Enhancing Fraud Detection Performance: The Interplay of Red Flag Awareness, Self-Efficacy, and Professional Skepticism

Abstract

This study aimed to examine the influence of red flag awareness and self-efficacy on the ability to detect fraud through professional skepticism. This study was conducted on commercial banks in Indonesia, due to the high level of fraud that occurs in the banking sector. This study used a quantitative method, and data were obtained from the results of a survey that distributed questionnaires to all internal auditors of commercial banks in Indonesia. The analysis tool used in this study was Smart PLS. The results show that red flag awareness and self-efficacy has an influence on the ability to detect fraud directly or through professional skepticism. This research contributes to bank managers and regulators improvement of the quality of internal auditor training, as well as strengthening the fraud detection system through the development of professional skepticism.

1. Introduction

Fraud refers to unethical or illegal actions in a company’s accounting processes aimed at manipulating financial reports or concealing true information. According to ACFE (2022), fraud is classified into three categories: asset misappropriation, fraudulent statements, and corruption.

In Indonesia, the financial sector has seen a decline in the number of commercial banks, which dropped from 115 in 2018 to 107 in 2021, partly due to fraud. A survey by ACFE Indonesia in 2020 recorded 239 fraud cases, with the financial and banking sectors accounting for 41.4% of these incidents, emphasizing the prevalence of fraud in this industry.

In response, the Financial Services Authority (OJK) issued Regulation No. 39/POJK.03/2019, requiring commercial banks to implement anti-fraud strategies focusing on prevention, detection, investigation, reporting, and monitoring. This regulation aims to enhance fraud prevention and improve reporting obligations. However, despite these measures, fraud cases continue to occur, as seen in the embezzlement case at Bank Kepulauan Riau. These ongoing incidents underline the need for continuous evaluation and strengthening of anti-fraud strategies.

Internal auditors play a crucial role in detecting fraud within organizations. According to ACFE (2022) internal audits provide an independent assessment of internal controls, risk management, and compliance with policies. To support this function, OJK’s Regulation No. 01/POJK.03/2019 requires banks to establish Internal Audit Units (SKAI) with the necessary skills and competencies to conduct thorough, independent audits. Factors influencing auditors’ ability to detect fraud include awareness of red flags, professional skepticism, and self-efficacy, which can enhance the effectiveness of fraud detection.

Auditor skepticism is a key factor in reducing fraud exposure, as it drives auditors to critically assess and investigate potential fraud. According to Laupe et al. (2022), skepticism strengthens investigative audits, while Henrizi et al. (2021) suggest it helps auditors identify information manipulation. Furthermore, red flag awareness plays a critical role in detecting fraud, as it enhances auditors’ ability to recognize patterns or inconsistencies (Craja et al., 2020). Self-efficacy also impacts professional skepticism, with higher self-efficacy leading to greater confidence and proactivity in audits (Svanberg et al., 2019), while auditors with low self-efficacy may be less likely to challenge information thoroughly (Sandhu, 2022).

Prior research has generally discussed fraud in broad terms without delving into the specific characteristics of fraudulent activities. As a result, such studies often lack the contextual depth needed to understand fraud in particular sectors. To fill this gap, the present study focused on the banking industry, a sector that is frequently susceptible to fraud due to its complex and high-stakes financial operations. This study aimed to investigate the factors influencing auditors’ ability to detect fraud within Indonesia’s banking sector. Specifically, it examined the direct effects of red flag understanding and self-efficacy on both professional skepticism and auditors’ fraud detection capabilities. Furthermore, the study explored the mediating role of professional skepticism in the relationship between red flag understanding and self-efficacy with auditors’ ability to uncover fraud. By adopting this approach, the research was expected to yield more nuanced and actionable insights that can enhance auditors’ effectiveness in identifying and addressing fraudulent practices in the banking industry.

2. Literature Review

Agency theory (Jensen & Meckling, 1976) explains the relationship between auditors (agents) and company owners (principals), in which auditors are responsible for ensuring that financial reports are transparent and accurately reflect the company’s true condition. However, in practice, auditors often face conflicts of interest, especially when pressure from management or shareholders influences their objectivity in detecting fraud. Management may attempt to conceal information or pressure auditors not to disclose findings that could harm the company’s reputation. If auditors lack independence and professional skepticism, they might overlook signs of fraud or even become complicit in financial statement manipulation. Therefore, the role of auditors extends beyond merely verifying financial reports; they must also ensure that they can work independently and objectively to prevent and detect fraud, even when facing external pressures.

The Theory of Planned Behavior (TPB) explains that an individual’s behavior, including an auditor’s ability to detect fraud, is influenced by three key factors: attitude, subjective norms, and perceived behavioral control. In the auditing context, red flag awareness represents the auditor’s attitude toward fraud indicators—the higher their awareness, the more likely they are to act in detecting fraud. Additionally, subjective norms, or social pressure from colleagues, superiors, and professional standards, encourage auditors to uphold integrity and perform their duties professionally. The third factor, self-efficacy, or the auditor’s confidence in their own abilities, affects how effectively they investigate and discover fraudulent activities. Auditors with high self-efficacy are more likely to make decisive judgments and take firm action when encountering fraud indicators. Thus, the combination of these three factors determines the likelihood of an auditor successfully detecting and reporting fraud in their audit practices.

Fraud is an unlawful act that is intentionally carried out to gain an advantage through manipulation, especially in financial reporting, and it can be subject to criminal and civil sanctions. Auditors play an important role in detecting fraud to maintain the integrity of financial statements and protect the interests of stakeholders, by relying on technical knowledge, creativity, risk sensitivity, and effective communication. Indicators of the ability to detect fraud include understanding accounting; data analysis; knowledge of information systems; understanding regulations; interview techniques; improvisation capabilities; analytical creativity; adaptation of new technologies; innovative questions; risk analysis; industrial risk awareness; active hearing; effective communication; transparency; and easy-to-understand language.

Professional skepticism is essential for maintaining the integrity and reliability of auditors and accountants, requiring a critical, cautious, and objective mindset to assess information, detect risks or fraud, and make well-founded decisions that add value to organizations and stakeholders. Indicators of professional skepticism include critical and analytical thinking; evidence-based skepticism; deferring conclusions; thorough analysis; flexibility of opinion; high curiosity; trusted source validation; professional emotion control; constructive working relationships; independent decision-making; professional integrity and ethics; report rigor; self-acceptance; strong self-efficacy; and confidence in one’s ability.

Red flags are early indicators that indicate potential problems or risks that require further attention. Although they do not always indicate a definite problem, red flags serve as a warning for individuals or organizations to conduct evaluations and take preventive measures. In the context of fraud, red flags are important signals to detect behavioral or financial irregularities, so a strong internal control system is essential. Understanding red flags includes six dimensions in the fraud hexagon: pressure, opportunity, rationalization, capability, ego/arrogance, and collusion, each of which describes a person’s ability to recognize fraud indicators based on situations, motivations, and risky social interactions.

Self-efficacy is an individual’s belief in their ability to overcome a particular task or challenge, as proposed by Albert Bandura. This belief is influenced by experience, observation, social feedback, and emotional states. Individuals with high self-efficacy tend to be more confident, more resilient to failure, and more motivated to achieve goals. Indicators of self-efficacy include confidence in achieving goals; confidence in facing challenges; belief in ability; situational beliefs; emotional confidence; and confidence in managing failure.

The following is the conceptual framework of this research (Figure 1):

Recognizing and utilizing red flags is crucial in auditing to enhance risk detection, reinforce evidence collection, and maintain a proactive attitude of skepticism. As highlighted by Munteanu et al. (2024), understanding and employing red flags in auditing supports the prevention, identification, and resolution of financial fraud by applying structured anti-fraud methodologies. Gizta (2020) found that red flags assist professionals in pinpointing areas of potential risk, ensuring thorough information assessment, and enabling appropriate preventive or corrective measures when necessary. Furthermore, Pramuki and Agustine (2023) demonstrated that knowledge of red flags heightens professionals’ awareness of potential fraud or non-compliance issues. Based on a review of previous studies, red flags are generally recognized as an important instrument in strengthening professional skepticism, increasing the effectiveness of risk detection, and supporting decision-making based on audit evidence. Although a few studies have reviewed the benefits of red flags in the context of fraud prevention and detection, there is still limited research that specifically examines the causal relationship between the level of awareness of red flags and the increase in auditors’ professional skepticism as a variable that can be measured empirically. Most of the existing literature tends to be descriptive or conceptual, without quantitatively exploring how auditors’ awareness of red flags contributes to professional attitudes in dealing with indications of fraud. Therefore, this study aimed to fill this gap by proposing the following hypotheses:

H1. 

Red flag awareness positively influences professional skepticism.

Individuals with high self-efficacy are more confident in questioning information, relying on their judgment, resisting external pressure, and honing analytical skills that enhance professional skepticism. Research by Tangke et al. (2020) indicated that self-efficacy positively influences professional skepticism. Auditors with strong self-belief can effectively overcome challenges, driven by internal motivation to achieve optimal results. Such auditors actively seek effective solutions, confidently manage tasks under various conditions (Qumaini & Aligarh, 2022), and are willing to request clarification when needed. They trust their ability to analyze information critically, question assumptions, and delve deeper into concepts (Djaddang & Lysandra, 2022). Research shows that self-efficacy supports auditors’ ability to overcome challenges, find solutions, and critique information. Although previous studies have focused on external auditors, the principles are relevant to internal auditors in the banking sector. However, differences in context, such as regulatory and management pressures, may affect internal auditors’ professional skepticism. Therefore, further research is needed to examine the effect of self-efficacy on internal auditor skepticism in the more stringent and stressful banking context. Based on the description above, the researchers developed it the following hypothesis:

H2. 

Self-efficacy positively influences professional skepticism.

Red flag awareness is crucial for auditors in detecting potential fraud in an entity’s financial statements. Asbi Amin (2019) emphasized that recognizing red flags helps auditors identify potential fraud risks. Similarly, Gizta (2020) highlighted the significant influence of red flag awareness on an auditor’s fraud detection capabilities. Edy (2021) noted that red flag awareness enhances auditors’ sensitivity to fraud risks and serves as a critical first step in fraud detection. Gunawan et al. (2022) added that identifying and interpreting red flags enables auditors to perform fraud detection audits effectively. Zakaria (2023) stressed that heightened awareness of warning signs improves auditors’ ability to detect fraud. Red flag awareness plays an important role in increasing auditor sensitivity and fraud detection effectiveness. However, previous findings have not highlighted the role of internal auditors in the banking sector, which has complex regulatory and transaction characteristics. Therefore, further research is needed to examine the relevance of red flag awareness in this context. Based on the description above, the researchers developed the following hypothesis:

H3. 

Red flag awareness positively affects the auditor’s ability to detect fraud.

Self-efficacy, or confidence in one’s abilities, directly influences an auditor’s capability to detect fraud. Auditors with high self-efficacy are more confident in decision-making, motivated to enhance their skills, capable of managing stress, proactive, and view failure as an opportunity for growth. Rustiarini et al. (2021) emphasized that self-efficacy enhances problem-solving abilities, while Haris et al. (2022) highlighted its role in fostering creativity and tackling complex challenges. Bagus Amlayasa and Riasning (2022) found that confidence directly improves auditors’ fraud detection responsibilities, and Al Natour et al. (2023) noted that auditors with high self-efficacy are more motivated, thorough, and effective in planning audits and detecting fraud. Based on the literature, it is emphasized that auditors with high self-efficacy demonstrate more proactive, confident, and strategic behavior in detecting fraud. Consequently, the researchers formulated the following hypothesis:

H4. 

Self-efficacy positively affects the auditor’s ability to detect fraud.

Skepticism plays a crucial role in enhancing auditors’ ability to detect and prevent fraud in financial statements. It encourages attention to suspicious details, promotes detailed examinations, and fosters vigilance against potential fraud (Idawati, 2019). Skepticism also enables professionals to make independent judgments based on evidence rather than blindly trusting information, thereby improving fraud detection and prevention (Agustina et al., 2021). A highly skeptical approach can boost confidence in identifying fraud or errors in financial statements (Wahidahwati & Asyik, 2022). Greater skepticism enhances auditors’ fraud detection capabilities, reducing fraud incidents and improving audit quality (Ningsih et al., 2022). Skeptical auditors are more likely to perform thorough testing of data and transactions (Alizadegan et al., 2023). Based on the description above, the researchers developed the following hypothesis:

H5. 

Professional skepticism positively affects the auditor’s ability to detect fraud.

Red flag awareness enhances auditors’ professional skepticism by highlighting potential risks or issues requiring deeper consideration. Gunawan et al. (2022) stated that recognizing fraud indicators helps auditors focus on high-risk areas and improve fraud detection through skepticism. Gizta (2020) emphasized that professional skepticism reduces bias and strengthens auditors’ ability to identify fraud in financial statements. Similarly, Pramuki and Agustine (2023) noted that red flag awareness aids in identifying fraud-prone areas for closer examination. Asbi Amin (2019) concluded that red flag awareness promotes critical thinking and vigilance during audits. While prior studies have shown a positive link between red flag awareness and auditor professional skepticism, empirical research on the mediating role of skepticism remains limited—particularly within the banking sector, which presents unique red flag indicators. Based on the description above, the researchers developed the following hypothesis:

H6. 

Red flag awareness positively affects the auditor’s ability to detect fraud through professional skepticism.

Self-efficacy, defined as an individual’s belief in their ability to complete a specific task, enhances auditors’ motivation and effort in detecting fraud. Saputra et al. (2021) highlighted that self-efficacy improves auditors’ fraud detection capabilities by boosting motivation and effort, while professional skepticism amplifies this effect by increasing caution and accuracy. Haris et al. (2022) noted that auditors with high self-efficacy are more confident, motivated, proactive, and creative in fraud detection. Bagus Amlayasa and Riasning (2022) emphasized that such auditors are more sensitive to risks and seek additional evidence of fraud. Similarly, Baatwah et al. (2023b) found that auditors with strong self-efficacy exhibit confidence, motivation, resilience, and effective behaviors in fraud detection. Auditors with high self-efficacy tend to be more responsive to risk and proactive in seeking evidence of fraud. In this context, professional skepticism acts as a mediator or amplifier, increasing auditors’ caution and accuracy in detecting indications of fraud. Although the positive relationship between self-efficacy and fraud detection ability has been widely studied, there is still a research gap, especially regarding the mediating or moderating role of professional skepticism. Most previous studies are also descriptive in nature and have not developed an empirical model that integrates the three variables into one conceptual framework. Based on the description above, the researchers developed the following hypothesis:

H7. 

Self-efficacy positively affects the auditor’s ability to detect fraud through professional skepticism.

3. Materials and Methods

This study employed a quantitative survey design to investigate the relationship between red flag awareness, self-efficacy, professional skepticism, and fraud detection capabilities in Indonesia’s banking industry. The eligible population of this study included 639 auditors registered in various commercial banks registered in Indonesia. This study used the Slovin formula for sampling, which helps determine the sample size from a large population. The error tolerance was set at 0.05 (5%), and the result was a sample of 246 with random selection. For the data collection, the researchers provided a questionnaire that could be filled in truthfully and can be accounted for by the respondents, and this questionnaire was tested on 50 people.

The data analysis was conducted using the Partial Least Square (PLS) method with Smart PLS software. PLS, a technique in Structural Equation Modeling (SEM), is preferred due to its advantages over other SEM methods. Its flexibility makes it particularly useful for research that integrates theory and data, including path analysis with latent variables. PLS is also notable for its ability to operate without relying on many assumptions. The measurement of variables in this study can be seen in the following (

Table 1. Indicator table.

Variable	Dimension	Indicator	Questionnaire Item
Red flag awareness (X1)	Fraud hexagon	Pressure	X1.1
		Chance	X1.2
		Rationalization	X1.3
		Capabilities	X1.4
		Ego	X1.5
		Collusion	X1.6
Self-efficacy (X2)	Self-efficacy Level:	Confidence in achieving goals	X2.1
		Confidence in facing challenges	X2.2
	Generality	Belief in ability	X2.3
		Situational beliefs	X2.4
	Strength	Emotional confidence	X2.5
		Confidence in managing failure	X2.6
Professional skepticsm (Y1)	Questioning mind	Critical and analytical thinking	Y1.1
		Evidence-based skepticism	Y1.2
	Suspension of judgment	Deferring conclusions	Y1.3
		Thorough analysis	Y1.4
		Flexibility of opinion	Y1.5
	Search for knowledge	High curiosity	Y1.6
		Trusted source validation	Y1.7
	Interpersonal understanding	Professional emotion control	Y1.8
		Constructive working relationships	Y1.9
	Autonomy	Independent decision-making	Y1.10
		Professional integrity and ethics	Y1.11
		Report rigor	Y1.12
	Self-esteem	Self-acceptance	Y1.13
		Strong self-efficacy	Y1.14
		Confidence in ability	Y1.15
Fraud detection capability (Y2)	Technical knowledge	Understanding accounting	Y2.1
		Data analysis	Y2.2
		Knowledge of information systems	Y2.3
		Understanding regulations	Y2.4
		Interview techniques	Y2.5
	Creativeness	Improvisation capabilities	Y2.6
		Analytical creativity	Y2.7
		Adaptation of new technologies	Y2.8
		Innovative questions	Y2.9
	Sensitivity to risk	Risk analysis	Y2.10
		Industrial risk awareness	Y2.11
	Effective communication	Active hearing	Y2.12
		Effective communication	Y2.13
		Transparency	Y2.14
		Easy-to-understand language	Y2.15

):

4. Results

This study obtained a sample of 312 internal bank auditors, surpassing the minimum required sample size of 246 using the Slovin formula. Among the 312 respondents, 196 were female and 116 were male. Regarding age, 134 respondents were aged 25–35 years, 62 were aged 36–45 years, 59 were aged 46–55 years, and 57 were over 55. In terms of the type of bank, 173 respondents worked in Private Banks (both Foreign Exchange and Non-Foreign Exchange), 95 in Regional Development Banks, 6 in Foreign and Mixed Banks, and 38 in Bank Persero. Concerning work experience, 134 respondents had 1–5 years of experience, 121 had 6–10 years, and 57 had more than 10 years.

Validity and reliability testing was conducted to ensure that the indicators in the questionnaire measured the construct accurately and consistently. Validity testing includes outer loading (ideal >0.70) and AVE (>0.50) for convergent validity, while reliability is evaluated through Composite Reliability and Cronbach’s Alpha, each with an ideal value >0.70.

Outer loading is a value that shows how strongly the indicator reflects the latent construct in the reflective model. This value describes the correlation between the indicator and the construction it measures. The higher the outer loading (ideally ≥0.70), the better the indicator is in representing the construct. The results of the outer loading analysis showed that the red flag awareness construct (X1) had six indicators, and four indicators (X1_1 = 0.823; X1_4 = 0.830; X1_5 = 0.851; X1_6 = 0.877) showed very good validity (loading > 0.7), while the other two indicators (X1_2 = 0.540 and X1_3 = 0.559) were below the minimum threshold of 0.6, so they needed to be considered for revision or elimination. The self-efficacy construct (X2) consisted of six indicators with quite good results; four indicators (X2_1 = 0.697; X2_3 = 0.704; X2_4 = 0.808; X2_5 = 0.765) had values approaching or exceeding 0.7, although the other two (X2_2 = 0.615 and X2_6 = 0.645) showed lower contributions and could be considered for substance strengthening. In the professional skepticism (Y1) construct consisting of 15 indicators, the majority of indicators showed supportive results, such as Y1.1 (0.759), Y1.4 (0.730), Y1.6 (0.715), Y1.7 (0.739), Y1.11 (0.712), Y1.12 (0.777), Y1.13 (0.716), and Y1.14 (0.724), all of which had values above 0.7, indicating good measurement quality. However, there were several indicators that had loading values below 0.7, namely Y1.2 (0.620), Y1.3 (0.611), Y1.5 (0.515), Y1.8 (0.586), and Y1.15 (0.537), which indicated a low contribution to the construction and could reduce the validity of the overall measurement. Finally, in the fraud detection capability (Y2) construct, which also had 15 indicators, there were a few indicators that showed high contributions, such as Y2.2 (0.707), Y2.3 (0.762), Y2.6 (0.751), Y2.8 (0.732), Y2.10 (0.750), and Y2.12 (0.699). However, there were also a number of indicators that had low and less than ideal loading values, such as Y2.1 (0.568), Y2.4 (0.520), Y2.9 (0.634), Y2.11 (0.535), Y2.13 (0.570), Y2.14 (0.425), and Y2.15 (0.581), which indicated that these indicators should be re-evaluated to ensure that only valid and relevant indicators were used in measuring fraud detection capability. The indicators with values below 0.7 were removed due to their less significant contribution to the construct analyzed. resulting in the following research model (

Table 2. Outer loading results using Smart PLS.

Construct	Indicator	Outer Loading
Red flag awareness	X1_1	0.913	Valid
	X1_4	0.917	Valid
	X1_5	0.916	Valid
	X1_6	0.774	Valid
Self-efficacy	X2_1	0.793	Valid
	X2_3	0.706	Valid
	X2_4	0.795	Valid
	X2_5	0.804	Valid
Fraud detection capability	Y2.10	0.842	Valid
	Y2.12	0.807	Valid
	Y2.2	0.733	Valid
	Y2.3	0.817	Valid
	Y2.6	0.796	Valid
	Y2.8	0.734	Valid
Professional skepticism	Y1.1	0.801	Valid
	Y1.11	0.765	Valid
	Y1.12	0.726	Valid
	Y1.13	0.771	Valid
	Y1.14	0.772	Valid
	Y1.4	0.762	Valid
	Y1.6	0.729	Valid
	Y1.7	0.772	Valid

):

R Square (R²) is a statistical measure that represents the proportion of variance in the dependent variable that is explained by the independent variables in a regression model. The following are the R Square results from this research (

Table 3. R Square results.

	R Square	R Square Adjusted
Fraud detection capability (Y2)	0.895	0.894
Professional skepticism (Y1)	0.811	0.810

):

The results show that the independent variables in the model had a strong influence on both fraud detection capability (Y2) and professional skepticism (Y1). The R Square value for fraud detection capability was 0.895, indicating that 89.5% of its variance was explained by the independent variables, demonstrating a very strong model fit. Similarly, professional skepticism had an R Square value of 0.811, meaning 81.1% of its variance was accounted for by the model, reflecting a strong relationship with the predictors. The Adjusted R Square values, which were only slightly lower, suggested that the model remained highly reliable, with minimal risk of overfitting. Overall, these findings confirm that the independent variables significantly contributed to explaining variations in both fraud detection capability and professional skepticism.

Construct reliability and validity are essential in research to ensure that the measurement model is both consistent and accurate in representing the underlying construction. Reliability assesses the internal consistency of the measurement items, often measured using Composite Reliability (CR) and Cronbach’s Alpha. The following are the construct reliability and validity results from this research (

Table 4. Construct reliability and validity.

	Cronbach’s Alpha	Rho_A	Composite Reliability	AVE
FDC (Y2)	0.878	0.881	0.908	0.623
RA (X1)	0.903	0.902	0.933	0.778
SE (X2)	0.780	0.792	0.858	0.602
PS (Y1)	0.897	0.899	0.917	0.581

Note: RA: red flag awareness; SE: self-efficacy; PS: professional skepticism; FDC: fraud detection capability.

):

The table presents the reliability and validity statistics for four constructs: FDC (Y2), RA (X1), SE (X2), and PS (Y1). The Cronbach’s Alpha and Composite Reliability (CR) values for all constructs exceeded 0.7, indicating strong internal consistency. RA (X1) had the highest reliability (Cronbach’s Alpha = 0.903, CR = 0.933), while SE (X2) had the lowest but still acceptable reliability (Cronbach’s Alpha = 0.780, CR = 0.858). The Average Variance Extracted (AVE) values were above the recommended threshold of 0.5, confirming adequate convergent validity, with RA (X1) having the highest AVE (0.778) and PS (Y1) the lowest (0.581). Overall, the constructs demonstrated strong reliability and sufficient validity for further analysis.

The following are the path coefficient values and p-values for testing the significance of direct effects (

Table 5. Direct effect path coefficients (mean, STDEV, T-values, p-values).

	Standard Deviation	T-Statistics	p-Values
Understanding red flags (X1) -> fraud detection capability (Y2)	0.037	2.895	0.002
Understanding red flags (X1) -> professional skepticism (Y1)	0.051	6.508	0.000
Self-efficacy (X2) -> fraud detection capability (Y2)	0.028	2.302	0.011
Self-efficacy (X2) -> professional skepticism (Y1)	0.042	7.945	0.000
Professional skepticism (Y1) -> fraud detection capability (Y2)	0.052	13.152	0.000

):

The analysis of various relationships showed significant results across different hypotheses. The relationship between understanding red flags (X1) and professional skepticism (Y1), with a standard deviation of 0.051, a t-statistic of 6.508, and a p-value of 0.000, was significant, indicating that H1 was accepted. The relationship between self-efficacy (X2) and professional skepticism (Y1), with a standard deviation of 0.042, a t-statistic of 7.945, and a p-value of 0.000, was also very significant, meaning H2 was accepted. The relationship between understanding red flags (X1) and fraud detection ability (Y2), with a standard deviation of 0.037, a t-statistic of 2.895, and a p-value of 0.002, showed a significant relationship, supporting the acceptance of H3. The relationship between self-efficacy (X2) and fraud detection ability (Y2), with a standard deviation of 0.028, a t-statistic of 2.302, and a p-value of 0.011, was significant, indicating that H4 was accepted. Finally, the relationship between professional skepticism (Y1) and fraud detection ability (Y2), with a standard deviation of 0.052, a t-statistic of 13.152, and a p-value of 0.000, was very significant, confirming the acceptance of H5.

The following are the path coefficient values and p-values for testing the significance of indirect effects: (

Table 6. Indirect effect path coefficients (mean, STDEV, T-values, p-values).

	Standard Deviation (STDEV)	T statistics (|O/STDEV|)	p Values
Red flag awareness (X1) -> professional skepticism (Y1) -> Fraud detection capability (Y2)	0.039	5.844	0.000
Self-efficacy (X2) -> professional skepticism (Y1) -> fraud detection capability (Y2)	0.030	7.473	0.000

).

The analysis of the paths showed significant relationships between various factors influencing fraud detection ability. For the path from red flag awareness (X1) to professional skepticism (Y1) and fraud detection ability (Y2), the STDEV was 0.039, with a t-statistic of 5.844 and a p-value of 0.000, indicating that red flag awareness significantly impacts fraud detection through professional skepticism and that H6 was accepted. Rebarding self-efficacy (X2) to professional skepticism (Y1) and fraud detection ability (Y2), the path showed an STDEV of 0.030 and a t-statistic of 7.473, with a p-value of 0.000, demonstrating that self-efficacy has a significant indirect relationship with fraud detection ability through professional skepticism and that H7 was accepted.

5. Discussion

This study demonstrates that understanding red flags enhances the professional skepticism of bank internal auditors, especially in detecting early signs of fraud. Auditors with a strong understanding of red flags are more alert and tend to investigate suspicious actions more thoroughly. Studies suggest that auditors who recognize red flags can quickly identify early fraud indicators and take preventive action (Sánchez-Aguayo et al., 2021) Fraud indicators can point to risks in financial statements, necessitating auditors to implement additional procedures and verification (Craja et al., 2020; Yuan et al., 2023). Internal auditors should be vigilant about repeated errors and conduct thorough reviews to detect fraud (Kumar et al., 2022). Identifying weaknesses in internal controls and digging for evidence are crucial for detecting fraud (Wahidahwati & Asyik, 2022). This is illustrated by several fraud cases in the banking sector in Indonesia, where internal auditors failed to detect early warning signs, such as unusual fund transfers, loan disbursements without proper collateral verification, and frequent overrides of internal control procedures. In these cases, significant financial losses occurred due to the auditors’ inability to recognize and respond to red flags in a timely manner. The frauds were often discovered only after customer complaints or external audits, indicating a reactive rather than proactive internal audit function. These real-world occurrences highlight the critical importance of red flag awareness in enabling auditors to act promptly and prevent fraud escalation. Agency theory emphasizes that auditors, by understanding fraud signs, investigate management’s information more thoroughly, thus helping the company reduce fraud risks (Panda & Leepsa, 2017). Internal auditors play a key role in identifying conflicts of interest, ensuring transparency and enhancing corporate governance (Bona-Sánchez et al., 2024; Hardies et al., 2024).

This study demonstrates that self-efficacy enhances the skepticism of bank internal auditors. Self-efficacy plays a key role in shaping auditors’ professional skepticism, affecting their judgment and decision-making. Auditors with high self-efficacy are less likely to rush to conclusions or make hasty audit statements. They tend to be more thorough and cautious in their audits, ultimately improving audit quality (Erlina & Muda, 2018). Auditors confident in their abilities act more skeptically, believing they are competent to identify and address discrepancies affecting audit quality, as supported by Hasson and Knechel (2019). High self-efficacy increases an auditor’s ability to evaluate evidence and make informed decisions (Baatwah et al., 2023a). Auditors with high self-efficacy can maintain skepticism under pressure, managing stress effectively and remaining calm and logical in decision-making, even when facing external challenges (Ghani et al., 2022; Kelly & Larres, 2023). Auditors with high self-efficacy are better equipped to handle complex and stressful situations, question assumptions, and make informed judgments (Djaddang & Lysandra, 2022; Haris et al., 2022; Qumaini & Aligarh, 2022).

This study demonstrates that understanding red flags enhances a bank’s internal auditors’ ability to detect fraud. A deeper knowledge of red flags allows auditors to more effectively identify high-risk areas for fraud. Afriyie et al. (2022) note that understanding fraud indicators enables auditors to examine financial statements and internal controls, boosting their fraud detection capabilities. Ziorklui et al. (2024) and Munteanu et al. (2024) highlight that auditors familiar with red flags are more likely to detect irregularities. Baader and Krcmar (2018) emphasize that recognizing red flags helps auditors identify potential fraud schemes and recommend preventive measures. Edy (2021) and Gizta (2020) argue that a solid understanding of red flags helps auditors be more sensitive to fraud risks and serves as a first step in detecting fraud. Gunawan et al. (2022) note that recognizing red flags leads to more efficient fraud detection. Zakaria (2023) asserts that greater awareness improves an auditor’s ability to identify fraud. According to agency theory, a strong understanding of red flags helps auditors protect the principal’s interests and reduce agency risk. Akinbowale et al. (2023) and Craja et al. (2020) argue that without this understanding, auditors may miss crucial fraud indicators, which could lead to undetected fraudulent actions that harm owners or shareholders.

Those with high self-efficacy trust their ability to detect fraud, even in complex or stressful situations. This belief positively impacts their attitude towards fraud detection tasks, fostering commitment. Auditors with high self-efficacy make decisions aligned with professional standards, supporting fraud detection effectiveness. High self-efficacy also boosts motivation, effort, and resilience, improving performance in fraud identification. Rustiarini et al. (2021) indicates that self-efficacy enhances problem-solving skills. Haris et al. (2022) find that confident auditors are more creative and better equipped to tackle complex challenges. Bagus Amlayasa and Riasning (2022) suggest that auditors’ responsibility in fraud detection is directly improved by confidence. Al Natour et al. (2023) report that high self-efficacy increases motivation, leads to more effective audit planning, and makes auditors more thorough and responsive in detecting fraud. According to Jonathan Muterera (2024), auditor confidence improves performance, decision-making quality, and fraud detection efficiency in dynamic conditions.

This study shows that professional skepticism enhances the ability of internal bank auditors to detect fraud. Auditors with high skepticism critically assess information, seeking additional evidence to validate its accuracy. They question the validity and reliability of information, helping identify discrepancies or anomalies in financial data. This approach also makes auditors more cautious about possible manipulation or concealment of information, ensuring that decisions are based on facts. Ali Al-Kake et al. (2024) and Gajewski et al. (2024) confirm that skeptical auditors critically evaluate information, while Varma et al. (2023) highlight that skepticism leads auditors to question rather than accept information as absolute truth.

Understanding red flags makes auditors more vigilant and critical of the information they receive, especially when it suggests irregularities or improper behavior. Auditors who are knowledgeable about red flags can identify suspicious patterns and data more accurately, leading to a more thorough analysis of the evidence. This critical approach helps auditors go beyond assumptions and apply comprehensive testing to objectively uncover facts. Ultimately, this improves fraud detection and strengthens the overall integrity of the audit process. Pramuki and Agustine (2023) highlight that understanding red flags is crucial for auditors in detecting fraud. Gunawan et al. (2022) also note that auditors, by focusing on high-risk areas with professional skepticism, improve their fraud detection abilities. Gizta (2020) emphasizes that maintaining skepticism helps auditors avoid bias and manipulation, enhancing their ability to identify fraud in financial statements.

High self-efficacy combined with strong skepticism enables auditors to better identify fraud. Auditors with high self-efficacy can maintain professionalism and skepticism under pressure, allowing them to perform their duties effectively. This self-assurance encourages them to ask critical questions and challenge information without fear of negative consequences, which is crucial for fraud detection and maintaining audit integrity. Self-efficacy increases motivation and effort, while professional skepticism boosts the auditor’s prudence and accuracy in detecting fraud (Saputra et al., 2021). Auditors with high self-efficacy are more confident in decision-making, motivated to improve, and proactive in fraud detection (Haris et al., 2022). They are also more risk-sensitive and eager to seek additional evidence of fraud (Bagus Amlayasa & Riasning, 2022). Baatwah et al. (2023a) indicate that auditors with high self-efficacy exhibit strong confidence in their fraud detection abilities, positively influencing their attitudes, behaviors, motivation, and resilience.

6. Conclusions

This study emphasizes the importance of red flag awareness and professional skepticism in enhancing the ability of internal auditors at banks to detect fraud. A solid understanding of red flags helps auditors identify high-risk areas, such as pressure, ego, collusion, and weaknesses in internal controls, signaling potentially fraudulent activities. Professional skepticism strengthens this capability by encouraging auditors to critically evaluate information, seek more evidence, and stay alert to anomalies in financial data. These traits enable thorough analysis, prevent premature conclusions, and maintain audit integrity, protecting stakeholders’ interests and minimizing fraud risks.

This research also highlights the role of self-efficacy in improving auditors’ professional skepticism and fraud detection abilities. Auditors with high self-efficacy are better equipped to handle complex, high-pressure situations, make informed judgments, and remain resilient under stress. This confidence boosts motivation, effort, and diligence, driving auditors to challenge assumptions and focus on key fraud indicators. Self-efficacy not only improves fraud detection but also promotes a proactive, creative, and adaptable mindset. When combined with professional skepticism, self-efficacy ensures that auditors remain objective and thorough, enhancing audit quality and fostering trust in the profession.

The results of this study indicate that understanding red flags and high self-efficacy significantly increases auditors’ professional skepticism. Although time pressure can reduce audit quality, experienced auditors are able to manage it well to maintain professional skepticism and improve fraud detection capabilities. This study develops applied theory by introducing the concept of “Integrated Fraud Detection”, which emphasizes that the effectiveness of fraud detection is influenced not only by technical capabilities, but also by psychological factors, work environment, and auditor experience. The research findings indicate that professional skepticism exerts the strongest influence compared to other variables. Therefore, organizations or institutions should foster a culture of professional skepticism in the audit process by implementing policies that promote critical evaluation of financial statements and transactions. To enhance auditors’ understanding of red flags and strengthen their self-efficacy, organizations must also invest in improving auditor competence through initiatives such as obtaining professional certifications, such as the Certified Fraud Examiner (CFE) from ACFE International, adopting effective time management strategies, and cultivating a supportive work environment that encourages the development of self-efficacy through mentoring and constructive feedback.

This study focused on red flag awareness, professional skepticism, and self-efficacy, without considering the influence of other external factors, such as organizational culture, managerial support, or regulatory pressures, which may also affect auditors’ ability to detect fraud. The use of quantitative methods limited the in-depth exploration of the reasons or mechanisms behind the influence of red flag awareness, professional skepticism, and self-efficacy. Qualitative approaches or case studies could provide deeper insights.