https://orcid.org/0000-0002-6859-4488
Figures
Abstract
Background
Online malicious attempts such as scamming continue to proliferate across the globe, aided by the ubiquitous nature of technology that makes it increasingly easy to dupe individuals. This study aimed to identify the predictors for online fraud victimization focusing on Personal, Environment and Behavior (PEB).
Methods
Social Cognitive Theory (SCT) was used as a guide in developing the PEB framework. Specifically, three factors were identified—Self-awareness (Personal), Attitude (Personal and Environment) and Safe Practice (Behavior) as the potential predictors for online fraud victimization. A self-reporting questionnaire was developed based on the PEB framework and used to collect data targeting Malaysian adults. The study reports result from two separate datasets collected across two separate timelines. Study I involved data collection in January 2023 (n = 820) whereas Study II was conducted with a modified questionnaire from November 2023 –January 2024 (n = 629). Study I identified the online fraud victimization predictors through an Exploratory Factor Analysis (EFA) and a hierarchical binary logistic regression. The dataset from Study II was used to validate the online fraud victimization model derived from Study I by executing another round of hierarchical binary logistic regression.
Results
Results from both the samples show that most of the respondents are aware of digital privacy. EFA from Study I yielded a five-factor solution with a total variance of 60.6%, namely, Self-awareness, Safe Practice, Bank Trust, Overconfidence and Social Influence. Hierarchical binary logistic regression results from both the studies were found to be consistent. Specifically, Overconfidence (β = 0.374; OR = 1.453; 95% CI [1.119, 1.887]; p = 0.005) and Social Influence (β = 0.332; OR = 1.225; 95% CI [1.077, 1.512]; p = 0.006) were found to significantly predict online fraud victimization as well as gender (β = 0.364; OR = 1.440; 95% CI [1.008, 2.016]; p = 0.045) with females exhibiting higher risks to victimization.
Implications
The emergence of Overconfidence and Social Influence as significant predictors can guide the development of targeted online fraud awareness campaigns and/or tools emphasizing critical thinking and skepticism. Policymakers can leverage this knowledge to implement regulations that reduce deceptive practices online, promote digital literacy programs, and mandate clearer consumer protections to mitigate the impact of social manipulation and overconfidence on fraud victimization.
Conclusion
This study identifies online fraud victimization predictors, hence improving our understanding of the factors behind this phenomenon—allowing for the development of effective preventive measures and policies to safeguard individuals and improve digital security. For instance, gender- specific educational campaigns can be developed to enhance awareness and equip women with strategies to detect and avoid scams. Additionally, addressing systemic factors like social norms and digital literacy gaps is crucial for creating equitable and effective solutions to reduce online fraud victimization.
Citation: Balakrishnan V, Ahhmed U, Basheer F (2025) Personal, environmental and behavioral predictors associated with online fraud victimization among adults. PLoS ONE 20(1): e0317232. https://doi.org/10.1371/journal.pone.0317232
Editor: Daniel Ioan Hunyadi, Lucian Blaga University of Sibiu: Universitatea Lucian Blaga din Sibiu, ROMANIA
Received: October 8, 2024; Accepted: December 24, 2024; Published: January 29, 2025
Copyright: © 2025 Balakrishnan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available from Open Science Framework: https://osf.io/m3gxr/. DOI 10.17605/OSF.IO/M3GXR.
Funding: This study was supported by the Ministry of Higher Education of Malaysia through the Fundamental Research Grant Scheme 2023-1 [FP002-2023; Reference Code: FRGS/1/2023/ICT07/UM/02/1] and partially by the Joint Research Scheme 2023 (SATU) [ST003-2023: UM.0001259/HRU.OP]. Both the fundings are awarded to the main author (VB) and they are external to VB’s organization. There is no additional internal or external funding received for this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The Internet and digital technology have become an integral part of our daily lives, a trend that is further amplified by the COVID-19 pandemic [1, 2]. Online consumer behavior witnessed a rapid change with majority of people resorting to digital technologies for most routine transactions including shopping, bill payments and procuring services, among others. This also includes the accelerated use of online platforms such as social media, as avenues to seek information, provide support, and to build relationships. However, these may have inadvertently increased the digital attack surface and amplified issues related to digital security and privacy. An increase in fraudulent activities ranging from scamming, phishing to data breach was observed worldwide. These activities are often accomplished through “social engineering” [3]–influencing/tricking people to perform unsafe actions such as clicking on fake website links, divulging personal and confidential details (e.g., username and passwords), opening malware enabled attachments, and downloading untrusted third-party apps for fraudulent payment transactions etc. for malicious gains [4].
Malaysia too, has witnessed a healthy rate of digitization following the COVID-19 outbreak. Reports on online fraud activities, particularly scamming are rampant worldwide, a trend that is considered worrying in Malaysia. For instance, there were 13,703 reported scam cases in the country involving losses of MYR 539 million (USD 124 million), a figure that increased to 20,701 reported cases with losses amounting to MYR 560.8 million (USD 129 million) in 2021 [5]. Most of these cases were associated with online trading, fake loans, Short Message Service (SMS) scams, African/Nigerian (request to help facilitate the illegal transfer of money) and Macau (tricking someone into disclosing personal banking details or transferring money into third party accounts) scams. The increase in such fraudulent activities was attributed to rising expertise in technology and low level of awareness on cybercrimes among the people [6].
The rise in digital technologies that are most often integrated, means online fraudulent activities to be a growing issue across any society. However, the present understanding of what makes an individual susceptible to online fraud is relatively poor. The concept of digital privacy is broad regardless of the platform. In the context of social media, for example, privacy may refer to one’s tendency to freely disclose information (public versus private), or concern about how personal data is being used (or misused) by others. It is pertinent to understand why online fraudulent victimization is on the rise, and this is probably best done by exploring some basic themes, namely, awareness, attitude and behavior towards digital privacy. The study therefore aims to explore predictors leading to online fraud victimization by focusing on personal, environment and behavior–the three main tenets of Social Cognitive Theory (SCT). Specifically, we aim to answer two main research questions (RQs):
RQ1: What is the awareness level of digital privacy among Malaysian adults?
RQ2: What are the significant predictors for online fraud victimization?
Theoretical framework and research hypotheses
This empirical study is guided by a theoretical framework encompassing three main factors, namely, Self-awareness, Attitude and Behavior, developed based on SCT that emphasizes on people, who are seen as active agents who can both influence and are influenced by their environment [7]. Specifically, SCT proposes that human behavior is influenced by three key tenets: personal such as individual beliefs, attitude, values etc., environment such as social norms and cultural influences, and behavioral including skills and abilities. For example, Self-awareness is deemed to be a critical personal factor that can influence decision making by affecting how individuals perceive and interpret information. According to SCT, this enables individuals to make informed decisions as they are more aware of their actions and potential consequences of their decisions.
According to Fenigstein et al. [8], the disposition for self-focused attention is an individual difference factor related to resistance to any form of influence. Individuals with a high sense of self-awareness, for example, are known to consider their personal knowledge, internal norms and attitudes to a greater degree when making decisions and thus leading to increased resistance to social influence and persuasion attempts [9]. Studies have shown that being aware of the risks associated with online mechanisms as well as the tactics used by fraudsters can help individuals be more vigilant, and help identify and assess risks, hence they are better equipped in protecting themselves [10–13]. Therefore, we posit that individuals with high self-awareness are less susceptible to online fraud victimization. Hypothesis one is given as follows:
- H1 –Self-awareness negatively predicts online fraud victimization
Another personal factor that is known to affect individuals’ behavior and decision-making is attitude, which refers to individual characteristics, portraying either positive, neutral or negative behavior to a certain concept or subject [14]. It can be influenced by several factors such as social norms (i.e., environment) personal values, beliefs and experiences. Attitude has been well-studied in various domains, ranging from those focusing on threat appraisal and coping strategies of online fraud [12, 13], technology adoption [15–17], to social issues including misinformation spread [18] and online hate [19]. These studies generally show that a positive attitude promotes good behavior, and vice-versa.
In the context of digital privacy, a negative attitude towards sharing personal information online was found to significantly predict phishing susceptibility [12, 13]. Similar findings were reported by those focusing on self-efficacy and online scams [20, 21]. We hypothesize that a negative attitude such as being overconfident or lackadaisical can lead to online fraud victimization, as it may result in individuals being careless or causing them to make quick decisions without considering the potential consequences of their actions. Further, environment has a pertinent role in how an individual behaves or reacts to a situation. For example, some individuals have a higher propensity in trusting people, hence making them more susceptible to fraud victimization. As stated in Introduction, most online fraud activities are done through social engineering [3, 22, 23] by gaining the trust of victims, and thus individuals who are overly trusting are more vulnerable to manipulation and may less likely take precautions to protect themselves from fraud/fraudsters. In this study, attitude covers a wide range of elements including personal and environment factors. In general, hypothesis two is given as below:
- H2 –A negative Attitude positively predicts online fraud victimization
Finally, the third factor is user behavior. Though this may encompass a wide spectrum of behavior, this study is specifically interested in secure or safe online practices adopted by individuals as a means of protection. For example, making it a norm to set and use strong passwords containing numbers, upper- and lower-cases, and/or special symbols, ensuring computer and mobile phone software are up to-date, being vigilant when using the public wi-fi or using a virtual private network (VPN) are some of the safe measures one can adopt to ensure internet security. As a matter of fact, studies have shown that adopting and practicing safe measures reduces breach of security, online scamming and other fraudulent activities [24–27]. Hypothesis three therefore, posits that individuals who practice safe measures are less susceptible to online fraud.
H3 –Safe Practice negatively predicts online fraud victimization
Materials and methods
This study explored the predictors for online fraud victimization by focusing on personal, environment and behavioral factors based on SCT. The conceptual framework was built based on previous literatures, specifically focusing on Self-awareness [10, 12, 13, 28], Attitude [12, 13, 20, 21] and Safe Practice [11, 25, 26], as illustrated in Fig 1.
Data collection, analysis and validation were done over the course of two timelines (i.e., January 2023 –January 2024), resulting in two sets of data. We differentiate the studies as Study I and II.
Study I
Study design and sample
The initial study (Study I) aimed to identify the predictors for online fraud victimization. The data collection was conducted from January 31 to February 23, 2023, using a self-administered online survey hosted on the Google Forms platform. The link to the survey was distributed through social media, and WhatsApp. Using a calculator based on Cochran’s equation [29], the appropriate sample size given the specified combination of precision (±5%), confidence (95%) and variability (0.5) was 385. Targeting Malaysian adults (i.e., 18 and above), a total of 820 respondents were recruited.
Instrument
A questionnaire was prepared in English and piloted among 36 respondents. Ambiguities were resolved, questions were rephrased and restructured to enhance the readability of the survey instrument. This was then translated into the national language (i.e., Malay), and verified by four linguistic experts. Both versions were provided to the respondents, with an option to pick one. The instrument had two main sections:
Section A–This was used to gather respondents’ demographic details such as age (continuous), gender, ethnicity, hours spent daily on social media etc. Respondents were also asked to rate their digital literacy (beginner, intermediate, advanced) and online risky behavior levels (low, medium, high). The level of digital privacy awareness was assessed through a single question (i.e., are you aware of digital privacy?). Further, another question determined online fraud victimization (i.e., have you been a victim of any online fraud activity, e.g., scammed, data breach, identity theft etc.?), which was used as the dependent variable in this study. The complete items can be found in Table 1 (Results).
Section B–This comprised three subsections measuring respondents’ perceptions, namely, attitude (n = 11; Cronbach’s alpha (α) = 0.87), behavior (n = 10; α = 0.76) and self-awareness (n = 11; α = 0.912). Example items for attitude were “I trust people easily”, “Online banking is safe”, “I will never fall victim as I am always careful” and “I trust the safety mechanisms adopted by my banks” etc. On the other hand, items for behavior focused on the safe/risky measures adopted such as “I always perform 2-step verification for my personal accounts”, “I browse online using a secure VPN”, and “I click on online links as they are generally safe” etc. whilst items for awareness include “I know the dangers of digital footprints” and “I am aware that my personal data are collected online”, among others. All the 32 items in Section B were measured using a five-point Likert scale indicating the extent of a respondent’s agreement with the items (5 = Strongly agree, 4 = Agree, 3 = Neutral, 2 = Disagree, and 1 = Strongly disagree). All the negative statements were re-coded prior to analysis. Cronbach’s alpha (α) was used to determine the internal consistency and reliability, and the scores for each of the factors ranged between 0.76 and 0.91 (i.e., > 0.7), hence indicating a good reliability. The complete questionnaire is attached as SI.
Data analysis
Frequency, percentage, mean and standard deviations (SD) were used to describe the sample. Exploratory Factor Analysis (EFA) was used to examine the underlying factor structures whereas a hierarchical binary logistic regression was performed to identify the significant predictors for online fraud victimization. The analyses were conducted using IBM SPSS version 28. Results are deemed significant at p ≤ 0.05.
Skewness and kurtosis values indicate values between −1.98 and +1.98 [30, 31], hence the data were considered normally distributed. Preliminary analysis revealed no multicollinearity issues as the communalities were above 0.5, the inter-item correlations and item-to-total correlations were above the recommended values of 0.3 and 0.5, respectively [32]. Further, the Kaiser-Meyer-Olkin (KMO) of sampling adequacy score was 0.869, indicating sample size adequacy. The Bartlett’s Test of Sphericity also supported the factorability of the correlation matrix, with x2 = 9233.34; degree of freedom (df) = 465; p < 0.001 [33]. The number of factors were determined based on the Eigenvalues (i.e., more than 1.00) whilst items with a factor loading less than 0.5 were removed.
Ethics approval
This study was approved by the Universiti Malaya Research Ethics Committee [UM.TNC 2/UMREC]. Written individual informed consent was obtained from all participants. All the data collection and analysis method used in this study complied with the terms and conditions for the source of the data.
Results
Demographics
Demographic profiles of the 820 Malaysian respondents are provided in Table 1. There was a total of 820 respondents aged between 18 and 67 years old (Mean = 27.14; SD = 11.4), mainly comprising Gen Z (69.8%; 18–26 years), followed by Gen Y (14.6%; 27–42 years) and Gen X (14.3%; 43–58 years), in line with the status whereby most of the respondents were tertiary students (65.2%). Gender segregation is balanced (male: 374 versus female: 446). Most of the respondents were also active as they spent between 1 and 5 hours on their phones (62.7%) and social media (66.6%) daily. Forty-two percent of the respondents reported to have been a victim of online fraud (at least once) (42%), of which the majority claimed the perpetration to have taken place on social media (26%) and phone calls/Short Messaging Services (SMS) (18.9%).
In terms of digital literacy, most of the respondents reported to be at the intermediate level for technology (56.1%), fake news literacy (62.9%) and fact-checking literacy (55.2%). A similar pattern was observed for online risky behavior level, with the majority reported to be intermediate (62.8%). Finally, Table 1 also shows that 81.6% (n = 669) are aware of digital privacy. This shows that the level of digital privacy awareness is high among our sample, hence answering our RQ1.
Underlying factors for online fraud victimization
To answer RQ 2, an EFA was performed using all the items in Section B. This is then followed by a hierarchical binary logistic regression. Using Principal Component Analysis (PCA) with Varimax rotation, the EFA produced a five-factor solution, with a total variance of 60.6%. Table 2 shows the results of EFA, along with mean, SD and α.
All 11 items assessing awareness loaded into the first cluster, hence the factor’s name was retained as Self-awareness. The high mean score (4.05) indicates most of the respondents were agreeable in having a high sense of self-awareness towards digital privacy, a finding that accords with the level of digital privacy awareness reported in Table 1. Similarly, seven items focusing on behavior loaded into the second cluster, therefore the name was retained as Safe Practice. Three items were removed as they had a loading below 0.5, namely, “I change my passwords regularly”, “I use the same password for every account”, and “I browse online using a secure VPN”. The mean score of 3.41 generally indicates the respondents to be somewhat agreeable in adopting safe online measures.
Items measuring Attitude split three ways–three items focusing on an individual’s confidence (i.e., overconfidence) loaded together, therefore this was named as Overconfidence, two specifically focusing on trust in banks loaded together, hence this was named as Bank Trust. Finally, three items on impact of social influence and being overly trusting loaded together, and thus this was named as Social Influence. Three items were removed due to a low load, that is, “The Internet is safe”, “Scammers only target older generations”, and “I do not care about my privacy setting”. The α values for Overconfidence and Bank Trust were 0.74 and 0.80, respectively. The α value for Social Influence was 0.65, however, since a reliability score between 0.5 and 0.7 is deemed to be moderate, this factor was retained.
Based on the results of EFA, hypothesis two is rephrased as follows:
- H2a –An overconfidence attitude positively predicts online fraud victimization
- H2b –A high trust in banking institutions positively predicts online fraud victimization
- H2c– A high social influence positively predicts online fraud victimization
Online fraud victimization predictors.
Finally, a hierarchical binary logistic regression was performed to identify the significant predictors for online fraud victimization. The dependent variable was online fraud victimization (Yes = 0; No = 1). Frequency of social media use, being aware of digital privacy, gender (male = 0; female = 1) and online risky behaviour level were entered in Block 1 (Model 1), followed by all five factors from EFA in Block 2 (Model 2). We report on the results for Model 2 in Table 3.
The overall model was found to be fit (x2 = 46.67; degree of freedom (df) = 16; p < 0.001), with a -2 log likelihood error of 1052.15 (Nagelkerke R2 = 7.6%; Hosmer and Lemeshow p-value = 0.807). The classification accuracy was 62% compared to 56.9% for Model 1, indicating an improvement in online fraud victimization prediction with the inclusion of the EFA factors.
Looking at the p-values in Table 3, two significant predictors were found for online fraud victimization, namely Overconfidence and Social Influence. A higher level of Overconfidence was associated with higher odds of online fraud victimization (β = 0.362; Odd-ratio (OR) = 1.437; 95% CI [1.234, 1.673]; p < 0.001), indicating that the more overconfident an individual is the more susceptible he/she is toward online fraud victimization. Similarly, results show that an individual to more likely fall victim online when there is a higher Social Influence (β = 0.263; OR = 1.169; 95% CI [1.061, 1.594]; p < 0.001). The rest of the predictors were insignificant. In short, H2b and H2c are supported.
Study II
The second study was conducted to validate the findings from the previous study, particularly in determining the significant predictors for online fraud victimization. All the approaches outlined in Study I were replicated, with minor modifications to the questionnaire. Specifically, items that were removed during EFA in Study I were excluded (i.e., “I change my passwords regularly”, “I use the same password for every account”, “I browse online using a secure VPN”, “The Internet is safe”, “Scammers only target older generations”, and “I do not care about my privacy setting”. No other changes were made. Data collection took place from November 11 to January 20, 2024, resulting in 630 valid responses. However, one response was excluded for not meeting the study’s criteria (i.e., not being Malaysian), leaving a final sample of 629 respondents. The demographics of this group closely resembled those from Study I, as shown in Table 4, which provides a detailed demographic profile.
Generation segregation was found to be similar with most of the respondents from Gen Z (80.2%), followed by Gen Y (11.7%) and Gen X (6%). Gender segregation is balanced, however there were higher number of males (346) than females (283). Most of the respondents were also active on social media spending between 1 and 5 hours daily (57.2%). Thirty-two percent reported to have experienced online fraud victimization with social media, chat applications and phone calls/SMS emerging as the top three mediums. Digital, fake news and fact-checking literacy patterns as well as online risky behavior level were similar with most respondents reported to be at the intermediate level. As in Study I, many Study II respondents were aware of digital privacy (83.1%), followed by those who were unsure (12.1%) and those unaware (4.8%).
Online fraud victimization prediction model
The hierarchical binary logistic regression also revealed Model 2 to be significant and fit (x2 = 22.96; df = 12; p = 0.028), with a -2 log likelihood error of 765.21 (Nagelkerke R2 = 5%; Hosmer and Lemeshow p-value = 0.743). A slight improvement was noted for the classification accuracy (68.2% compared to 67.6% for Model 1), resembling a similar pattern with Study I.
Model 2 produced similar results for the second dataset as shown in Table 5, with Overconfidence (β = 0.374; OR = 1.453; 95% CI [1.119, 1.887]; p = 0.005) and Social Influence (β = 0.332; OR = 1.225; 95% CI [1.077, 1.512]; p = 0.006) emerging as significant predictors. Additionally, gender was also found to be significant (β = 0.364; OR = 1.440; 95% CI [1.008, 2.016]; p = 0.045), indicating that females are more likely to be victimized online compared to males.
Discussion
Considering the spike in online fraud victimization cases, and the lack of understanding of the predictors leading to this phenomenon, this study explored the predictors associated using a self-administered questionnaire survey developed based on SCT. Using a sample size of 820 Malaysian adults, the proposed model was tested and further validated with a new dataset (n = 629) using hierarchical binary logistic regression. The study found Overconfidence, Social Influence and gender (female) to significantly predict online fraud victimization, whilst Bank Trust and Self-awareness did not. Further, a high level of digital privacy awareness was also observed among both the study respondents.
Findings indicate a positive link between Overconfidence and the likelihood of being a victim of online fraud, a finding that was reflected in previous studies [34–37]. Individuals may be overconfident in their ability to handle a risky situation (e.g., bullying victimization) even when the situation increases the likelihood of victimization [38]. One plausible reason is that the ability to distinguish between legitimate and fraudulent information may be lacking in overconfident people [39], and they may also be less likely than others to follow the news about financial frauds, which would reduce their awareness of the problem and leave them open to deception in the future [25, 40]. Another possibility is that excessive self-assurance fosters the appearance of control, which discourages people from adopting preventative measures such as seeking financial advice, hence making them more susceptible to financial frauds [41]. This is in fact, reflected in our survey items whereby respondents somewhat agreed that they are in control of their data and privacy, and believe that they are immune to online frauds and scams. Such an attitude is worrying as fraudsters often rely on their targeted victims’ overconfidence and apply psychological tactics to manipulate their victims [35]. This may also indicate that the respondents may somewhat lack digital privacy awareness despite claiming otherwise, or they may perceive that they are too savvy to be taken advantage of. This, therefore, warrants further investigation.
Social Influence predicts online fraud victimization, in line with previous studies that found the factor to positively affect online fraud victimization [42, 43]. In this study, Social Influence refers to one’s tendency to be influenced, manipulated and overly trusting others–a tendency that results in a higher odd of being victimized online. Similar patterns of observations were reported in other cybercrime studies such as gang membership and offending desistance [44], cyberbullying [45], and adolescent engagement in risky behaviors associated with health risk or illegal behavior [46]. This indicates the impact of an individual’s psychological nature in being highly susceptible to online frauds. Research has shown that people high in agreeableness trait are more likely to fall victim to frauds as they may be reluctant to question or challenge information presented to them [47, 48]. Similarly, those who are more susceptible to manipulation, perhaps due to a lack of knowledge or experience, may be more likely to be taken advantage of. Educational programmes and regular reminders and awareness campaigns may help in ensuring such vulnerable individuals to be more vigilant online.
In terms of demographic, being a female implies a higher tendency to be victimized online compared to males. This could be due to a combination of social, psychological, and behavioral factors. Studies suggest that women tend to engage more frequently in online activities such as social networking and e-commerce such as online shopping [49], which can expose them to a higher number of potential threats. In fact, according to Statista [50], more than 60% of women reported discovering products through social media compared to more than 50% of men in 2023. Additionally, social engineering tactics used by fraudsters [3], such as emotional manipulation or trust-building, may disproportionately affect females, who are often perceived as more empathetic or community-oriented [51]. This aligns with findings by [52] who highlight that women are more likely to engage in empathetic and trusting online behaviors, which may make them more likely to fall victim to schemes like phishing or romance scams. Furthermore, studies by Anderson [53] and Leukfeldt et al. [54] emphasize that fraudsters may exploit gender stereotypes, assuming that women are less familiar with technical aspects of online security, making them easier targets.
This study found that falling victim to online frauds was not significantly predicted by having a high level of Bank Trust. This finding is consistent with earlier research [48, 55], however it is at contrast with Farkhondeh et al. [56]. Likewise, this observation supports the notion that trust did not predict fraud vulnerability [57]. One probable explanation is that Internet fraudsters or scammers frequently used strategies such as using the names of reputable companies or people and sending fictitious messages to their targeted victims [58]. People will specifically lose trust in an institution if they think genuine messages are fraudulent ones, in addition to financial losses. The victimization of online frauds may therefore not be significantly predicted by Bank Trust.
Similarly, Self-awareness did not significantly predict online fraud victimization. This concurs with [59], however in contrast with [60, 61]. This could be attributed to the high level of digital privacy awareness among our respondents (81.9%; 83.2%). Respondents are mainly urbanite educated individuals who are also digitally active, with intermediate and advanced levels of literacy for technology, fact checking and fake news, therefore they may be more knowledgeable in terms of digital privacy and Internet safety, a pattern that was reflected in [62]. This probably explains why Self-awareness was insignificant in predicting online fraud victimization.
Safe Practice was found to be insignificant in mitigating online fraud victimization in this study, in contrast to some other studies [63–65]. The mean score (i.e., 3.41, Table 2), suggests that most respondents across both studies displayed a moderate to high level of agreement in exercising caution and adhering to good online practices. This indicates that most of the respondents are already somewhat cautious in their online behavior. Furthermore, given that a significant portion of the respondents belong to the younger generation, they are likely to be more tech-savvy and familiar with adopting standard online safety measures. Their digital fluency and regular exposure to online environments may have contributed to the widespread practice of basic security protocols, thus diminishing the predictive power of Safe Practice as a unique factor in preventing online fraud. Essentially, Safe Practice may not have emerged as a significant predictor because it is already widely implemented by this demographic, making other factors more relevant in explaining variations in fraud victimization.
Conclusion
This study explored and determined critical predictors predicting online fraud victimization. Using the SCT as the foundational theory for the framework development, EFA was conducted, resulting in the identification of five distinct independent variables: Safe Practice, Overconfidence, Social Influence, Bank Trust, and Self-awareness. Key findings related to our two main RQs show that most of the respondents sampled are aware of digital privacy, however knowledge alone does not prevent them from becoming a fraudulent victim online. Findings show that the propensity for an individual to be influenced by their social environment (Social Influence), high level of confidence (Overconfidence) and being female significantly increase the risk of falling victim to online frauds. This highlights the need for targeted interventions such as improving critical thinking skills, enabling individuals to better analyze and evaluate the credibility of information they encounter online. Further, fostering skepticism towards online communications can empower people to recognize and resist fraudulent schemes more effectively. Finally, gender-sensitive prevention programs can be designed to address the unique ways in which different genders, especially women might experience and respond to online scams, thus ensuring more inclusive and effective protective measures.
This study has limitations that need to be acknowledged. The study used a survey methodology to gather the quantitative data needed, but this method lacks the depth or richness of data that can only be obtained through longitudinal research using qualitative data, which can produce a stronger inference of causality [66]. Further, the sample is deemed skewed as many respondents were young, educated and from urban areas. This may have affected the outcome of the study. Further, digital privacy awareness as well as literacy levels were assessed based on self-perceptions, hence this may not reflect the picture. Future studies could explore other methods to determine individuals’ awareness and their literacy levels (e.g., using experiments). Finally, the study may also lack the generalizability desired as the respondents are of Malaysian origin. Future studies involving participants from various backgrounds or nationalities may yield similar or contrasting results that are worth pursuing.
References
- 1. Seetharaman P. Business model shifts: impact of COVID-19. Int J Inf Manag. 2020;54:102173.
- 2. Zhang Y, Wu Q, Zhang T, et al. Vulnerability and fraud: evidence from the COVID-19 pandemic. Humanit Soc Sci Commun. 2022;9:424. pmid:36466703
- 3. Atkins B, Huang W. A study of social engineering in online frauds. Open J Soc Sci. 2013;1(3):23–32.
- 4.
Greitzer FL, Strozer JR, Cohen S, Moore AP, Mundie D, Cowley J. Analysis of unintentional insider threats deriving from social engineering exploits. In: 2014 IEEE Security and Privacy Workshops; 2014 May; San Jose, CA. New York: IEEE; 2014. p. 236–250.
- 5. Basyir M, Harun HN. Online scam cases increasing in Malaysia. NST. 2022 [cited 2023 Oct 1]. Available from: https://www.nst.com.my/news/nation/2022/09/834531/online-scam-cases-increasing-malaysia.
- 6. Mardiah A. Ramp up security amid rise of financial fraud. The Malaysian Reserve. 2023 [cited 2023 Oct 1]. Available from: https://themalaysianreserve.com/2023/01/11/ramp-up-security-amid-rise-of-financial-fraud/.
- 7.
Bandura A. Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall; 1986.
- 8. Fenigstein A, Scheier MF, Buss AH. Public and private self-consciousness: assessment and theory. J Consult Clin Psychol. 1975;43(4):522–527.
- 9. Hutton DG, Baumeister RF. Self-awareness and attitude change: seeing oneself on the central route to persuasion. Pers Soc Psychol Bull. 1992;18(1):68–75.
- 10. Barth S, de Jong MD, Junger M, Hartel PH, Roppelt JC. Putting the privacy paradox to the test: online privacy and security behaviors among users with technical knowledge, privacy awareness, and financial resources. Telemat Inform. 2019;41:55–69.
- 11. Chen H, Beaudoin CE, Hong T. Securing online privacy: an empirical test on Internet scam victimization, online privacy concerns, and privacy protection behaviors. Comput Hum Behav. 2017;70:291–302.
- 12. Lee YY, Gan CL, Liew TW. Thwarting instant messaging phishing attacks: the role of self-efficacy and the mediating effect of attitude towards online sharing of personal information. Int J Environ Res Public Health. 2023;20:3514. pmid:36834209
- 13. Martens M, Wolf RD, Marez LD. Investigating and comparing the predictors of the intention towards taking security measures against malware, scams, and cybercrime in general. Comput Hum Behav. 2019;92:139–150.
- 14. Scherer R, Tondeur J, Siddiq F, Baran E. The importance of attitudes toward technology for pre-service teachers’ technological, pedagogical, and content knowledge: comparing structural equation modeling approaches. Comput Hum Behav. 2018;80:67–80.
- 15. Ahmad A, Madi Y, Abuhashesh M, Nusairat NM. The knowledge, attitude, and practice of the adoption of green fashion innovation. J Open Innov Technol Mark Complex. 2020;6(4):107.
- 16. Rahimi A, Azimi G, Asgari H, Jin X. Adoption and willingness to pay for autonomous vehicles: attitudes and latent classes. Transp Res Part D: Transp Environ. 2020;89:102611.
- 17. Santiago J, Borges-Tiago MT, Tiago F. Embracing RAISA in restaurants: exploring customer attitudes toward robot adoption. Technol Forecast Soc Change. 2024;199:123047.
- 18. Balakrishnan V., Ng K. S., & Rahim H. A. (2021). To share or not to share–The underlying motives of sharing fake news amidst the COVID-19 pandemic in Malaysia. Technology in Society, 66, 101676.
- 19. Weber M, Viehmann C, Ziegele M, Schemer C. Online hate does not stay online: how implicit and explicit attitudes mediate the effect of civil negativity and hate in user comments on prosocial behavior. Comput Hum Behav. 2020;104:106192.
- 20. Choi HS, Kim SS. The role of self-efficacy in consumers’ vulnerability to online scams: a moderated mediation model. J Consum Behav. 2021;20(5):663–671.
- 21. Kim J, Park H. Effects of perceived self-efficacy and risk perception on consumers’ responses to phishing scams. Int J Hum-Comput Interact. 2021;37(11):1009–1019.
- 22. Steinmetz KF, Pimentel A, Goe WR. Performing social engineering: a qualitative study of information security deceptions. Comput Hum Behav. 2021;124:106930.
- 23. Vrhovec S, Bernik I, Markelj B. Explaining information seeking intentions: insights from a Slovenian social engineering awareness campaign. Comput Secur. 2023;125:103038.
- 24. Pour MS, Nader C, Friday K, Bou-Harb E. A comprehensive survey of recent Internet measurement techniques for cyber security. Comput Secur. 2023;103123.
- 25. Thongmak M. Protecting privacy in Pokémon Go: a multigroup analysis. Technol Soc. 2022;70:101999.
- 26. Cerf S, Robu B, Marchand N, Bouchenak S. Privacy protection control for mobile apps users. Control Eng Pract. 2023;134:105456.
- 27. Wang D, Duan Y, Jin Y. Navigating online perils: socioeconomic status, online activity lifestyles, and online fraud targeting and victimization of old adults in China. Comput Hum Behav. 2025;162:108458.
- 28. Chen R, Gaia J, Rao HR. An examination of the effect of recent phishing encounters on phishing susceptibility. Decis Support Syst. 2020;133:1–14.
- 29.
Cochran WG. Sampling techniques. New York: John Wiley & Sons; 1977.
- 30.
Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis: a global perspective. 7th ed. Upper Saddle River (NJ): Pearson; 2010.
- 31.
Mooi E, Sarstedt M. A concise guide to market research: the process, data, and methods using IBM SPSS Statistics. New York: Springer; 2011. https://doi.org/10.1007/978-3-642-12541-6
- 32.
Field A. Discovering statistics using IBM SPSS. 4th ed. London: SAGE; 2013.
- 33. Kaiser HF. An index of factorial simplicity. Psychometrika. 1974;39(1):31–36.
- 34. Cheng JT, Tenney ER, Moore DA, Logg JM. Overconfidence is contagious. Harvard Business Review. 2020 [cited 2023 Oct 1]. Available from: https://hbr.org/2020/11/overconfidence-is-contagious.
- 35. Xiao X, Li X, Zhou Y. Financial literacy overconfidence and investment fraud victimization. Econ Lett. 2022;212:110308.
- 36. Fischer P, Lea SE, Evans KM. Why do individuals respond to fraudulent scam communications and lose money? J Appl Soc Psychol. 2013;43(10):2060–2072.
- 37. Bekkers L, Goede S, Huurne EM, Houten YV, Spithoven R, Leukfeldt ER. Protecting your business against ransomware attacks? Explaining the motivations of entrepreneurs to take future protective measures against cybercrimes using an extended protection motivation theory model. Comput Secur. 2023;127.
- 38. Kokkinos CM, Kipritsi E. The relationship between bullying, victimization, trait emotional intelligence, self-efficacy, and empathy among preadolescents. Soc Psychol Educ. 2012;15(1):41–58.
- 39. Engels C, Kumar K, Philip D. Financial literacy and fraud detection. Eur J Finance. 2020;26:420–442.
- 40. Wei L, Peng M, Wu W. Financial literacy and fraud detection: evidence from China. Int Rev Econ Finance. 2021;76:478–494.
- 41. Porto N, Xiao JJ. Financial literacy overconfidence and financial advice seeking. J Fin Serv Prof. 2016;70(4):78–88.
- 42. Bakar SN, Zakaria NH. The impact of fear and rational appeal scam techniques on individual susceptibility. Baghdad Sci J. 2021;18(2):871–883.
- 43. Modic D, Anderson R, Paloma J. We will make you like our research: the development of a susceptibility-to-persuasion scale. PLoS ONE. 2018;13(3):1–21. pmid:29543845
- 44. Ashton SA, Loannou M. The relationship between gang membership and psychological risks to offending desistance in a sample of adolescent and young adult males. J Gang Res. 2022;29(2):1–25.
- 45. Pyzalski J, Plichta P, Szuster A, Barlińska J. Cyberbullying characteristics and prevention: what can we learn from narratives provided by adolescents and their teachers? Int J Environ Health Public Health. 2022;19:1–26. pmid:36141856
- 46. Nesi J. Transformation of adolescent peer relations in the social media context: part 2 –application to peer group processes and future directions for research. Clin Child Fam Psychol Rev. 2018;21:295–319. pmid:29627906
- 47. Frauenstein ED, Flowerday S. Susceptibility to phishing on social network sites: a personality information processing model. Comput Secur. 2020;94:101862. pmid:32501314
- 48. Judges RA, Gallant SN, Yang L, Kang L. The role of cognition, personality, and trust in fraud victimization in older adults. Front Psychol. 2017;8(3):275–282. pmid:28450847
- 49. Reyes Marc Eric & Morales Belen & Gabriella Javier & Rachel Ng & Ágnes Zsila. (2022). Social Networking Use Across Gender: Its Association with Social Connectedness and Happiness Amidst the COVID-19 Pandemic. Journal of Technology in Behavioral Science. 7. pmid:35761849
- 50. Statista. Main social media activities as part of the shopping journey 2023, by gender. 2024 [cited 2023 Oct 1]. Available from: https://www.statista.com/statistics/1455179/social-media-shopping-activities-by-gender-worldwide/.
- 51. Kamas L, Preston A. Empathy, gender, and prosocial behavior. J Behav Exp Econ. 2021;92:101654.
- 52. Whitty MT. Do you love me? Psychological characteristics of romance scam victims. Cyberpsychol Behav Soc Netw. 2018;21(2):105–9. pmid:28657792
- 53. Anderson KB. Mass-market consumer fraud in the United States: A 2017 update [Staff report]. Federal Trade Commission, Bureau of Economics; 2019. Available from: https://www.ftc.gov/system/files/documents/reports/mass-market-consumerfraud-united-states-2017-update/p105502massmarketconsumerfraud2017report.pdf
- 54. Leukfeldt ER, Holt TJ. Examining the social organization practices of cybercriminals in the Netherlands online and offline. Int J Offender Ther Comp Criminol. 2020;64(5):522–38. pmid:31880178
- 55. Ebner NC, Ellis DM, Lin T, Rocha HA, Yang H, Dommaraju S, et al. Uncovering susceptibility risk to online deception in aging. J Gerontol B Psychol Sci Soc Sci. 2020;75(3):522–33. pmid:29669133
- 56. Farkhondeh H, Harminder S, Jocelyn W. The role of contextualization in users’ vulnerability to phishing attempts. Australas J Inf Syst. 2020;24:1–32.
- 57. Shao J, Du W, Lin T, Li X, Li J, Lei H. Credulity rather than general trust may increase vulnerability to fraud in older adults: a moderated mediation model. J Elder Abuse Negl. 2019;31(2):146–62. pmid:30614418
- 58. Mohd ZN, Mohd AM. Phishing as cyber fraud: the implications and governance. Hong Kong J Soc Sci. 2021;57:120–33.
- 59.
Sugianingrat IA, Suartina W, Indraswarawati SA, Suputra GA. Organizational ethical climate for reducing fraud behavior. In: Proceedings of The First International Conference on Financial Forensics and Fraud; 2019; Bali, Indonesia. EAI; p. 1–9.
- 60.
Dewata E, Farah EH, Jauhari H, Sari Y. Good governance and internal control on the prevention of fraud in the procurement of goods and services for government agencies. In: 5th FIRST T3 2021 International Conference (FIRST-T3 2021); 2021; Atlantis Press. p. 150–54.
- 61. Mamade BK, Dabala DM. Exploring the correlation between cybersecurity awareness, protection measures, and the state of victimhood: the case study of Ambo University’s academic staff. J Cyber Secur Mobil. 2021;10(4):699–724.
- 62. Ribeiro L, Guedes IS, Cardoso CS. Which factors predict susceptibility to phishing? An empirical study. Comput Secur. 2024;136:103558. Available from: https://doi.org/10.1016/j.cose.2023.103558.
- 63. Anderson CL, Agarwal R. Practicing safe computing: a multimethod empirical examination of home computer user security behavioral intentions. MIS Q. 2010;34(3):613–43.
- 64. Choi KS, Cho SJ, Lee JR. Impacts of online risky behaviors and cybersecurity management on cyberbullying and traditional bullying victimization among Korean youth: application of cyber-routine activities theory with latent class analysis. Comput Human Behav. 2019;100:1–10.
- 65. Simeon O, Joseph Ekanem AJ, Ezeh GN. Smart phone security threats and risk mitigation strategies. J Multidiscip Eng Sci Stud. 2022;8(7):4585–97.
- 66.
Saunders M, Lewis P, Thornhill A. Research methods for business students. 5th ed. Pearson Education Limited; 2009.