The Influence of Personality Traits and Domain Knowledge on the Quality of Decision-Making in Engineering Design

Abstract

In engineering design, the decision-making process holds significant importance as it plays an important role in determining the outcomes of a task. The decision-making process is notably influenced by various factors, with particular focus on the personality traits and information available. The purpose of this study is to comprehensively investigate the effects of these factors on quality and confidence in decision-making within the context of engineering design. To achieve this objective, we utilized a simulated design environment that can capture decision-making information. The analysis of personality traits was carried out utilizing the complete Big Five model, while the estimate of the structural equation model was executed by employing partial least squares structural equation modeling (PLS-SEM) and a machine learning model for quality estimation. The available empirical research indicates that individuals who have a lower degree of extraversion and agreeableness, and higher levels of conscientiousness and openness, are more likely to make decisions of higher quality. These characteristics have been found to have no significant effect on the levels of confidence during the process of making decisions. Furthermore, it was found that the trait of neuroticism has a negative impact on the quality of decision-making but does not have a significant impact on decision-making confidence. The noticeable finding was the strong impact of test-assessed knowledge on decision quality and confidence, in contrast to the lack of significant effect of self-assessed knowledge. This highlights the importance of carefully aligning tasks with individual personality traits in organizations working in the engineering design sector and prioritizing factual demonstrated knowledge rather than subjective self-assessment when assigning decision-making positions to individuals. These findings highlight the importance of considering personality traits and domain knowledge in educational and professional settings to enhance decision-making quality and confidence among engineering students, potentially informing targeted training and assessment practices.

1. Introduction

Decision-making is an important element in today’s engineering context within the engineering design process. It is not a decision made in isolation by an individual but a complicated procedure dependent on several variables and other contextual elements [1]. Computer-based tools have led to a significant improvement in making effective decisions due to the emergence of computational technology. These computational tools can be further categorized into three major groups of knowledge-based engineering, workflow management systems, and collaborative decision support systems. Each of these classes has an important contribution it can make towards the enhancement of the quality of decisions, giving the complexity and importance of the process of decision-making in engineering design [2,3].

The study of behavior among individuals in operations and engineering design, in mechanical, systems, and industrial engineering, is an area that is growing but has not received much attention in research, especially in complicated decision-making contexts. The importance of human factors in engineering for enhancing production quality is evident, highlighting the substantial contributions made by individuals in design processes [3]. Acquiring a thorough understanding of human behavior is considered vital for improving operational efficiency and gaining insights into how behavior influences the quality and reliability of decisions across various engineering disciplines [4].

The study of engineering decision-making has advanced by investigating the complexities of human behavior, particularly by examining perceptual, cognitive, and decision-making systems within the engineering domain [5]. Several studies have identified a relationship between personality traits, as described by the Big Five model, and knowledge specific to various fields. Research consistently demonstrates the impact of behavioral traits on decision-making outcomes, with a significant influence on the quality and confidence associated with these decisions [6]. Decision quality can be defined as a comprehensive evaluation that includes the procedural aspects of decision-making and ensures positive outcomes that affect an organization’s performance. In contrast, confidence in decision-making can be described as the degree of information that a decision-maker knows rather than achieving the desired objective [7].

Extensive research has been conducted in the domain of engineering design, with particular emphasis on the examination of decision-making procedures. A framework for facilitating robust decision-making in the context of engineering design emphasizes the important role of rigorous methodology throughout the design process [8,9], and decision-making plays a crucial role in the field of engineering design [10]. An analytical framework can be utilized to find robust methods in the context of decision-making [11,12,13]. Of these methodological advancements, there is a notable lack of research examining the influence of personality factors on the quality of decision-making within the context of engineering design. The assessment of quality in engineering design has been influenced by a range of methodological techniques. The significance of design science in the context of decision support systems has been expanded upon, providing a framework for evaluating and assessing these systems [14]. Another approach in the form of a weighted decision matrix for effectively monitoring and documenting design decisions inside service compositions was utilized [15]. There has been significant emphasis on the importance of enhancing the quality and relevancy of design-science research within the field of decision-support systems [16].

Although these studies have made substantial contributions to the discipline, their dependency on the evaluation of a single author for quality assessment is noticeable [6]. In contrast to previous studies, the current study used a comprehensive methodology that combines the assessment of a single author with expert evaluations, resulting in a more rigorous evaluation of decision quality. Moreover, the utilization of machine learning techniques, as described in [17] establishes a robust framework for assessing the quality of decisions in the field of engineering design. This approach will improve the credibility of these assessments using a single evaluator; a Naive Bayes machine learning model is utilized to make predictions about the Quality of Decisions (QODs) by analyzing textual decision statements.

The current study utilizes a comprehensive sampling strategy to further improve the generalizability of its results, a methodological approach that is consistent with modern advances in research design [18]. The paper entitled “Sampling in Software Engineering Research: A Critical Review and Guidelines” offers a comprehensive evaluation of the current state of sampling in research of high quality within the field of software engineering. Additionally, the article provides suggestions that are relevant and applicable beyond the domain of software engineering [19]. The study applies purposeful sampling procedures [20] to assure the relevance and generalizability of its findings across different research settings. Moreover, a comprehensive overview of the current state of sampling in the recent literature [21], which offers supplementary recommendations for enhancing sample methodologies, utilizes qualitative research methods to strengthen its research design [22]. The implementation of a comprehensive approach to sampling in this study aligns with the investigation of the influence of personality factors and domain-specific knowledge on decision quality. Moreover, this approach is consistent with current standards in research design.

This study builds upon prior research that has examined the influence of personality traits and decision-making on decision quality in various domains [23], focusing on the impact of decision-making and problem-solving quality in engineering design contexts [24], and how certain personality traits may play a role in this process [25]. Our research focuses on the integration of decision-making processes and personality traits, highlighting the significance of decision-makers internal features as strong indicators of decision quality within the domain of engineering design.

This paper’s research gap is the lack of examination of how people behave during the engineering design process, particularly when provided with difficult choices. Even though it is acknowledged that decision-making plays a crucial role in enhancing organizational performance, decision-making is typically carried out by individuals using their time and knowledge. This article emphasizes the need for more research that examines the “people dimension” of engineering design and investigates how dispositional features like personality traits and domain expertise affect decision-making confidence and quality. Organizations can perform more effectively and make better decisions by having a deeper grasp of these behavioral concerns. The significance of age and how it affects decision-making is a significant component that has not been evaluated in this study. Due to variations in individual and professional exposure, age differences can have a substantial influence on confidence and decision quality. By assessing the impact of age on decision-making, future studies can close this knowledge gap and offer a more thorough grasp of the variables affecting engineering decisions. Furthermore, previous research frequently uses homogeneous participant groups, which limits the findings’ generalizability in a variety of scholarly and cultural contexts.

The methodology employed in this study addresses these challenges and demonstrates its feasibility and importance in advancing the field. By combining subjective evaluations with machine learning validation (e.g., Naive Bayes model) and expert assessments, the study ensures robust and reliable results, reducing biases associated with single-author evaluations. The inclusion of participants from diverse academic and cultural backgrounds further enhances the generalizability of the findings, making the study relevant to a broader range of engineering contexts. Additionally, this research integrates the assessment of the Big Five personality traits alongside both self-assessed and test-assessed knowledge, providing a nuanced understanding of how these factors influence decision-making quality and confidence. This dual focus enables a more precise evaluation of the relationship between dispositional characteristics, knowledge, and decision outcomes.

This study’s practical implications make it particularly valuable. By identifying actionable strategies, such as tailoring educational interventions to align with individual personality traits and emphasizing test-assessed knowledge, this research offers concrete pathways for improving decision-making quality and confidence. Furthermore, the use of simulation games creates a realistic decision-making environment, enhancing the ecological validity of the findings and ensuring their applicability to real-world engineering challenges. In addressing critical gaps in the literature, this study not only contributes to theoretical advancements in the field but also provides practical insights that can be utilized in academic curricula and organizational training programs, thereby establishing its significance and relevance.

The primary objective of our research is “to quantify the impact of designers’ domain knowledge on quality of decision within the context of engineering design”. Furthermore, as a supplementary study, we will also assess confidence in decisions as a secondary objective.

This study seeks to explore the relationship between personality traits, domain knowledge, and decision-making outcomes in engineering contexts. Specifically, it addresses the following research questions:

• How do individual personality traits influence the quality and confidence of decisions made by engineering students?
• What is the impact of domain knowledge, both self-assessed and test-assessed, on decision-making quality and confidence?
• How do personality traits and domain knowledge interact to affect decision-making outcomes in engineering tasks?

This paper is organized in the following manner. The introductory section of this study emphasizes the importance of personality traits and the competence of decision-makers in the domain of engineering design decision-making. Afterward, the research model and hypotheses will be presented. Following that, the data sources employed, and the methodology utilized for data analysis are outlined. The following part summarizes the results of the effects of various personality traits, as well as self-reported and test-assessed knowledge, in the field of engineering design. The discussion section offers a comprehensive analysis of the theoretical improvements and practical implications. The study concludes by proposing potential avenues for future research.

A distinguishing feature of this study is its comprehensive approach to examining personality traits in the context of engineering design. Unlike previous research that primarily addressed personality traits in general contexts, this study specifically evaluates their impact on decision-making quality and confidence within the complex and task-specific environment of engineering design. By integrating the Big Five personality model with both self-assessed and test-assessed knowledge, this research offers a deeper and more precise understanding of how dispositional factors and knowledge interact to influence decision outcomes.

Furthermore, the inclusion of machine learning validation (e.g., Naive Bayes model) alongside subjective and expert evaluations enhances the robustness of decision quality assessments, overcoming the limitations of earlier studies that relied heavily on single-evaluator subjectivity. The study’s focus on diverse academic and cultural participants, expanding the generalizability of our findings and addressing the homogeneity of samples in prior research.

2. Literature Review and Development of Hypotheses

2.1. Decision-Making in Engineering Design

Engineering design is an analytical process that reduces the time and resources required for final production while simultaneously improving the quality of the end product [26]. The quality of judgments made during the engineering process has a major impact on how well engineering products turn out [27]. Most of the time, manufacturing-specific information is insufficient. The cost-effective design of manufacturing-oriented concepts is a challenging task. There is no assessment of manufacturing capability. Early phases of design face obstacles due to the unknown complexity of cost factors [1]. The most important aspect of handling the design process is the decisions taken and the information incorporated in the decision-making process. This significance leads to the development of the idea of robust decision-making [8]. Decision-making in engineering design can be used by engineers for making judgments about product design, educators or students working on assignments, and researchers looking into new possibilities for the basis and principles of decision-making in product design, providing instances of successful decision-based design implementation. The latest developments in decision-based design theory and application arise from careful observations on demand modeling, distributed design, preferences, uncertainty, and validation [10].

The emphasis on individual personality qualities is a way to predict job performance and is one of the most notable developments in corporate organizations today. To better understand the relationship between personality traits and important success behaviors in project design services for the engineering and architectural professions, various factors are being studied [24]. According to research, people choose their careers and themselves in part based on their personalities, which has an impact on how well they fit in with their surroundings. The impact of personality congruence in occupations on job happiness is insufficiently understood, despite research showing that personality congruence between the individual and the environment is critical to job satisfaction [28]. Employers request recruiters to select candidates who vary in seven areas: future salary range, professional competence, and the Big Five personality qualities. We discover that every personality attribute influences the likelihood of hiring a job applicant, with agreeableness and conscientiousness having the most advantages. Recruiters typically prefer applicants who are more open, conscientious, and agreeable to analytical positions, while favoring those who are open, extroverted, and agreeable to collaborative roles [29].

The decision-making process in engineering design is influenced by a wide range of factors, including both internal and external elements that impact the decision-makers. Internally, cognitive qualities, such as expertise that is specific to a particular domain [30] and individual personality traits [31] play important roles. Furthermore, the quality of decision-making is influenced by cognitive biases, and it has been determined that cognitive biases—systematic patterns of divergence from norms or rationality in judgment—are prevalent factors influencing decision-making in a variety of contexts. A thorough research investigation should examine the cognitive biases that are pertinent to decision-making in the domains of machine learning, systems engineering, industrial engineering, and mechanical engineering. Confirmation bias, anchoring bias, and availability heuristic are a few examples [32]. The study becomes even more complex when considering the personal growth component of experience. People’s decision-making processes change as they gain greater expertise in their respective industries. The influence of experience on reducing or enhancing cognitive biases can be addressed by longitudinal research that monitors the quality of decision-making during a person’s career development [33]. The complex interactions between external influences are highlighted by the fields of industrial engineering, systems engineering, and machine learning. In these domains, decision-making scenarios are dynamically shaped by regulatory frameworks, political settings, sector-specific issues, and economic conditions. This research seeks to comprehend the influence of regulatory standards and geopolitical disruptions, investigate sector-specific difficulties impacting decision dynamics, and uncover the effects of economic volatility on strategies for adaptation and risk management. Theoretically, case studies, longitudinal research, and empirical investigations are suggested to represent the complex interplay between outside influences and decision-making processes. In the end, this thorough study aids in the creation of frameworks that improve decision-making effectiveness and resilience in the face of the changing external environments that these engineering domains are known for [34], shifting the decision-making scenarios.

Within the domain of engineering design, the act of decision-making assumes a fundamental role, supported by a well-organized process that involves gathering relevant information about various options, variables, and constraints that are inherent to the design context [35]. The significant impact of domain knowledge on design results and allocation of resources in design procedures, where a rigorous utilization of mathematical concepts and decision theory facilitates the advancement of computational approaches for engineering design [36]. The concept of a relevant knowledge system (RKS) is a fundamental aspect of engineering decision-making. It emphasizes the significance of understanding nature, creation, and evaluation of knowledge that is necessary for making well-informed judgments. This concept is particularly demonstrated in the field of structural engineering [27]. The significance of decision-making quality is emphasized as a critical element, with the recognition of effective decision support to enhance a designer’s decision-making capacity, thereby improving the design process and facilitating the creation of high-quality designs. Furthermore, existing research emphasizes the importance of knowledge modeling in the context of design decision-making. It supports for inclusion of more layers of conceptualization in computer-aided design (CAD) systems, as well as the integration of evaluation and validation activities within design processes [26]. The utilization of decision-based design principles strengthens the narrative by providing insights into the management of preferences, uncertainty, design, estimation, simulation, and validation. Thus, it provides a comprehensive analysis of the decision-making context in engineering design [10,26].

2.2. Role of Personality Traits

The exploration of how personality traits impact decision-making within the context of engineering designs is a complex and multidimensional field of study. Since personality traits are innate and consistent in terms of patterns of behavior, they have been shown to have a major impact on several cognitive and decision-making processes in humans. In the field of engineering, personality traits have a particularly strong influence since complex problem-solving and critical decision-making are crucial [37]. The selection of creative concepts in design teams is influenced by personality traits and risk attitudes, emphasizing the significance of individual diversity in the outcomes of the design process within the complex field of design teams; a crucial stage is the selection of creative concepts, which is heavily impacted by the interaction between team members’ tolerance for risk and personality qualities. This dynamic interplay emphasizes how important individual variety is in influencing how the design process turns out. An in-depth investigation into this relationship provides insightful information on the complexity of decision-making in design teams [38] and the influence of personality factors on cognitive biases, specifically focusing on the disposition effect and herding behavior. It highlights the significant importance of risk attitude as a moderator in these biases [39]. The study conducted by [40] examines the impact of improving personality factors on the entrepreneurial intentions of immigrants. The researchers employed structural equation modeling techniques to analyze the data and draw conclusions. Prior studies indicate that the association between personality traits and decision-making is contingent upon the specific environment and is subject to additional factors, including risk attitudes and cognitive biases.

Most of the decision-making processes in organizational settings depend on the evaluation approaches, including personality characteristics, which have a significant importance [41]. The Big Five model implementation is not only limited to personality assessment, and their applications have been made in many fields like decision-making, performance, and the role of culture [42]. These studies support the role of personality traits, i.e., those classified within the Big Five model, as prominent in decision-making processes within different domains. Cultural, environmental, and context-based scenarios may influence the decision-making process [43].

2.3. Existing Methodologies in Quality Assessment

A multivariate method is used in scientific research to assess and test multivariate causal links in structural equation modeling or SEM. Compared to other modeling techniques, SEM assess the impacts, both direct and indirect, on intended causal connections [44]. SEM is a statistical technique that has evolved over three generations and is around 100 years old. Using path analysis, the first generation of SEM created the causal modeling logic [45,46,47]. Factor analysis was subsequently added to SEM by the social sciences. SEM increased its capability with the second generation. The “structural causal model” [48], created by Judea Pearl in 2000, marked the beginning of the third generation of SEM, and after that, Bayesian modeling was then included [49].

The Naïve Bayes algorithm, which solves classification problems, is a supervised learning algorithm based on Bayes theorem. The primary use for it is in-text classification. The two terms Naïve and Bayes, which together make up the Naïve Bayes algorithm, are defined as follows: Naive: It receives its name from the naive assumption that one feature’s occurrence stands alone from the occurrence of other features. For instance, if some fruit is defined based on its color, shape, and taste, then an apple is identified as a red, spherical, sweet fruit. Thus, each characteristic works independently of the others to help distinguish that it is an apple. Bayes: The reason it is named Bayes is that it relies on Bayes’ Theorem [50].

2.4. Impact of Knowledge on Decision-Making

This study examines the associations between the Big Five personality traits, various forms of knowledge evaluation, and their impact on decision-making outcomes within the context of engineering design. Figure 1 shows the structural model, which consists of two outcomes: Quality of Decision (QOD) and Confidence in Decision (CID). It is postulated that the outcomes are impacted by the Big Five personality characteristics, namely extraversion, agreeableness, conscientiousness, neuroticism, and openness. Furthermore, two types of knowledge assessment are addressed: self-assessed knowledge (SAK) and test-assessed knowledge (TAK). Figure 1 shows the directional linkages between the constructs and decision-making as indicated by the arrows representing the structural paths.

Figure 2 visually represents the nine latent variables in our study: extraversion (EXC), agreeableness (AGF), conscientiousness (COA), neuroticism (NEQ), openness (OPA), Confidence in Decision (CID), Quality of Decision (QOD), Self-Assessed Knowledge (SAK), and Test-Assessed Knowledge (TAK).

Each latent variable is shown with its corresponding indicators, for instance, EXC1, EXC2, and EXC3 for extraversion, and similarly for other constructs.

2.5. The Impact of the Big Five Personality Characteristics on Decision Quality and Confidence

Extraversion. The study of personality traits, specifically extraversion, within the framework of engineering design performance, provides a comprehensive viewpoint. The construct of extraversion, which encompasses traits such as social skills, assertiveness, and energy, has been the subject of investigation in diverse fields of work, including the field of engineering design [51]. Individuals exhibiting higher levels of extraversion tend to display an aptitude for pleasant affect and a preference for engaging in interpersonal relationships [52] and often hold positions of leadership [53]. Besides having minimal social interaction, individuals show the characteristics of supremacy, ambition, and activity [54]. In the context of occupational achievement, extraversion has been recognized, alongside neuroticism and conscientiousness, as a significant contribution element [55]. It is important to acknowledge that the influence of extraversion on the quality of decision-making is not universally positive. A study conducted on the subject of supply chain management revealed that individuals with lower degrees of extraversion tend to have superior decision-making abilities [6]. This finding is consistent with other studies showing that although extraversion and agreeableness may contribute to the development of trust, they do not necessarily improve the standard of making decisional choices or the quality of decisions [56]. Hence, although extraversion may offer benefits in specific social and professional settings, its impact on decision-making quality, notably within supply chains and in engineering design, has a more complex nature. This brings us to our following hypotheses:

Hypothesis 1 a.

A negative association between the extraversion characteristics and the quality of a decision.

Research findings have suggested that individuals who display extraverted tendencies tend to possess greater levels of confidence in their decision-making processes [57,58]. The increased level of confidence is imputed to the extrovert’s tendency for external stimulation and their focus on the external environment [59]. Furthermore, extroverts’ experience shows enhancement in their decision confidence because of engaging in social interactions. This gain in confidence is often followed by an associated uplift in positive emotional states [60].

The level of confidence can decrease when individuals with extroverted tendencies find themselves in circumstances that demand introverted conduct. Extraverts, when confronted with situations that require them to behave in a manner that contradicts their natural outgoing and interactive tendencies, encounter cognitive and emotional limitations, resulting in decreased confidence regarding their decision-making abilities [61,62]. Therefore, it may be argued that the impact of extraversion on decision-making confidence is contingent upon the specific context and can be influenced by situational elements that are inconsistent with an extrovert’s inherent tendencies. In the case of tasks that primarily involve individual effort, it can be postulated as the hypothesis below.

Hypothesis 1 b.

A negative association between the extraversion characteristics and the level of confidence in decision.

Agreeableness. Agreeableness is a cooperative trait that is essential to developing and maintaining meaningful relationships with other people [63]. Individuals who possess significant levels of agreeableness are frequently characterized as exhibiting qualities such as kindness, thoughtfulness, and warmth. These attributes inherently contribute to the promotion of collaboration and the facilitation of harmonious interpersonal exchanges [64]. On the other hand, individuals characterized by lower levels of agreeableness are more likely to encounter challenges in their interpersonal relationships and may be viewed as contentious [65].

It is important to acknowledge that within the framework of the Big Five personality traits, agreeableness and openness are regarded as the least robust factors. This is primarily because of the presence of contradictions in their theoretical structure [66]. However, a strong correlation exists between higher levels of agreeableness and efficacy. Individuals revealing such traits tend to engage in behaviors that promote interpersonal attractiveness, facilitate transparent communication, and create cooperation. Therefore, these behaviors play a significant role in dispute resolution and have a positive impact on the overall functioning of a team [67].

The relationship between agreeableness and performance quality is not definitively established. While it is widely acknowledged that agreeableness can have a positive impact on interpersonal dynamics within teams, there is still a need for additional research to fully understand its effects on decision-making quality and individual task performance. In the context of individual tasks, it is reasonable to make the assumption below.

Hypothesis 2 a.

A negative association between the agreeableness features and the quality of a decision.

The trait of agreeableness is often linked to cooperative and sociable conduct; persons with higher levels of this feature are more inclined towards collaboration and interpersonal connections [63]. Although these attributes can enhance communication and conflict resolution within group dynamics, they may also impact individuals’ levels of confidence in decision-making. According to existing research, persons who possess a high level of agreeableness may demonstrate a tendency towards “tender-mindedness”, rendering them more sensitive to external influences and perhaps resulting in decision biases [68]. The presence of vulnerability has the potential to compromise the rationality of individuals’ decision-making processes, hence diminishing their confidence in the outcomes of their choices. Additionally, the lack of team communication could lead to the decline of their trust in decision-making, as they frequently depend on external information [69]. Hence, while agreeableness may offer advantages in contexts that need collaboration, it can also lead to difficulties in terms of building trust in decision-making. Therefore, it can be hypothesized by the statement below.

Hypothesis 2 b.

A negative association between the personality attribute agreeableness and the level of confidence in decision.

Conscientiousness. Conscientiousness is a robust and established construct within the area of the development of personalities and the examination of individual differences [70]. This concept is a recognized method of managing impulsive behavior that promotes engagement in tasks and behaviors aligned with certain objectives. These include engaging in thoughtful actions, delaying immediate satisfaction, and adhering to established social norms [71]. Individuals who display conscientiousness attributes are commonly characterized by their self-discipline, organizational skills, and persistent effort. Conscientiousness is a strong predictor of job success in various professional domains, and it also demonstrates predictive power with other performance indicators [72,73]. The existing studies support the notion that conscientiousness has a significant impact on the quality of decision-making, yielding beneficial outcomes [55]. Therefore, Hypothesis 3 (a) is proposed.

Hypothesis 3 a.

A positive relation between the conscientiousness attribute and the quality of a decision.

Individuals who possess a high level of conscientiousness frequently demonstrate a tendency to engage in deliberate processes when confronted with critical issues, highlighting their drive towards thorough examination. Previous studies have established a positive relationship between conscientiousness and confidence, namely in areas such as reading, writing, and time management abilities, which are separate from the process of decision-making. Conscientious individuals tend to attain higher levels of achievement, but their overall well-being may be negatively impacted, particularly when faced with failure. Research has shown a significant decrease in satisfaction among conscientious individuals in such circumstances [71]. Conscientiousness exhibits an intriguing negative correlation with risk-taking behaviors, maybe indicating a prudent and deliberate attitude towards uncertainty that is commonly observed among conscientious individuals [74]. In situations characterized by complexity and a lack of knowledge about future outcomes, researchers have observed a notable inverse relationship between conscientiousness and confidence. The complicated relationship between conscientiousness and confidence is shown by this fascinating interaction, suggesting that the impact of conscientiousness on confidence may depend on the specific circumstances and characteristics of the activity being undertaken. So, we can hypothesize the statement below.

Hypothesis 3 b.

There exists a negative relation between conscientiousness and confidence in decision.

Neuroticism. The complex interaction between neuroticism and other work-related performance measures presents a captivating area for academic research [75]. Neuroticism is a consistent and reproducible personality trait that shows a notable lack of emotional adjustment, namely with stress, anxiety, and despair, observed in diverse situations [54]. Individuals having high levels of neuroticism are frequently acknowledged as showing symptoms of anxiety, depression, anger, worry, insecurity, and emotional reactivity. Studies conducted in the field of engineering design have shown evidence that neuroticism is a significant predictor of unfavorable work-related performance. More specifically, those who exhibit a higher degree of neuroticism, specifically those in the top quintile, demonstrate an inclination to make errors at a rate that is twice as high as individuals in the lower quintiles. Nonetheless, a notable difference arises in situations where persons with high levels of neuroticism outperform their emotionally stable counterparts, particularly in demanding job contexts or when they demonstrate a significant degree of exertion [76]. The establishment of this variation is dependent upon the nature of the task being performed. Additional investigation reveals a significant inverse relationship between neuroticism and designers’ level of motivation for performance; however, the association between neuroticism and procrastination seems to be somewhat weaker [77]. The varied and occasionally contradictory results from different studies highlight the complex effects of neuroticism on decision-making quality in engineering design contexts. This calls for a more detailed investigation to better understand the underlying dynamics of action. Thus, we can formulate the hypothesis below.

Hypothesis 4 a.

A negative association between the personality attribute of neuroticism and the quality of a decision.

Individuals who show high levels of neuroticism tend to display an inclination towards increased levels of worry, as indicated by both self-reported attitudes and behavioral choices in experimental settings. There exists a positive correlation between neuroticism and an increased propensity for engaging in risk-taking behavior, particularly when individuals are exposed to acute stress [78]. Occurrence of adverse emotional states such as anxiety, aggression, and sadness might potentially influence the cognitive processes involved in decision-making [79]. Moreover, the presence of neuroticism has been found to have an effect on both the psychological and physiological reactions to stress, which in turn can potentially affect an individual’s ability to make decisions in stressful situations [80]. Therefore, Hypothesis 4 (b) is proposed.

Hypothesis 4 b.

There exists a negative correlation between the personality attribute of neuroticism and the level of confidence in deciding.

Openness. Openness refers to a characteristic of an individual’s personality that involves actively seeking and valuing novel experiences, as well as demonstrating a willingness to tolerate and explore unexpected situations. Enhanced creativity has been closely linked to the trait of openness. Individuals who exhibit high levels of openness are more likely to actively participate in activities that promote creativity. This, in turn, improves their self-perceived level of creativity. The research placed significant emphasis on the mediation function of intrinsic motivation in cultivating creativity among individuals who exhibit high levels of openness [81], creative behavior [82], adaptability, and enhanced decision-making [83]. Individuals who possess a high level of openness, due to their inclination towards innovation and willingness to take risks, are more likely to succeed in dynamic environments. This enables them to make insightful business choices that can positively impact the growth and performance of organizations. It can be hypothesized with the statement below.

Hypothesis 5 a.

The openness aspect and the quality of decision-making have a positive association.

Openness and inventive performance within organizations imply a cultivating culture of creativity and potentially support trust in decision-making processes, particularly within the context of engineering design [84,85]. The influence of qualities such as openness on decision-making might exhibit diverse effects across distinct organizational contexts [86]. The potential influence of openness on the process of decision-making could be influenced or monitored by additional elements, such as emotional intelligence [87]. A hierarchical approach to decision-making is proposed, wherein higher- and lower-order decisions are integrated. The complexity of decision-making models may be subject to impacts from personality factors, such as openness [88].

Hypothesis 5 b.

A positive association between openness aspects and an individual’s level of confidence in making a decision.

2.6. Influence of Domain Knowledge on the Quality of Decision-Making and Confidence in Decision

How individuals use knowledge to solve problems differs significantly from one another. Individuals who are more knowledgeable in an area of work tend to be more confident. The acquisition of knowledge about something in particular might result in the appearance of a cognitive bias that causes individuals to rely on their prior knowledge before making decisions [89]. The type of information that a decision-maker needs to make an informed decision highlights the value of having access to more information. It implies that having a broad understanding might help one make better decisions, especially in fields that require a broad range of expertise, like engineering design [27]. The quality of decisions may be impacted if self-evaluation is occasionally out of synchronization with actual capabilities, according to research on self-assessment of knowledge and abilities. Although it raises the possibility that self-assessed knowledge may be altered in the context of driver education, affecting the decisions made [90]. We therefore postulate the following:

Hypothesis 6 a.

Self-reported engineering design knowledge is negatively associated with decision quality.

Engineering design decision-making is multifaceted; it takes into account the designer’s skill, and it is influenced by environmental settings [91]. Making knowledgeable choices is the major responsibility of a designer, and assistance is essential for improving skills and expediting the creation of high-quality designs [35]. Implicit information is used by experienced designers, considerably influencing engineering design decision-making [92]. Making informed decisions is made easier by expressing design information using a design-rule framework, which simplifies maintaining extensive knowledge bases [93]. Furthermore, experienced engineering design individuals generally exert more cognitive effort, which improves decision-making [94]. Thus, we propose the following hypothesis:

Hypothesis 6 b.

Test-evaluated engineering design knowledge’s positive association with the quality of a decision.

The distinction between self-reported information, commonly referred to as subjective knowledge and knowledge assessed by testing, known as objective knowledge, holds considerable significance due to the unique influence that each form of knowledge may exert on the decision-making process. Understanding this distinction is crucial for gaining a deeper comprehension of the decision-maker’s capacity [6]. Despite their simplicity and cost-effectiveness, self-report measures have demonstrated limited agreement with objective performance. It follows that the existence of this phenomenon can influence decision-making processes whenever one relies on self-reported information [95]. In empirical software engineering research and education, it is often necessary to ask students to assess their own perceived level of experience. Empirical research has shown that self-rated experience has predictive validity in predicting performance in programming activities. The application of self-report measures is significantly growing, particularly in knowledge assessment. As mentioned, there is a chance that over-reliability on self-reported information may lead to an incorrect representation of knowledge and, therefore, reduce decision-making confidence levels [96].

Further research needs to be conducted to critically investigate various influences and effects of self-reported and test-evaluated competence on decision-making within the engineering design field. This exploration is important to gain a better understanding of how these factors influence individuals’ confidence in their decision-making abilities.

Hypothesis 7 a.

A positive relation between Self-reported knowledge in engineering design and confidence in a decision.

Hypothesis 7 b.

A positive relation between test-evaluated knowledge in engineering design knowledge confidence in a decision.

3. Research Methodology

3.1. Research Design

The present study utilizes a research design that combines qualitative and quantitative methodologies to examine the influence of personality traits and domain knowledge on the process of decision-making tasks as shown in Figure 3. The study utilizes a simulation game as a replication source for modeling decision-making scenarios, specifically in the context of design problems. Simulation games are considered to be valuable resources in the field of engineering education, as they enable a transition from the conventional method of learning through lectures to a more engaging and immersive learning approach [97].

The program offers a practical methodology for creating roller coasters, involving students in realistic engineering problems, potentially mirroring the decision-making procedures inherent in engineering design [98].

3.2. Experimental Tasks

Figure 3 shows the view of the experimental task of track design problem where constraints are specified (TDPCS) [36]. Participants are required to maximize the “enjoyment experienced by the rider of the track” under a constraint that the centripetal acceleration should not exceed 4 g.

The objective function for the enjoyment is mathematically modeled but not disclosed to the participants to simulate real-world conditions where exact mathematical models are often unknown as shown in Figure 4. Each task is divided into seven periods, and each period permits seven decision-making tries, cumulating to 2205 (45 participants × 7 periods × 7 tries); these decisions were used to analyze qualitative and quantitative data. Alongside the quantitative metrics, participants also need to write cognitive (thought in mind, approach to decisions, or analytic) explanations for each of their decisions.

3.3. Participants

The sample includes 45 participants from diverse academic and cultural backgrounds, primarily majoring in various engineering departments at both the undergraduate and graduate levels.

The participants for the study were recruited through various methods including advertisements, flyers, and notices posted on campus boards, targeting a diverse group of engineering students. The recruitment strategy was designed to ensure a representative sample of participants from different academic and cultural backgrounds, majoring in various engineering disciplines at both the undergraduate and graduate levels.

To motivate participants and align their efforts with the study objectives, an incentive structure based on performance was implemented. Participants were compensated based on the ratio of the maximum function value they achieved during the task to the theoretical maximum value possible. This ratio was multiplied by a constant rate of $2.50, allowing participants the opportunity to earn up to $2.50 per period. To control potential wealth effects, as in [31], participants were informed that their compensation for each task would be based on their performance in two randomly selected periods. Given three tasks in total, participants could earn a maximum of $15.00 in performance-based incentives, in addition to a participation fee of $5.00. This methodological approach not only incentivized performance but also controlled extraneous variables affecting the reliability of the data collected.

3.4. Quality Assessment Method

Unlike other metrics, decision quality is evaluated through a methodologically rigorous, multi-faceted approach designed to overcome the limitations of subjectivity encountered in previous research. The procedure can be broken down into the following key steps:

Initial Author Evaluation: Each of the 2205 decision statements across all participants was initially assessed by one of the study’s authors. The evaluations were completed using a Likert scale ranging from 1 to 5, with 1 representing the lowest quality and 5 representing the highest quality. These assessments established a foundational baseline for the study.

Machine learning Validation—Naïve Bayes: the study further trains and tests a Naive Bayes model using textual decision statements for each participant to validate the initial author evaluations concerning the quality of the decision statements.

Participant-Level Aggregation: A single aggregated quality score for each of the 45 participants was calculated following the initial author evaluations. An average of 49 individual decisions corresponding to each participant were assessed by the authors in this study and were computed to generate this aggregated quality score. This aggregate score represents the cumulative decision quality for each participant during the experiment.

Expert Validation: To further strengthen the robustness of these quality measurements, the averages of all quality scores for the 45 participants were subsequently reviewed and then validated by domain experts in the field. This triangulation between initial author assessments and expert evaluations significantly reduces subjectivity, hence strengthening the reliability of this study’s findings [99,100,101].

3.5. Data Collection and Assessment Methods

Big Five Inventory (BFI): The Big Five Inventory is utilized to evaluate the personality characteristics of individuals. It assesses five primary aspects of personality, namely extraversion, agreeableness, conscientiousness, neuroticism, and openness [51,102,103].

Self-Assessed Knowledge (SAK): participants self-assess their engineering design knowledge and roller coaster game-related expertise using a Likert scale from 1 to 5, where 1 shows low knowledge and 5 shows high knowledge.

Test-Assessed Knowledge (TAK): written questions graded by the authors serve to test the participants for specific engineering design domain knowledge, especially for knowledge that is relevant to roller coaster track design.

Confidence Assessment: After making each decision, participants rate their confidence in that decision on a Likert scale from 1 to 5, where 1 shows less confidence and 5 shows the highest confidence. Each of the 45 participants made 49 decisions across seven periods, cumulating to 2205 decisions. For each participant, the following metrics were averaged to produce single, representative values:

• Width of the track design;
• Enjoyment value associated with the track;
• Self-assessed confidence in each decision.

These aggregated metrics facilitate a more streamlined quantitative analysis by reducing the dimensionality of the dataset.

3.6. Data Analysis

Jupyter Lab: the analyses and visualizations in this study were conducted using Jupyter Lab, which provided an interactive environment for implementing models, processing data, and generating the visual representations presented in this article.

Structural Equation Modeling: Partial least squares (PLSs) is a statistical method commonly used in multivariate analysis. It is particularly useful when dealing with datasets that include many variables. A structural model comprising nine latent variables is estimated with PLS-SEM to assess the relationships among personality traits, domain knowledge, decision quality, and confidence [96].

Naive Bayes Validation: Textual decision statements are used to train and evaluate a Naive Bayes model to validate the initial author evaluations of decision quality. Performance metrics like mean squared error (MSE) and R-squared ($R^2$) are used to assess the model’s predictive accuracy. The Naive Bayes model further validates these evaluations, enhancing the study’s reliability. Naive Bayes has proven to be effective in text classification tasks and is less prone to overfitting, especially when the dataset is not large. Naive Bayes serves as a baseline model. Its simplicity and usability make it an effective baseline for comparison with complex models. If the Naive Bayes model performs sufficiently well, it may not be necessary to move to more complex algorithms, thereby holding to the principle of Occam’s razor [104], which recommends the simplest possible model that performs adequately.

4. The Results of Hypothesis Testing

4.1. Descriptive Analysis

The latent variables in our study, presented in Figure 1, include Extraversion Characteristics (EXCs), Agreeableness Features (AGFs), Conscientiousness Attributes (COAs), Neuroticism Factors (NEFs), Openness Aspects (OPAs), Quality of Decision (QOD), Confidence in Decision (CID), Self-Assessed Knowledge (SAK), and Test-Assessed Knowledge (TAK).

Each latent variable is assessed through specific indicators.

Table 1. Construct, means, standard deviations, and loadings.

Construct	Indicator	Mean	Std. Dev.	Loading
Extraversion Characteristics (EXC)	EXC1	1.69	0.701	0.963 *
	EXC2	1.73	0.688	0.940 *
	EXC3	1.73	0.751	0.899 *
Agreeableness Features (AGFs)	AGF1	1.91	0.733	0.931 *
	AGF2	1.93	0.720	0.956 *
	AGF3	1.93	0.688	0.963 *
Conscientiousness Attributes (COAs)	COA1	2.82	0.777	0.988 *
	COA2	2.84	0.737	0.995 *
	COA3	2.80	0.815	0.983 *
Neuroticism Factors (NEFs)	NEF1	1.82	0.806	0.991 *
	NEF2	1.82	0.777	0.991 *
Openness Aspects (OPAs)	OPA1	3.76	0.883	0.994 *
	OPA2	3.76	0.933	0.986 *
	OPA3	3.80	0.894	0.984 *
Self-Assessed Knowledge (SAK)	SAK1	2.84	1.021	0.962 *
	SAK2	2.78	1.042	0.933 *
Test-Assessed Knowledge (TAK)	TAK1	3.11	1.005	0.985 *
	TAK2	3.13	0.991	0.987 *

Note: * significant at the <0.001 level (two-tailed test).

includes seven of these latent variables, with QOD and CID detailed later due to their unique role as outcome variables in the structural model.

The Big Five Inventory (BFI) was used to measure personality traits (extraversion, agreeableness, conscientiousness, neuroticism, and openness). For each personality construct, the indicators are represented as follows:

• EXC1, EXC2, and EXC3 for Extraversion Characteristics;
• AGF1, AGF2, and AGF3 for Agreeableness Features;
• COA1, COA2, and COA3 for Conscientiousness Attributes;
• NEF1 and NEF2 for Neuroticism Factors;
• OPA1, OPA2, and OPA3 for Openness Aspects.

The Big Five Inventory (BFI) was used to measure personality constructs. Each latent variable is measured via observed variables corresponding to individual items within the BFI. Detailed descriptions of these items are provided in the Supplementary Materials.

The mean, standard deviations, and loadings of the construct for each indicator under consideration are shown in Table 1. The mean values exhibit a range from 1.69 for the indicator EXC1 to 3.80 for OPA3. The highest mean values are evident in the Openness Aspects (OPA) construct, while the lowest is observed in the Extraversion Characteristics (EXC) construct. The means for most of the indicators are dispersed across a spectrum that leans towards the higher end of the scale, particularly for the OPA and TAK constructs.

As for the standard deviations, these values oscillate between 0.688 for EXC2 and 1.042 for SAK2. Indicators within the Extraversion Characteristics (EXCs) construct have the least variability, as indicated by their lower standard deviations. Indicators belonging to the Self-Assessed Knowledge (SAK) construct display the highest variability, denoting a broader range of responses.

4.2. The Assessment of Validity and Reliability

The assessment of validity and reliability of measures for multi-item constructs in models is an essential factor in ensuring the strength and precision of the constructs being evaluated.

Table 2. Construct measures of reliability and validity.

Construct.	Cronbach’s Alpha	Composite Reliability	Average Variance Extracted (AVE)
Extraversion Characteristics (EXCs)	0.928	0.952	0.873
Agreeableness Features (AGFs)	0.946	0.953	0.902
Conscientiousness Attributes (COAs)	0.988	0.989	0.977
Neuroticism Factors (NEFs)	0.982	0.983	0.982
Openness Aspects (OPAs)	0.988	0.993	0.976
Self-Assessed Knowledge (SAK)	0.889	0.940	0.898
Test-Assessed Knowledge (TAK)	0.971	0.975	0.972

provides the results of the constructs, and a visual presentation is provided in Figure 5. In the context of partial least squares structural equation modeling (PLS-SEM), the values are organized according to their indicator’s reliability, ensuring that each latent variable within the model is reliably measured [105]. A Cronbach’s alpha value exceeding the threshold of 0.7 is indicative of good internal consistency, ensuring that the items within the construct are well correlated and consistently measure the same characteristic [106,107]. An AVE higher than 0.5 demonstrates high convergent validity, indicating that the constructs in the model are unidimensional and well measured [107] shown in Figure 6.

4.3. The Estimation Results of Model

The Naive Bayes model validation for decision quality assessment, such as mean squared error (MSE) and R-squared ($R^2$), and to assess the model’s predictive accuracy are presented in

Table 3. Naive Bayes model validation.

Metric	Value
Nested Cross-Validation Score (Mean $R^2$)	0.688
Mean Squared Error (MSE)	0.015
R-squared ($R^2$)	0.989

, and a graphical representation is shown in Figure 7.

In the context of Quality of Decision assessment (QOD), the Naive Bayes model exhibits good predictive efficacy, as evidenced by an $R^2$ value of approximately 0.989 and a small mean squared error (MSE) of 0.0151.

This high $R^2$ value is indicative of the model’s ability to account for the variance in the dependent variable, thereby offering a highly accurate representation of the underlying data. The model’s generalizability is further supported through a rigorous nested cross-validation approach, which yielded a robust mean $R^2$ value of 0.688. These empirical results validate the model’s applicability and reliability, rendering it reliable and effective for tool decision-making scenarios, thereby contributing a methodologically rigorous, statistically robust, and practically relevant model to the extensive literature on decision analytics.

Table 4. Square roots of the mean–variance and correlations between latent variables.

Construct	EXC	AGF	COA	NEF	OPA	SAK	TAK
EXC	0.934	−0.124	0.003	−0.049	−0.229	−0.122	−0.362
AGF		0.950	−0.174	0.117	−0.082	0.078	0.056
COA			0.988	−0.348	0.244	0.165	0.425
NEF				0.991	−0.286	−0.119	−0.131
OPA					0.988	0.122	0.216
SAK						0.948	0.150
TAK							0.986

shows the correlation between latent variables, and a heterotrait–monotrait (HTMT) ratio graph is provided in Figure 8 and a heatmap visualization graph is presented in Figure 9. It shows a considerable negative correlation between extraversion (EXC) and various other characteristics. This suggests that individuals with higher levels of extraversion may exhibit lower performance on knowledge tests. The trait of conscientiousness (COA) is found to have a significant positive association with neuroticism (NEF) while displaying a negative association with Self-Assessed Knowledge (SAK). This suggests that persons who score high in conscientiousness may also have neurotic tendencies but tend to underestimate their level of knowledge. The presence of a strong association between Neuroticism (NEF) and Openness (OPA) implies that persons with higher levels of neuroticism may potentially exhibit more acceptance of new experiences.

The heterotrait–monotrait (HTMT) ratio graph is provided in Figure 8 and provides valuable insights into the discriminant validity of the constructs. Evaluating discriminant validity is essential to confirm that the constructs measured are distinct and do not overlap significantly.

Most construct pairs depicted in the graph have HTMT values significantly below the 0.90 threshold, which is commonly accepted for establishing discriminant validity.

A few construct pairs, such as EC5 and SIC, OC3 and CC1, demonstrate higher HTMT values nearing or exceeding 0.50. These instances suggest a potential overlap and require further scrutiny to ensure that the constructs are adequately distinct.

Comparative Analysis with Fornell–Larcker Criterion: The Fornell–Larcker criterion assesses discriminant validity by comparing the square root of the average variance extracted (AVE) with the correlations between the constructs. While effective, it can be less sensitive in cases where constructs are highly correlated but still distinct. The application of HTMT in our study reinforces the discriminant validity of most constructs, with a few exceptions that warrant further investigation. By using HTMT alongside traditional methods like Fornell–Larcker, we can achieve a more comprehensive validation of our measurement model, enhancing the reliability and interpretability of our research findings. This dual approach ensures that our constructs are both distinct and relevant, providing a solid foundation for the study’s conclusions and implications.

There exists a significant positive correlation between Self-Assessed Knowledge (SAK) and Test-Assessed Knowledge (TAK), suggesting an association between individuals’ self-perceived knowledge and their actual performance in knowledge evaluations.

The determination coefficient ($R^2$) of the equations that reveal the inherent components, specifically, the quality of the decision (QOD) and confidence in the decision made (CID), is presented in

Table 5. Effect sizes and structural model results.

Criterion	Indicators	${\mathit{R}}^2$	Path Coefficient	${\mathit{f}}^2$
Quality	EXC	0.628	$-0.213$	0.090
	AGF		$-0.210$	$0.109$
	COA		$0.260$	0.118
	NEF		$-0.185$	0.075
	OPA		$0.277$	0.168
	SAK		$-0.030$ (ns)	0.002
	TAK		$0.266$	0.122
Confidence	EXC	0.333	$0.010$ (ns)	0.000
	AGF		$-0.211$	0.061
	COA		$-0.002$ (ns)	0.000
	NEF		$-0.199$	0.048
	OPA		$0.232$	0.066
	SAK		$0.026$ (ns)	0.001
	TAK		$0.322$	0.100

Note: non-significant (^ns) at the <0.005 level (one-tailed test).

. The model shows a higher level of explanatory power for decision quality (0.628) but shows a comparatively lower level of explanatory ability for decision confidence (0.333). Furthermore, Table 5 provides path coefficients and effect sizes, denoted by $f^2$ of each indicator.

The effect of Extraversion Characteristics (EXC) on the quality of a decision (QOD) was found to be significant, with a path coefficient of $(-0.213; p<0.05).$ Thus, Hypothesis H1 (a) is confirmed. The effect of Extraversion Characteristics (EXCs) on confidence in a decision (CID) was found to be non-significant with a path coefficient of $(0.010; p>0.05).$ Thus, Hypothesis H1 (b) is not confirmed. For extraversion, it is noted that while it might be associated with positive leadership outcomes and social interaction, its direct impact on decision-making quality might not be significant. This aligns with findings that the interaction between extraversion and expertise, and it can predict group decision quality, suggesting a more complex relationship dependent on other factors like group dynamics and task type [108].

The effect of Agreeableness Features (AGFs) on the quality of a decision (QOD) was significant, with a path coefficient of $(-0.210; p<0.05).$ Hypothesis H2 (a) was confirmed. The effect of Agreeableness Features (AGFs) on confidence in a decision (CID) was significant, with a path coefficient of $(-0.211; p>0.05).$ Hypothesis H2 (b) was not confirmed. Regarding agreeableness, research has shown that high levels of agreeableness correlate with cooperation and empathy but may not directly enhance decision quality. This could be due to the trait’s association with a preference for harmony and consensus over optimal or challenging decision-making scenarios [109]. Furthermore, a quantitative review of agreeableness highlights its multifaceted influence on behavior, which can have varying implications for decision-making depending on the context [110].

The effect of Conscientiousness Attributes (COAs) on the quality of a decision (QOD) was found to be significant, with a path coefficient of $(0.260; p<0.05)$. Therefore, Hypothesis H3 (a) is confirmed. Conversely, the effect of Conscientiousness Attributes (COAs) on confidence in a decision (CID) was non-significant, with a path coefficient of $(-0.002; p>0.05)$. Thus, Hypothesis H3 (b) is not confirmed.

The effect of Neuroticism Factors (NEFs) on the quality of a decision (QOD) was significant, with a path coefficient of $(-0.185; p<0.05).$ Therefore, Hypothesis H4 (a) is confirmed. However, the effect of Neuroticism Factors (NEFs) on confidence in a decision (CID) was non-significant, with a path coefficient of $(-0.199; p>0.05)$. Thus, Hypothesis H4 (b) is not confirmed.

The effect of Openness Aspects (OPAs) on the quality of a decision (QOD) was significant, with a path coefficient of $(0.277; p<0.05).$ Hypothesis H5 (a) is thus confirmed. The effect of Openness Aspects (OPAs) on confidence in a decision (CID) was non-significant, with a path coefficient of $(0.232; p>0.05)$. Therefore, Hypothesis H5 (b) is not confirmed.

The effect of Self-Assessed Knowledge (SAK) on the quality of a decision (QOD) was found to be non-significant, with a path coefficient of $(-0.030; p>0.05).$ Thus, Hypothesis H6 (a) is not confirmed.

The effect of Test-Assessed Knowledge (TAK) on the quality of a decision (QOD) was found to be significant, with a path coefficient of $(0.266; p<0.05)$. Therefore, Hypothesis H6 (b) is confirmed.

The effect of Self-Assessed Knowledge (SAK) on confidence in a decision (CID) was found to be non-significant, with a path coefficient of $(0.026; p>0.05).$ Consequently, Hypothesis H7 (a) is not confirmed.

The effect of Test-Assessed Knowledge (TAK) on confidence in a decision (CID) was found to be significant, with a path coefficient of $(0.322; p<0.05)$. Hence, Hypothesis H7 (b) is confirmed.

The innovations of this study lie in its comprehensive approach to analyzing decision-making quality and confidence within engineering design. A key novelty is the integration of the Big Five personality traits with both self-assessed and test-assessed domain knowledge, providing a holistic understanding of how these factors interact to influence decision outcomes. This approach goes beyond prior research that typically examines personality or knowledge in isolation. The study also introduces a multi-faceted evaluation methodology that combines subjective assessments, expert validation, and machine learning techniques (Naive Bayes model), ensuring robust and unbiased decision quality assessments. Additionally, the use of simulation games to replicate real-world decision-making scenarios adds ecological validity to the findings, making them highly applicable to professional engineering challenges.

Unlike previous studies that often rely on homogenous participant groups, this research includes diverse participants from various academic and cultural backgrounds, enhancing the generalizability of its findings. Furthermore, the differentiation between test-assessed and self-assessed knowledge offers new insights, revealing that decision confidence is more strongly influenced by validated knowledge than by perceived competence. This distinction underscores the importance of factual knowledge validation in training and professional development.

Overall, the study’s practical implications are significant, emphasizing the alignment of tasks with individual personality traits and the prioritization of objective knowledge in decision-making roles. These contributions provide a framework for improving decision-making processes in educational and professional settings, making the study both innovative and impactful. The summary is shown in

Table 6. Summary of the hypotheses results. Null rejection is indicated by ✔, while failure is indicated by ✘.

Hypotheses	Results	Hypotheses	Results
(H1a) For QOD by EXC	✔	(H4b) For CID by NEF	✘
(H1b) For CID by EXC	✘	(H5a) For QOD by OPA	✔
(H2a) For QOD by AGF	✔	(H5b) For CID by OPA	✘
(H2b) For CID by AGF	✘	(H6a) For QOD by SAK	✘
(H3a) For QOD by COA	✔	(H6b) For QOD by TAK	✔
(H3b) For CID by COA	✘	(H7a) For CID by SAK	✘
(H4a) For QOD by NEF	✔	(H7b) For CID by TAK	✔

.

5. Discussion

This study provides the complex relationship between personality traits and decision-making aspects, particularly focusing on decision quality and confidence. Showing consistency with previous research, our findings indicate that extraversion and agreeableness negatively affect decision quality, aligning with the hypothesis that these traits favor cooperative and communicative behaviors over task-oriented performance [55,111]. This reinforces the notion from earlier studies that such traits may detract from decision quality in contexts requiring high cognitive focus and less interpersonal interaction.

In contrast, conscientiousness and openness demonstrated a positive impact on decision quality, showing their relevance in roles that demand keen attention to detail and innovation—traits that are highly valued in engineering design tasks. This aligns with the trait activation theory, which posits that the relevance of personality traits to job performance outcomes is activated by corresponding job demands [112].

Interestingly, our study extends the existing literature by showing that none of these personality traits significantly influence the confidence level in decision-making. This is a departure from some past findings and suggests that confidence may be influenced more by other factors, such as the complexity of the task or the individual’s experience rather than their stable personality traits.

Furthermore, while neuroticism was found to negatively impact decision quality, consistent with its established association with poorer job performance across many studies [113], it did not significantly affect confidence levels. This provides a new perspective to the debate on the role of neuroticism in professional settings, suggesting that while it may reduce performance quality, it does not necessarily alter an individual’s confidence in their decisions.

A notable contribution of our study is the differentiation between the impacts of test-assessed and self-assessed knowledge on decision quality and confidence. We found that test-assessed knowledge was a significant predictor of both, reinforcing the value of objective measures of knowledge in predicting task performance [114]. In contrast, self-assessed knowledge did not show a similar influence, highlighting the potential discrepancies between perceived and actual knowledge and their effects on decision-making outcomes.

Our research thus offers new insights into how specific personality traits and knowledge assessments correlate with key aspects of decision-making within the context of engineering design. It suggests that while certain personality traits are consistently linked with decision quality, their influence on decision confidence is less pronounced, and the role of actual knowledge proficiency (as opposed to perceived knowledge) is critical for both decision quality and confidence.

5.1. Contributions

This study presents a novel approach to addressing the issue of subjectivity in quality assessment, which has been identified as a significant concern in existing scholarly works. In contrast to prior investigations that primarily relied on the assessment of a single author, the present study adopts a more thorough methodology. The process involves incorporating assessments from experts alongside evaluations provided by a single author to enhance the evaluation of decision quality. To improve the credibility of these assessments. The utilization of a multi-faceted strategy in this study not only enhances the credibility of the findings but also establishes an academic pathway for incorporating both human experience and machine learning methods in future investigations related to the evaluation and prediction of quality in decision-making.

We expanded the applicability of our findings by involving a more diverse group of individuals in comparison to previous investigations. Our study includes a total of 45 participants who come from a range of academic and cultural backgrounds. These participants are enrolled in different engineering departments at both the undergraduate and graduate levels. The utilization of this comprehensive sampling technique is consistent with the objective of the study, which is to investigate the influence of personality factors and domain-specific knowledge on decision quality and confidence in the field of engineering design. The fact that self-reported knowledge in the field of engineering design for roller coaster track design does not significantly increase the decision-maker’s confidence may seem a bit unexpected. It is often anticipated that the perception of knowledge will contribute to an increase in confidence. In contrast, the validation of knowledge through testing significantly enhances both confidence levels and the overall quality of decision-making. As the participants’ understanding of engineering design grows stronger their sense of confidence also increases. The significant influence of knowledge on both confidence and decision-making highlights the potential for substantial improvement in both confidence levels and decision quality, even through a minor allocation of resources towards training.

5.2. Theoretical Implications

This study’s findings contribute to the understanding of personality traits’ impact on decision-making within the context of engineering design, reinforcing the theoretical framework that personality significantly influences professional behaviors and outcomes. This is aligned with the Big Five personality theory, which suggests that certain traits such as conscientiousness and openness are particularly relevant in contexts requiring high levels of cognitive engagement and problem-solving abilities [111,115].

Moreover, this research highlights the importance of context in the manifestation of personality traits’ effects on decision-making. The differential impact of traits like agreeableness and extraversion across various decision-making scenarios supports the situational strength theory, which posits that the influence of personality on behavior is contingent on the strength of situational cues [116]. In engineering design, where tasks can greatly vary in terms of complexity and novelty, personality traits interact with the task environment to influence decision quality and confidence.

Additionally, by demonstrating that specific personality traits like openness and conscientiousness are more predictive of decision-making quality in complex tasks, this research provides empirical support for the trait activation theory [112]. This theory suggests that the presence of specific job-relevant cues in the environment can ‘activate’ certain personality traits, enhancing their predictive power regarding job performance.

As the participants’ understanding of engineering design grows stronger, their sense of confidence also increases. This significant influence of knowledge on both confidence and decision-making highlights the potential for substantial improvement in both confidence levels and decision quality, even through a minor allocation of resources towards training. Importantly, the impact of such training may vary by age, as participants with fewer life and professional experiences might respond differently, underlining the need to tailor educational interventions to different age groups to maximize their effectiveness.

5.3. Implications for Practice

There is a need to integrate an individual’s personality qualities with the context of the task, particularly within the engineering design domain [29]. These traits exhibit not only a beneficial relationship with job satisfaction across various professions but also exert an important impact on decision-making approaches within a professional environment and academic settings.

Within the field of engineering design, the task requirements can vary significantly, ranging from conventional roles to those that are highly specialized and tailored to the unique needs of a certain organization. The existing information possessed by a designer can have a significant impact on conventional work roles, hence improving the quality of decision-making. Conversely, within occupations that require a high level of specialization, certain personality traits, particularly openness and conscientiousness, may have greater importance. Research findings indicate that persons who possess higher levels of conscientiousness and openness tend to exhibit a stronger capacity for making optimal decisions.

5.4. Limitations and Future Research

In this study, we aim to go further into the subject of decision-making. Our objective is to conduct a comprehensive investigation that will contribute to the existing body of knowledge in this field. By examining many factors and variables that influence decision-making.

Additional research is necessary to fully understand the long-term effects and possible wider use of including simulation games, varied participants, and sample size as means of assessing decision-making. The advancement and inclusion of decision support systems have the potential to enhance the efficiency of decision-making in the field of engineering design. The inclusion of domain knowledge in machine learning can be approached in a more comprehensive manner, which has the potential to reveal new insights and improve the interpretability of models. The continued development and enhancement of hybrid modeling techniques, which integrate machine learning with simulation, have the potential to yield predictive models that are both more precise and informative.

This study offers significant advancements over previous studies by integrating personality traits and knowledge assessments to provide a comprehensive understanding of decision-making in engineering design. Unlike earlier research, which often examines personality or knowledge in isolation, this study explores their interaction, offering nuanced insights into how they jointly influence decision quality and confidence. Furthermore, the use of a multi-faceted evaluation methodology—combining subjective assessments, expert validation, and machine learning (Naive Bayes model)—ensures robust and reliable findings, addressing the limitations of single-author evaluations seen in prior studies.

The use of simulation games is another innovative aspect, allowing participants to engage in realistic, constraint-driven scenarios that mirror real-world engineering challenges. This contrasts with the theoretical or static experimental approaches commonly used in earlier research, enhancing the ecological validity of the findings. The study also stands out for its inclusion of diverse participants from various academic and cultural backgrounds, improving the generalizability of its results compared to studies with homogenous groups.

Additionally, the differentiation between self-assessed and test-assessed knowledge highlights the importance of validated knowledge in decision-making, a factor often overlooked in previous work. By focusing on actionable insights, such as aligning tasks with individual personality traits and emphasizing test-assessed knowledge, this study bridges the gap between theoretical understanding and practical application, making it a valuable contribution to improving decision-making processes in education, training, and professional settings.