeking redress. Second is private responses, which relate to indirect complaint behaviour, including word-of-mouth communication. Third is third-party responses, which means taking legal action or reporting a complaint to a regulatory body. The two adopted models have some limitations due to their static and single post-purchase focus (Mayombo 2014:4). For instance, the Day and Landon (1977) and Singh (1988) models do not show the complaint behaviour sequence clearly. These models only point out clients' options when clients encounter dissatisfaction and need to complain. This study adopted the two models by considering clients who take action and voice their complaints, as outlined by Singh (1988), not only by voicing concerns but also by seeking redress with a bank. Seeking redress may provide an opportunity for banks to correct mistakes and preserve the quality of the bank-client relationship, including aspects such as trust, satisfaction Clients Seeking Redress … A South African Banking Fraud Perspective 54 and loyalty towards the bank. This means that after clients’ experience banking fraud, they may seek redress directly from the bank first. 3.1 SEEKING REDRESS Seeking redress occurs when clients express their dissatisfaction with an expectation that their complaints will be resolved or that a firm will fix the problem (Ong & Teh 2016:1). Frasquet et al. (2021:1643) also assert that clients who seek redress are clients who complain to a firm and require an explanation. They expect to be compensated for the loss, request a refund or an exchange, or for the product to be repaired. According to Iram and Iqbal (2023:1315), seeking redress is more likely to be used early and depends on clients' ability and willingness to exercise it. Clients may seek redress directly from a firm involved in their dissatisfaction experience or from a third party. When a client seeks redress directly from a firm, it enables the firm to acknowledge the complaint and address the matter. Moreover, a firm will have an opportunity to prevent the problems from recurring and preserve the relationship with its clients (Carlson et al. 2023:1). Various factors influence clients’ decisions to seek or not seek redress. Examples include the perceived likelihood of success and clients’ attitudes towards complaining might play a significant role (Blodgett & Anderson 2000:322; Turaga et al. 2020:1076). George (2021) notes that dissatisfied clients choose to seek redress based on the perceived likelihood of success, their attitudes towards complaining and perceived justice. Similarly, George (2021) asserts that clients often avoid seeking redress when they believe their efforts will not yield a successful outcome. 3.2 RELATIONSHIP QUALITY DIMENSIONS Dai (2025:155) explains relationship quality as an overall state of the relationship, viewed as a concept with several dimensions. Ballantyne et al. (2003:14) argue that in an environment where firms offer similar products and services, the quality of an ongoing relationship becomes a way of gaining a competitive advantage. According to Roy et al. (2023), relationship quality refers to clients’ perceptions of how well the relationship fulfils their expectations, predictions, goals and desires. The most commonly used dimensions to examine relationship quality are satisfaction trust and loyalty (Zietsman 2017:2). Dai (2025:155) studied relationship quality by using the factors of satisfaction, trust and loyalty from the perspective of clients. In the current study, relationship quality is examined as clients’ overall evaluations of the long-term interactions of satisfaction, trust and loyalty. Satisfaction can be defined in various situations that are associated with products and services. According to Sabir et al. (2014:1015), satisfaction occurs when a client believes that a firm’s products and services meet their expectations. Singh et al. (2023:3458) also explain satisfaction as clients’ perceptions that a firm has fully, efficiently and promptly met their expectations. Satisfaction is also the result of a comparison between a client's purchase of a product or service and its expected performance versus its actual performance (Hertzberg et al. 2020). Satisfaction describes a generally positive experience of a purchase, which includes the service and product quality(Singh et al. 2023:3458). The satisfaction is based on a client’s judgement and perception of whether the bank has met or exceeded the client’s expectations. One way to achieve strong relationships is to ensure that clients are satisfied. This is because dissatisfied clients will defect, leading to the end of a relationship (Abdella & Indradewa 2024:370). Trust is also one of the most widely defined and examined relationship quality dimensions in marketing literature (Dai 2025; Roy 2023; Telang & Somanchi 2016). Trust largely determines the establishment and maintenance of relationships between clients and firms. Uruena and Hildalgo (2015:6) explain trust as a set of beliefs in the competence and integrity of the other party. Trust is also based on the level of clients’ confidence in a firm’s competence and 55 The Retail and Marketing Review: Vol21 Issue 2 (2025) ISSN:2708-3209 Seeking Redress Satisfaction Trust Loyalty H1 H2 H3 H4 H5 H6 performance. Trust has been generally accepted as a critical element in positive relationships (Akhlaq & Kiran 2022:582). The strength and quality of a relationship rely on the level of trust, and the higher the trust level, the stronger the relationship will be, as trust is vital in building quality relationships, especially in service firms or banks (Roy et al. 2023). Banks offer banking services and products that are complex and thus clients need to have confidence that the banking products and services will deliver as promised (Chen et al. 2012:7). Developing clients’ trust is especially essential where vulnerability and risk of fraud are eminent (Ndubisi & Wah, 2005; Roy et al. 2023). As an example, when clients experience monetary loss due to banking fraud, they might not trust their banks to keep their money and other financial assets safe and may blame them for the loss. Several studies have highlighted a direct link between client satisfaction and client loyalty (Kusumawati & Rahayu 2020; Suttikun & Meeprom 2021; Zephan 2018:2). Rosário and Joaquim (2023:52) define loyalty as "a deeply held commitment to a product or service of a firm despite situational influences and marketing efforts having the potential to cause switching behaviours”. Loyalty is also viewed as the continued connection clients have with their banks (Khadka & Maharjan 2017). According to Rosário and Joaquim (2023:52), loyalty is focused on the aspect of a relationship with a firm, and a loyal client is a person who makes regular purchases, uses a firm’s products and services and is treated as a partner. Loyal clients also have an impact on the creation of new products, tailored to their individual and specific needs. Loyalty is considered as the function of the share of total purchases, the function of buying frequency or buying pattern or the function of buying probability (Kuusik 2017:5). According to Rahman and Ramli (2016:607), loyalty has been recognised as a key determinant of clients’ retention and plays a significant role in longterm profitability. A loyal client will stay with the same firm, is likely to engage in positive word-of-mouth communication, recommend the product and influence the decision behaviour of friends and family members (Zhao et al. 2023). Loyalty in a bank indicates clients’ intention to revisit the bank and to re-use the bank’s products and services in the future despite a situation that may make them want to switch (Rosário & Joaquim 2023:52). Based on the literature above, Figure 1 illustrates a hypothesised model between the independent variable (Seeking redress), and the dependent variables (Satisfaction, Trust and Loyalty). FIGURE 1: HYPOTHESISED MODEL Source: Researcher’s construction Clients Seeking Redress … A South African Banking Fraud Perspective 56 The following hypotheses are proposed: H1: There is a significant relationship between seeking redress and satisfaction. H2: There is a significant relationship between seeking redress and trust. H3: There is a significant relationship between seeking redress and loyalty. H4: There is a significant relationship between satisfaction and trust. H5: There is a significant relationship between trust and loyalty. H6: There is a significant relationship between satisfaction and loyalty. 4. RESEARCH DESIGN AND METHODOLOGY A quantitative research methodology was adopted. A quantitative methodology was most suitable given the study's primary objective, which involved developing hypotheses and testing relationships between the selected variables. The population of the study was the clients of any bank operating in South Africa who had been previously affected by fraud in their bank accounts and had reported this to the bank. The data for the study was collected by a reputable research agency (CINT). CINT has a digital research database consisting of more than 155 million participants in over 130 countries, including South Africa. The profile of the participants on the database includes banking and non-banking clients. The questionnaire was captured on Microsoft Forms, and the URL link was sent to CINT to distribute on behalf of the researchers. CINT distributed the link to the survey through its online platform. The known population for the current study is 2.6 million South African respondents from the CINT database. According to Rahman (2023), when the population size is known, a sample size of 384 is sufficient for a population of 1 million. Therefore, a sample size of 384 (minimum) respondents at a confidence level of 95% was considered sufficient for the current study, to reach a maximum of 400 (Collis & Hussey 2014:198). A non-probability judgmental or purposive sampling was adopted because no sampling frame was available and due to strict bank-client confidentiality and the POPIA, the researcher had no access to the bank-client database. The sample size was 400 South African bank clients who had previously been affected by fraud and complained to their respective banks. Both secondary data and primary data methods were used for data collection. Secondary data were gathered by extensively reviewing various academic journals, articles, and textbooks. For this study, the secondary data review focused on topics, such as banking fraud, complaint behaviour, complaint management and relationship quality. The primary data was collected using a closed-ended online questionnaire focusing on South African bank clients who had previously been affected by fraud. Before primary data was collected, ethical issues were considered. For this study, ethics approval was received from the relevant university’s committee. The survey of the study included a consent form, and the respondents were also given an option to opt out of the survey. To ensure anonymity, the respondents were not required to provide their personal information, such as name, surname or ID number. The questionnaire had two sections, A and B. Section A used a categorical scale to collect information about demographics such as age, gender, population group and educational level. Section B used a five-point Likert-type scale containing statements relating to clients ‘seeking redress’ and ‘relationship quality dimensions’ (satisfaction, trust and loyalty). Existing items were used but a few new items were also developed by the researcher based on literature. Overall, 24 statements were posed in the questionnaire, 8 statements related to the ‘seeking redress’ variable (Frasquet et al. 2021; Mayombo 2014; Researcher’s construction), statements relating to relationship quality dimensions: 5 statements related to satisfaction (Khadka & Maharjan 2017; 57 The Retail and Marketing Review: Vol21 Issue 2 (2025) ISSN:2708-3209 Researcher’s construction; Rosário & Joaquim 2023), 6 statements related to ‘trust’ (Dai 2025; Ndubisi & Wah, 2005; Researcher’s construction; Roy et al. 2023) and 5 statements related to ‘loyalty’ (Akhlaq & Kiran 2022; Maxham & Metemeyer 2002; Puente-Cavazos et al. 2025; Researcher’s construction; Sousa & Voss 2009; Uruena & Hidalgo 2015). This study’s questionnaire items were developed by considering validity and reliability. The data was analysed using both descriptive and inferential statistical techniques. Various types of descriptive analyses, such as frequency and percentages were used to summarise the clients' demographic information. Descriptive analyses, including mean and standard deviation, were used to summarise the demographic variables and the factors of the study, namely seeking redress and relationship quality. Inferential statistics were used to measure validity and reliability as well as establish correlations between variables. Furthermore, the inferential statistical method used to analyse the data was the structural equation model (SEM). SEM was applied to examine the relationships among the variables in the study. SEM is a multivariate technique that tests and evaluates multivariate relationships between independent, mediating and dependent variables (Hair et al. 2014:565; Moqbel et al. 2020). SEM enables a test of complex patterns of relationships and the simultaneous testing of multiple hypotheses within a single comprehensive model, an approach that would otherwise require several separate analyses using other statistical techniques (Bryan, 2015; Sathyanarayana & Mohanasundaram, 2024:562). Hair et al. (2014:565) also highlight that one of the advantages of using SEM is that it allows the simultaneous use of several indicator variables per construct, which leads to more valid conclusions on the construct level. According to Randall and Lomax (2022:136), SEM addresses measurement error by directly modelling the error terms linked to observed variables. This approach ensures that the estimated relationships between constructs remain accurate and free from bias caused by measurement inaccuracies, effectively reflecting relationships among perfectly reliable variables. SEM was therefore used to examine the relationships among the variables of the study. Before SEM is assessed for overall model fit, a CFA is conducted to validate the measurement model. CFA assesses whether the observed variables load significantly into their respective latent, as specified in the model (Hair et al. 2010:567-567). In other words, CFA is used to evaluate the structure and validity of the measurement model (Hair et al. 2020:109). In CFA, regression weights are determined. Regression weights (also referred to as factor loadings) represent the strength and direction of the relationship between observed variables and their underlying latent factors (Hair et al. 2010:567-567; Tabachnick & Fidell, 2019:476). Once CFA is conducted, the SEM model is assessed for overall model fit. Various indices were used in the study to test the goodness of fit of the proposed hypothesised model. The goodness of fit indicates how well the specified model reproduces the observed covariance matrix among the indicator items (Kline, 2023:141; Sathyanarayana & Mohanasundaram, 2024:563). Goodness of fit is evaluated, using a number of model fit indices, which examines the relationships between the observed data and the theoretical model which would be expected from the model (Hair et al. 2010:567; Sathyanarayana & Mohanasundaram, 2024:563). The first step in CFA is to assess convergent validity, average extracted variance (AVE) and composite reliability, as Cheung et al. (2024:746) suggest. Convergent validity is a concept of SEM that assesses whether different measures that are meant to measure the same constructs correlate well with each other. It is confirmed through high loadings (Hair et al. 2021:78). The AVE is a measure used to assess a construct’s convergent validity. It represents the mean of the squared loadings of the indicators linked to latent construct and reflects the construct’s commonality (Hair et al. 2021:109). An AVE value of at least 0.50 is deemed acceptable, indicating that a construct explains at least 50% or more of the variance in its indicators (Bryan, 2015; Cheung et al. 2024:750). Thus, AVE measures the proportion of variance captured by a latent construct relative to that attributed to measurement error (Hair et al. 2021:77). It summarises the overall convergent validity of a set of items. The high loadings of 0.50 and more confirmed Clients Seeking Redress … A South African Banking Fraud Perspective 58 convergent validity in the study. Once validity is checked, the next step is to assess reliability. Composite reliability indicates the extent to which a set of items consistently measures the same latent construct (Hair et al. 2021:77). Unlike Cronbach’s alpha, composite reliability does not assume equal indicator loadings. A high composite reliability (above 0.70) suggests that the items are well aligned in the measuring instrument. Composite reliability uses factor loadings of convergent validity to examine the internal consistency of the constructs (Cheung et al. 2024:750; Kline, 2023:141). 5. 5 EMPIRICAL RESULTS 5.1 DEMOGRAPHIC DATA The agency conducted the recruitment process of respondents, based on the researchers’ inclusion criteria. Using the question link, CINT invited suitable respondents (South African bank clients) to participate in the study through email and their mobile app (OpinionAPP). Respondents used the encrypted link, provided by CINT, to access the online questionnaire. Before completing the questionnaire, a consent letter and introductory section was provided, detailing the objective of the study. The link to the questionnaire was open to all South Africans bank clients in the platform. A total of 400 bank clients in South Africa completed the questionnaire. Of the 400 completed questionnaires, 399 (one had missing data) were useful for data analysis. Section A of the questionnaire consisted of demographic questions and gathered data, such as the gender, age, population group and educational level of the respondents. Descriptive statistics were used to calculate and summarise the demographic data of the respondents. Table 1 presents the demographic data of the respondents. TABLE 1: DEMOGRAPHIC DATA OF RESPONDENTS Gender Frequency Percent (Rounded off) Male 141 35.3 Female 258 64.7 Total 399 100 Age <20 years old 9 2.3 20-29 years old 164 41.1 30-39 years old 137 34.3 40-49 years old 70 17.5 50-59 years old 11 2.8 60+ years old 8 2.0 Total 399 100 Population Asian 20 5.0 Black 283 70.9 Coloured 30 7.5 White 64 16.0 Others 0 0.0 Not willing to say 2 0.5 Total 399 100 59 The Retail and Marketing Review: Vol21 Issue 2 (2025) ISSN:2708-3209 Educational level Frequency Percent (Rounded off) Matric certificate (Grade 12) 92 23.1 Diploma 54 13.5 Postgraduate diploma 12 3.0 Bachelor's degree 158 39.6 Honours degree 31 7.8 Master’s degree 45 11.3 Doctorate 7 1.8 Total 399 100 Source: Statistical results As shown in Table 1, the majority of respondents were females (64.7%), and 35.3% were males. A large share of the sample consisted of younger bank respondents with those aged 20 and 29 years making up 41.1%. Respondents younger than 20 years old constituted 2.3% of the sample. Those aged 30 and 39 years made up 34.3%, while respondents between 40 and 49 years accounted for 17.5%. The 50 and 59 age group represented approximately 2.8% and respondents aged 60 years and above comprised approximately 2.0% of the sample. In terms of the population group, Black respondents were the majority at 70.9%, followed by the White population group (16%) and the Coloured population group (7.5%). The Asian population group represented 5.0% of the sample, and a small fraction of respondents (0.5%) chose not to disclose their population group. Lastly, the majority of respondents selected a bachelor’s degree as their highest qualification (39.6%), followed by Matric certificate (Grade 12) (23.1%), Diploma (13.5%), Masters (11.3%), Honours degree (7.8%), Post graduate diploma (3.0%), and Doctorate degree (1.8%). 5.2 VALIDITY AND RELIABILITY The validity and reliability of the measuring instrument were assessed using CFA. TABLE 2: VALIDITY AND RELIABILITY RESULTS Items Statement Loading P-values AVE CR SR7 I requested the bank to resolve the matter. 0.922 *** 0.692 0.878 SR3 I requested the bank to investigate the fraudulent transaction. 0.880 *** SR1 I complained directly to the bank. 0.868 *** SR8 I requested the bank to ensure there will be no re-occurrence of the same problem. 0.850 *** SR2 I requested an explanation from the bank. 0.843 *** SR5 I requested the bank to recover my stolen money. 0.823 *** SR4 I requested a refund from the bank. 0.725 *** SR6 I requested the bank to improve the security related to digital banking facilities. 0.720 *** Sat1 I am satisfied with the bank. 0.930 *** 0.792 0.948 Sat2 The services of the bank meet my expectations. 0.912 *** Sat4 I am satisfied with the bank’s products and services. 0.895 *** Sat5 Overall, my experience with the bank has been good. 0.892 *** Sat3 The digital banking options offered by the bank, for example internet or mobile banking, meet my expectations. 0.817 *** Tru6 I have confidence in how the bank delivers its services. 0.931 *** 0.768 0.934 Tru5 I believe that the bank delivers its services as expected. 0.930 *** Tru4 I believe that the bank’s products perform as expected. 0.906 *** Tru3 I trust the bank with keeping my money safe. 0.902 *** Tru1 I trust the bank's digital banking facilities. 0.811 *** Tru2 I trust the bank's staff. 0.764 *** Clients Seeking Redress … A South African Banking Fraud Perspective 66 Mayombo, M.B. 2014. The influence of customer complaint behaviour, firm responses and service quality on customer loyalty of mobile telephone subscribers in Uganda. University of Nairobi, Kenya. Menon, V., Grover, S., Gupta, S., Indu, P.V., Chacko, D. & Vidhukumar, K. 2025. A primer on reliability testing of a rating scale. Indian Journal of Psychiatry, 67(7): 725-729. Moqbel, M., Guduru, R. & Harun, A. 2020. Testing mediation via indirect effects in PLS-SEM: A social networking site illustration. Data Analysis Perspectives Journal, 1(3): 1–6. Ndubisi, N.O. & Wah, C.K. 2005. Factorial and discriminant analyses of the underpinnings of relationship marketing and customer satisfaction. International Journal of Bank Marketing, 23(7): 542–57. Oluwasanmi, O., Adeyemi, E.A. & Akanji, E.0. 2023. Customer complaint management, service quality and marketing performance in the service industry. Advances in Management, 22(1): 20-39. Ombudsman for Banking Services. 2023. Banking fraud [Online]. Available from: https://nfosa.co.za/ombudsmansounds-alarm-on-virtual-banking-card-fraud/ [Accessed 12 April 2024]. Ong, C. E. & Teh, D. 2016. Redress procedures expected by consumers during a business-to-consumer ecommerce dispute. Electronic Commerce Research and Applications, 19(4):1–13. Parikh, P. & Dutt, C.S. 2022. Impacts of staff demographics on complaint management behaviour in five-star hotels in Dubai. International Hospitality Review, 36(2): 199-219. Phiri, J., Lavhengwa, T. & Segooa, M.A. 2024. Online banking fraud detection: A comparative study of cases from South Africa and Spain. South African Journal of Information Management, 26(1):1763. Pio, P. G. C., Sigahi, T., Rampasso, I. S., Satolo, E. G., Serafim, M. P., Quelhas, O. L. G., Leal Filho, W. & Anholon, R. 2024. Complaint management: Comparison between traditional and digital banks based on ISO 9001 requirements. International Journal of Productivity and Performance Management, 73(4): 1050–1070. Puente-Cavazos, C.D., Cavazos-Arroyo, J. & Puente-Díaz, R. 2025. The impact of perceived financial security in neobanks on customer retention and the cost of switching to a traditional bank: the mediating role of trust in neobanks. Brazilian Journal of Business Management, 27(1):2024-0079. Rahman, M.M. 2023. Sample Size Determination for Survey Research and Non-Probability Sampling Techniques. Journal of Economics, Business and Entrepreneurship, 2(2): 55–60. Rahman, M.A. & Ramli, M.F. 2016. The influence of relationship quality on customer loyalty in the dual-banking system in the northern states of peninsular Malaysia. Procedia-Social and Behavioral Sciences, 219(2):606-613. Randall, E.S. & Lomax, R.G. 2022. 3rd edition. A Beginner's Guide to Structural Equation Modelling. Routlage press, London. Rosário, A. & Casaca, J.A. 2023. Relationship marketing and customer retention-a systematic literature review. Studies in Business and Economics, 18(3): 44-66. 67 The Retail and Marketing Review: Vol21 Issue 2 (2025) ISSN:2708-3209 Roy, S.K., Singh, G., Gruner, R.L., Dey, B.L., Shabnam, S., Muhammad, S.S. & Quaddus, M. 2023. Relationship quality in customer-service robot interactions in Industry 5.0: An analysis of value recipes. Information Systems Frontiers, 1-20. Rudd, H.D., Huo, H. & Xu, G. 2022. Improved churn causal analysis through restrained high-dimensional feature space effects in financial institutions. Human-Centric Intelligent Systems, 2(3): 70-80. Sabir, R.I., Ghafoor, O., Akhtar, N., Hafeez, I. & Rehman, A.U. 2014. Factors affecting customer satisfaction in banking sector of Pakistan. International Review of Management and Business Research, 3(2): 1014-1025. SABRIC. 2023. Annual crime stats [Online], Available from: https://www.sabric.co.za/media-and-news/pressreleases/ [Accessed 16 March 2024]. Sathyanarayana, S. & Mohanasundaram, T. 2024. Fit indices in structural equation modeling and confirmatory factor analysis: reporting guidelines. Asian Journal of Economics, Business and Accounting, 24(7): 561-577. Shantikuma, S. 2018. Validity, reliability and generalisability [Online]. Available from: https://www.healthknowledge.org.uk/content/validity-reliability-and-generalisability [Accessed 19 March 2024]. Singh, J. 1988. Consumer complaint intentions and behavior: Definitional and taxonomical issues. Journal of Marketing, 52(1): 93-107. Singh, V., Sharma, M.P., Jayapriya, K., Kumar, B.K., Chander, M.A.R.N. & Kumar, B.R. 2023. Service quality, customer satisfaction and customer loyalty: A comprehensive literature review. Journal of Survey in Fisheries Sciences, 10(45): 3457-3464. Smith, B. 1998. Buyer‐seller relationships: Bonds, relationship management, and sex‐type. Canadian Journal of Administrative Sciences, 15(1): 76-92. Sofia, R.D., Afiff, A.Z. & Aruan, D.T.H. 2023. The effect of attributions and failure severity on consumer complain behaviors in sharing economy. Cogent Business & Management, 10(3): 2275848. Sousa, R. & Voss, C.A. 2009. The effects of service failures and recovery on customer loyalty in e‐services: An empirical investigation. International Journal of Operations and Production Management, 29(8): 834-864. Suttikun, C. & Meeprom, S. 2021. Examining the effect of perceived quality of authentic souvenir product, perceived value, and satisfaction on customer loyalty. Cogent Business and Management, 8(1): 1976468. Tabachnick, B.G. & Fidell, L.S. 2019. Using multivariate statistics. 7th edition. Pearson. Tanko, D., Tuncer, T., Dogan, S. & Akbal, E. 2022. Regression methods for social media data analysis. Mugla Journal of Science and Technology, 8(1):31-40. Tax, S.S., Brown, S.W. & Chandrashekaran, M. 1998. Customer evaluations of service complaint experiences: Implications for relationship marketing. Journal of Marketing, 62(2):60-76. Clients Seeking Redress … A South African Banking Fraud Perspective 68 Telang, R. & Somanchi, S. 2016. Security, fraudulent transactions and customer loyalty: A field study.Carnegie Mellon University, USA. Turaga, N.K., Patibandla, J.K., Turaga, M. & Bandaru, S.R. 2020. Factors influencing to complain by first time customers and loyal customers in DTH services. J Crit Rev, 7(12):1076-1080. Uruena, A. & Hidalgo, A. 2015. Justice and emotions in service recovery: A complaint in B2C e-commerce. International Journal of Engineering Management and Economics, 5(1-2):3-18. van Deventer, M. & Redda, E. H. 2023. Customer loyalty and trust in South African retail banking. Banks and Bank Systems, 19(2): 211–222. Xia, Y. & Yang, Y. 2019. RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: The story they tell depends on the estimation methods. Behavior research methods, 51(1): 409-428. Zephan, N. 2018. Relationship between customer satisfaction and customer loyalty. Case study: Hilton hotel Yaoundé, Cameroon. (Unpublished dissertation), Centria University of Applied Science, Cameroon. Zhao, S., Jiang, Z., Li, Y. & Meng, J. 2023. How does Word-of-mouth Marketing Affect Customer Loyalty. Advances in Economics, Management and Political Sciences, 10(1): 51-57. Zietsman, P.J. 2017. The relationship between relationship quality constructs, overall satisfaction and loyalty: An emerging market perspective of SME banking customers. (Unpublished dissertation), University of Pretoria, South Africa.