Dark Triad traits and subjective career success mediated by emotional intelligence in the South African SMME sector

1 Introduction

Personality traits remain an important area of investigation in the workplace, as they may enhance or impede success (Eshete et al., 2025). While research has traditionally focused on positive psychological traits that enhance performance and wellbeing, there is increasing recognition that socially aversive personality characteristics affect career outcomes (Oldewage and Jonck, 2025). Among these, the Dark Triad traits (DTT), namely Machiavellianism, narcissism, and psychopathy, have attracted considerable scholarly attention due to their association with manipulation, self-interest, reduced empathy, and strategic interpersonal behavior (Khan et al., 2025). Although these traits are generally regarded as maladaptive, emerging research suggests that they may, under certain conditions, confer advantages such as assertiveness and strategic thinking (Blasco-Belled et al., 2024; Furnham and Treglown, 2021). However, their implications for career success remain mixed, particularly when career success is conceptualized subjectively rather than objectively. Unlike objective indicators, such as salary or promotion, subjective career success (SCS) reflects individuals' personal evaluations of their careers in terms of recognition, quality of work, meaningful work, influence, authenticity, as well as growth and development (Shockley et al., 2016). Understanding how personality traits influence these subjective dimensions of career success is therefore necessary to develop a more holistic understanding of career success and its antecedents.

This study is grounded in Trait Activation Theory (Tett and Burnett, 2003), which postulates that personality traits are expressed in response to situational cues, and that behavioral outcomes depend on contextual demands and individual regulatory capabilities. From this perspective, the effects of DTT are contingent on how these traits are enacted within specific organizational environments. In work contexts characterized by close interpersonal interactions, such as small, medium and micro-enterprises (SMMEs), the expression of these traits might influence how individuals perceive their career success.

Emotional intelligence (EQ) plays a key role in career advancement and is often regarded as more important than analytical and technical skills (Arora, 2017). In this study, EQ is concerned with the perception of emotional abilities, thus an individual's evaluation of their ability to understand, regulate, and express emotions to adapt to their environment (Petrides et al., 2016). This approach aligns closely with personality research and is typically measured using self-report instruments (Gerhardt et al., 2026). A meta-analysis by Michels and Schulze (2021) found that Machiavellianism and psychopathy are negatively related to trait EQ, whereas narcissism is positively related to trait EQ. Similarly, a systematic review by Walker et al. (2021) confirmed predominantly negative associations between Dark Triad traits and EQ. Thus, EQ can be understood as a mechanism through which personality traits are expressed, ultimately influencing SCS (Sharma and Tiwari, 2024).

Given the critical role of SMMEs in economic development and decent employment creation (Rambe and Nyagadza, 2026), understanding the psychological factors that influence career experiences in this sector is of both theoretical and practical importance. Despite this, limited empirical research has examined how DTT relate to SCS within the South African SMME context, particularly when considering the mediating role of EQ. Against this background, the present study investigates the relationship between DTT and SCS among employees in the South African SMME sector, with a specific focus on the mediating role of EQ.

2 Problem statement

Despite the burgeoning body of knowledge on Dark Triad personality traits, much of the existing literature has focused on the influence thereof on organizational outcomes such as leadership, performance (Bueno-de la Fuente et al., 2025), and counterproductive workplace behavior (Harrison and Summers, 2025), with limited attention given to their role in subjective career experiences. In particular, the extent to which these socially aversive traits influence SCS remains insufficiently understood (Oldewage and Jonck, 2025), especially given the multidimensional nature of SCS, which extends beyond objective indicators.

At the same time, EQ has been identified as a key predictor of positive workplace outcomes (Woime and Shato, 2025), including interpersonal effectiveness, wellbeing, and personal and professional success (Tjimuku et al., 2025). However, its role as an explanatory mechanism in the relationship between maladaptive personality traits and SCS remains underexplored. While emerging evidence suggests that EQ might influence how personality traits are expressed in organizational contexts (Antonopoulou, 2023), limited research has examined its mediating role in the relationship between DTT and the various dimensions of SCS. This gap is particularly evident within the South African SMME sector, where organizational structures are less formalized and resource-constrained conditions are evident (Rambe and Nyagadza, 2026). In such environments, personality traits and emotional regulation capabilities are likely to play a significant role in shaping career experiences. Consequently, there is a need for empirical research to examine how DTT influence SCS, and the extent to which EQ mediates these relationships within the South African SMME context.

The rationale for the mentioned is that the SMME sector in the South African context contribute between 52 and 57 percent to the country's Gross Domestic Product (GDP) and it is extrapolated that this would increase to 90 percent. Despite the pivotal role of SMMEs over 50 percent of SMMEs foreclose within five years (Nkwinika, 2023). Thus, the sector is characterized by volatility, unpredictability, resource constraints, high levels of uncertainty, interpersonal dependence, and dynamic work environments. The Dark Triad traits are postulated condition-sensitive adaptations to socio-ecological unpredictability. Consequently, the socio-economic conditions that are significantly associated with the Dark Triad traits are unpredictability, resource constraints and austerity (Jonason et al., 2016). The Dark Triad traits are highly adaptive in volatile and uncertain environments such as the those indicative of the SMME sector. Consequently, the contextual conditions of the SMME sector activate or motivate the expression of Dark Triad traits as suggested by the Trait Activation Theory. Jonason et al. (2016) explain that the Dark Triad traits might be condition-dependent coping mechanisms to solve difficulties in unpredictable, resource constrained and volatile conditions.

3 Literature review

3.1 Dark Triad traits

The Dark Triad (DT) consists of subclinical personality traits, namely Machiavellianism, narcissism, and psychopathy (Khan et al., 2025; Verma et al., 2026). Research on DT traits indicates that these characteristics share common features, such as interpersonal manipulation, self-interest, and a lack of empathy, while remaining distinct in their behavioral manifestations (Motavalli et al., 2025; Gholami et al., 2026). Furthermore, recent literature suggests that these traits are underpinned by a common dark core of personality, characterized by the prioritization of personal gain at the expense of others (Moshagen et al., 2018; Bader et al., 2021). However, some scholars, such as Furnham and Treglown (2021), argue that the typically adverse characteristics of DT traits may be advantageous in competitive environments. Moreover, contemporary research proposes that individuals who exhibit these traits in moderation may be better able to navigate complex organizational settings, pursue ambitious goals, and maintain resilience in highly competitive professional contexts (Nübold et al., 2017; Parker et al., 2018).

Individuals who score high on Machiavellianism typically exhibit characteristics such as strategic manipulation, emotional detachment, emotional regulation, and a pragmatic approach to interpersonal relationships (Nübold et al., 2017; Jahangir et al., 2025). In addition, these individuals tend to employ calculated strategies to influence others and achieve personal objectives, particularly in competitive organizational contexts (Marbut et al., 2025). By contrast, individuals who score high on narcissism are characterized by grandiosity, a strong sense of entitlement, and a need for admiration and recognition (Abdelrahman et al., 2024). Despite this, traits associated with Machiavellianism and narcissism, such as strategic thinking, confidence, and goal orientation, have been linked to leadership emergence, career advancement, and effective negotiation outcomes (Bueno-de la Fuente et al., 2025).

In contrast, psychopathy is generally regarded as the most maladaptive of the DT traits and is associated with impulsivity, callousness, and a diminished capacity for guilt or remorse (Cardoso et al., 2023). In organizational environments, psychopathic tendencies have been linked to risk-taking, unethical behavior, and reduced concern for the wellbeing of colleagues (Kirrane et al., 2025). Moreover, the literature indicates that the interpersonal deficits associated with psychopathy and other DT traits may undermine long-term relational outcomes, particularly in contexts where emotional awareness and social competence are required (Kirrane et al., 2025).

Building on this, the following section shifts focus to SCS, examining how such interpersonal characteristics may influence career-related outcomes over time.

3.2 Subjective career success (SCS)

The customary definition of career success has long been conceptualized in terms of indicators such as salary progression, hierarchical advancement, and occupational status (Dai and Song, 2016), that is, objective career success. While objective career success focuses on measurable outcomes, SCS emphasized the psychological and experiential aspects of career development (Shockley et al., 2016). Recent studies increasingly recognize that career success also encompasses individuals' subjective evaluations of their career experiences and achievements (Shockley et al., 2016). Shockley et al. (2016) assert that SCS is determined by an individual's personal interpretation of satisfaction and fulfillment derived from career trajectories, reflecting internal evaluations rather than externally observable achievements.

Consequently, as modern careers have become less structurally defined, subjective evaluations have assumed a central role in understanding how individuals assess the quality and meaning of their work lives (Shockley et al., 2016; Tripathi and Singh, 2021). Subjective career success (SCS) includes dimensions such as meaningful work, authenticity, work-life balance, and opportunities for personal growth (Shockley et al., 2016). In addition, recent research suggests that individuals increasingly evaluate their careers based on intrinsic criteria, such as alignment with personal values, opportunities for development, and contributions to overall wellbeing (Zhou et al., 2016; Syahrizal and Marisaa, 2023).

Accordingly, SCS can be understood as a multidimensional construct encompassing both professional achievements and broader life satisfaction (Zhou et al., 2016). To conceptualize this construct, Shockley et al. (2016) developed the Subjective Career Success Inventory (SCSI), which includes several dimensions, notably, recognition, quality work, meaningful work, influence, authenticity, personal life, and growth and development. Recognition refers to the extent to which individuals feel acknowledged and respected for their contributions, whereas quality work reflects the perception that one's work is performed to a high standard. Meaningful work denotes the sense that work activities contribute to a greater purpose, while influence refers to the ability to shape decisions and organizational outcomes. Authenticity reflects alignment between personal values and work roles, whereas personal life represents the balance between professional and non-work domains. Growth and development relate to opportunities for learning, advancement, and skill development across the career lifespan (Shockley et al., 2016).

The contextual theoretical underpinning of SCS include, for example Boundaryless Career Theory, Protean Career Theory and the Chaos Theory of Careers. Within contemporary career environments, Boundaryless Career Theory (BCT) offers an additional lens for understanding SCS. BCT suggests that careers are created by individual agency, mobility, career competencies, and self-directed management, while still being influenced by structural and institutional conditions (Tams and Arthur, 2010; Guan et al., 2019). In this context, emotional capabilities become essential resources for navigating dynamic and uncertain work environments. In addition, a study by Gerli et al. (2015) indicates that emotional competencies positively influence physical career mobility, while social competencies are associated with career advancement and objective career success in a boundaryless career context. Accordingly, the present study argues that EQ may facilitate the management of Dark Triad tendencies, thereby influencing SCS within South African SMMEs especially since careers in the mention sector could be deemed boundaryless.

Protean Career Theory (PCT) further strengthens the theoretical foundation of the present study by emphasizing self-directed career management and psychologically driven definitions of success. PCT, as indicated by Hall et al. (2018), proposes that individuals evaluate career success mainly in terms of psychological success, including personal growth, authenticity, fulfillment, and alignment with personal values, rather than relying solely on objective career indicators. In protean career contexts, emotional and interpersonal competencies are increasingly important for sustaining adaptive career behaviors and career satisfaction. More recent scholars argue that individuals with a stronger self-directed, values-driven career orientation are better positioned to manage a changing work environment (Wiernik and Kostal, 2019; Sultana et al., 2024). Therefore, EQ may buffer the potentially detrimental interpersonal effects of Dark Triad traits, thus shaping SCS within South African SMMEs.

In addition, Chaos Theory of Careers (CTC) also provides a relevant framework for understanding SCS within contemporary organizational settings. The theory proposes that individual characteristics, in addition to changing environmental conditions, lead to careers being non-linear and unpredictable (Pryor and Bright, 2022). Within such uncertain contexts, psychological adaptability and emotional regulation become critical resources for navigating career challenges and transitions. Consequently, EQ may assist individuals in regulating the interpersonal and behavioral expressions associated with Dark Triad traits, thereby influencing perceptions of SCS within the dynamic and unpredictable South African SMME sector.

In organizational contexts, subjective career outcomes vary, as they are influenced by a range of psychological and interpersonal factors, including personality traits, emotional capabilities, and workplace relationships (Ng and Feldman, 2014; Spurk et al., 2020). Moreover, Miao et al. (2017) suggest that (EQ) is associated with improved interpersonal functioning, enhanced work relationships, and greater career satisfaction. Consequently, individuals with higher levels of EQ may experience more favorable career perceptions, as they are better equipped to navigate social interactions, manage workplace challenges, and cultivate supportive professional networks (Miao et al., 2017).

3.3 Emotional intelligence (EQ)

According to Gkintoni et al. (2025), EQ refers to the ability to effectively regulate one's emotions by perceiving, understanding, managing, and utilizing emotions in oneself and others. Accordingly, it is hypothesized that EQ might play a pivotal role in shaping the relationship between DTT and career outcomes (Valeriia, 2025). The literature suggests that individuals with higher levels of EQ are better equipped to regulate the interpersonal expressions of dark personality traits, thereby influencing perceptions of career success (Miao et al., 2019).

In organizational contexts, EQ is increasingly recognized as a critical competency that facilitates adaptive functioning, interpersonal effectiveness, and career-related outcomes (Dulewicz and Higgs, 2001; Coronado-Maldonado and Benítez-Márquez, 2023; Appiah-Kubi et al., 2025).

Research consistently demonstrates that EQ is a significant predictor of workplace behaviors and career experiences. Emotionally intelligent individuals are better equipped to navigate high-emotion labor environments, build positive relationships, and manage stress (Miao et al., 2017). As a result, individuals with higher levels of EQ are more likely to regulate emotional responses, resolve conflict, and engage in constructive interpersonal interactions, which are essential for achieving favorable career outcomes (Stein and Book, 2011; Kakarla, 2025).

Conversely, individuals who exhibit DTT tend to demonstrate reduced empathy and emotional detachment, which may result in ineffective emotional processing and impaired social functioning (Valeriia, 2025). Against this background, the present study investigates the influence of EQ on SCS and examines the extent to which EQ mediates the relationship between DT personality traits and subjective career outcomes within the South African SMME sector.

4 Hypothesis development

4.1 Dark Triad traits (DTT) and subjective career success (SCS)

Trait Activation Theory posits that personality traits are expressed in response to situational cues and that the behavioral outcomes thereof depend on the interaction between individual dispositions and contextual demands (Tett and Burnett, 2003). From this perspective, the influence of DTT, namely Machiavellianism, narcissism, and psychopathy, on career outcomes is not inherently fixed, but is contingent on how these traits are activated and expressed within organizational environments. Given that SCS reflects individuals' internal evaluations of their career experiences, it is particularly sensitive to interpersonal dynamics.

Nonetheless, the literature presents mixed findings regarding the relationship between DDT and career success. For example, Machiavellinism has been found to be positively related to SCS, whereas psychopathy has been shown to be negatively related (Spurk et al., 2016). Volmer et al. (2016) found that leader narcissism had a positive effect on SCS, Machiavellianism had a negative effect, and psychopathy was unrelated to follower SCS. In contrast, Paleczek et al. (2018) reported no significant association between Machiavellianism and career success. Similarly, Furnham and Treglown (2021) found that SCS was positively associated with narcissism. More recently, Oldewage and Jonck (2025) reported that psychopathy had a statistically significant negative effect on career success, narcissism had no significant impact, and Machiavellianism had a detrimental effect in a similar context. Paleczek et al. (2018) suggest that DTT may, in some instances, be beneficial for career advancement, highlighting that these traits can exhibit adaptive characteristics under certain conditions.

Despite these mixed findings, DTT are expected to influence SCS within the SMME context, where interpersonal interactions and contextual demands may amplify the expression of these traits.

4.2 Emotional intelligence (EQ) as mediator between Dark Triad traits (DTT) and subjective career success (SCS)

A growing body of research suggests that EQ plays a fundamental role in the relationship between DTT and SCS (Miao et al., 2019; De Beer, 2024). From a theoretical perspective, EQ reflects an individual's ability to perceive, regulate, and utilize emotions effectively in both interpersonal and intrapersonal contexts (Elfenbein and MacCann, 2017; Xuan, 2025). These capabilities are particularly important in workplace environments that require social competence and adaptive functioning (Elfenbein and MacCann, 2017; Miao et al., 2019; De Beer, 2024).

Recent research further indicates that individuals exhibiting high levels of DTT, namely Machiavellianism, psychopathy, and narcissism, may achieve short-term gains (Nuzulia and Why, 2020; Jiang et al., 2025). However, the characteristics associated with these traits, such as emotional detachment, manipulation, and reduced empathy, are often linked to long-term relational difficulties and diminished career fulfillment in collaborative environments (Nuzulia and Why, 2020; Jiang et al., 2025). Michels and Schulze (2021) suggest that EQ may serve as a regulatory mechanism that mitigates these adverse effects by enhancing emotional awareness and behavioral adaptability.

Building on this perspective, EQ can be conceptualized as a psychological resource that buffers the negative influence of maladaptive personality traits on subjective evaluations of career success. In doing so, EQ not only reduces the impact of interpersonal deficits associated with DTT but also facilitates more adaptive career behaviors and perceptions. Consequently, EQ is posited to function as a mediating mechanism in the relationship between DTT and SCS, ultimately promoting more sustainable and fulfilling career outcomes.

Based on the preceding discussion, this study proposes a theoretical model, as illustrated in Figure 1.

5 Research methodology

5.1 Research approach and procedure

A quantitative, cross-sectional survey design was employed in this study. This research design was considered appropriate due to its ability to collect, analyze, and interpret quantifiable data for hypothesis testing (Ghanad, 2023). The cross-sectional nature of the design refers to data collected from a sample at a single point in time (Wang and Cheng, 2020). Such designs are observational and are frequently used to examine relationships between variables. In the present study, determinants of psychological wellbeing were investigated; therefore, a cross-sectional approach was deemed suitable.

Instead of a web-based survey, a paper-and-pencil survey was administered by trained enumerators. These enumerators were responsible for recruiting participants, administering informed consent forms, and following up with participants during the data collection phase (Kombe et al., 2019).

5.2 Participants and sampling methods

The study targeted individuals employed in SMMEs within the greater Bloemfontein area, a provincial capital located in central South Africa. SMMEs were defined as: (i) micro-enterprises (one to four employees), (ii) small enterprises (five to nine employees), and (iii) medium enterprises (10 to 49 employees), in line with Bhorat et al. (2018).

In terms of scale, the Free State province had approximately 27,000 SMMEs (Johannesburg Stock Exchange (JSE), 2024), with an estimated 771,000 individuals employed during the fourth quarter of 2025 (Statistics South Africa, 2026). A total of n = 396 questionnaires were collected using convenience sampling. Convenience sampling is a non-probability sampling technique in which participants who meet the inclusion criteria are selected based on accessibility (Golzar et al., 2022). The primary inclusion criterion was current employment within a SMME at the time of data collection.

A sample of n = 396 yields an estimated effect size of 0.35 and statistical power of 0.92 at a 5% margin of error (Sedar et al., 2021), indicating that the sample was sufficient for the purposes of this study.

Of the 396 participants, males (n = 222; 56.3%) slightly outnumbered females (n = 170; 43.1%). One participant identified as “other” (n = 1; 0.3%), and one participant opted not to disclose gender (n = 1; 0.3%). The age distribution was relatively balanced across categories, with 39.1% (n = 154) aged 36–45 years, followed by 27.7% (n = 109) aged 25–35 years, and 25.4% (n = 100) aged 46–55 years. Participants aged 56–65 years constituted 7.6% (n = 30), while one participant (0.3%) was older than 65 years.

Regarding educational attainment, most participants held a Grade 12 qualification (n = 184; 46.6%), followed by a bachelor's degree (n = 76; 19.2%), an honors degree (n = 62; 15.7%), a postgraduate qualification (n = 59; 14.9%), and a doctoral degree (n = 5; 1.3%). A small portion selected “other” (n = 9; 2.3%).

In terms of work experience, most participants (n = 185; 46.7%) had between 1 and 5 years of experience, followed by 6–10 years (n = 127; 32.1%), 11–15 years (n = 34; 8.6%) and 16–20 years (n = 21; 5.3%). Additionally, 5.1% (n = 20) reported 21–25 years of experience, and 2.3% (n = 9) had more than 25 years of experience.

Regarding SMME classification, 47.8% (n = 185) were employed at micro-enterprises, 34.0% (n = 137) in small enterprises, and 17.3% (n = 68) in medium enterprises. In terms of organizational roles, 38.7% (n = 147) were in middle management, 29.5% (n = 112) in senior management, and 31.8% (n = 121) were business owners.

5.3 Measuring instrument

The instruments used in this study included: (i) the Dark Triad Dirty Dozen (DTDD), (ii) the Subjective Career Success Inventory (SCSI), (iii) the Rahim Emotional Quotient Index (REQI), and (iv) a biographical questionnaire capturing participants' demographic information.

The DTDD was used to assess the prevalence of Dark Triad traits (DTT). The scale was developed by Jonason and Webster (2010) to measure three dimensions, namely narcissism, Machiavellianism, and psychopathy. The instrument consists of 12 items rated on a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The DTDD has been validated for a South African sample by Oldewage and Jonck (2025), reporting a Cronbach's alpha of 0.865 and a composite reliability of 0.938. In addition, discriminant validity has been established for the scale (Oldewage and Jonck, 2025).

Subjective career success (SCS) was measured using the SCSI, developed by Shockley et al. (2016), which comprises 24 items. The SCSI measures seven dimensions: recognition, quality work, meaningful work, influence, authenticity, personal life, and growth and development. Responses were recorded on a five-point Likert-type scale ranging from 1 (not at all) to 5 (a great deal). Example items include: “I have been able to pursue work that meets my personal needs and preferences” (authenticity) and “I have continuously improved by developing my skillset” (growth and development). The psychometric properties of the SCSI have been confirmed in a South African context, with a Cronbach's alpha of 0.938, composite reliability of 0.963, and established discriminant validity (Oldewage and Jonck, 2025).

EQ was measured using the Rahim Emotional Quotient Index (REQI) 22-item scale. The REQI was developed by Rahim et al. (2002) to assess trait EQ (Petrides and Furnham, 2000). The instrument consists of 22 items measuring five dimensions: self-awareness, self-regulation, motivation, empathy, and social skills, rated on a seven-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). Example items include: “I am aware of which emotions I am experiencing and why” (self-awareness) and “I keep my distressing emotions in check” (self-regulation).

Research suggest that EQ as measured by the REQI can be treated as a unidimensional construct in applied research, particularly in the South African context (Nel et al., 2015). Moreover, the suitability of the REQI for use in South Africa has been supported by several studies (see, for example Kotzé and Nel, 2015; Nel et al., 2015). Nel et al. (2015) confirmed the psychometric properties of the REQI for a South African sample.

5.4 Analysis method

The data were analyzed using the Statistical Package for the Social Sciences (IBM SPSS) version 30, SPSS Amos version 30, and SPSS PROCESS Macro version 4.3 (Model 4). Prior to hypothesis testing, the psychometric properties of the scales were assessed. Cronbach's alpha coefficients were calculated to evaluate reliability, with values above 0.70 considered acceptable (Tavakol and Dennick, 2011).

Convergent validity was assessed using composite reliability (CR > 0.70) and average variance extracted (AVE > 0.50) (Ali et al., 2020). An AVE value of 0.50 indicates that the latent construct explains at least 50% of the variance in its observed indicators (Van Tonder et al., 2026). Discriminant validity was evaluated using the heterotrait-monotrait ratio (HTMT), with values below 0.90 considered acceptable, while a more conservative threshold of 0.85 is preferred (Ali et al., 2020).

Common method variance (CMV) was identified as a potential limitation due to the use of self-report measures. CMV may arise when the same measurement method is used for both independent and dependent variables, potentially inflating parameter estimates (Polas, 2025). To assess CMV, Harman's single-factor test was employed. Specifically, an unrotated exploratory factor analysis (EFA) was conducted to determine whether a single factor accounted for more than 50% of the total variance (Aguirre-Urreta and Hu, 2019).

Subsequently, the theoretical model was estimated using structural equation modeling (SEM) in SPSS Amos (Ramlall, 2017). Model fit was evaluated using multiple indices, including the Tucker-Lewis index (TLI), Comparative Fit Index (CFI), chi-square statistic, and Root Mean Square Error of Approximation (RMSEA) (see, for example Ali et al., 2020 and Scholtz et al., 2024). Threshold values of CFI ≥ 0.95, TLI ≥ 0.95, and RMSEA < 0.08 were considered indicative of acceptable model fit (Scholtz et al., 2024).

Both descriptive and inferential statistical analyses were conducted. Descriptive statistics included the mean, median, standard deviation, skewness, and kurtosis. Skewness reflects the degree of asymmetry in a distribution, whereas kurtosis indicates the extent to which a distribution is peaked or flat (Kim, 2013). According to Kline (2015), values within ± 3 for skewness and ± 10 for kurtosis indicate acceptable normality.

Inferential analyses included regression analysis to assess the direct relationships between variables. Mediation analysis was conducted using the SPSS PROCESS Macro version 4.3 (Model 4) (Field, 2024). A bias-corrected bootstrap procedure was employed to estimate indirect effects, with 95% confidence intervals (lower-level confidence interval [LLCI] and upper-level confidence interval [ULCI] that excluded zero indicating significant mediation (McCallaghan et al., 2019). Statistical significance was assessed at the 95% (p ≤ 0.05) and 99% (p ≤ 0.01) confidence levels.

5.5 Ethical considerations

Ethical clearance was obtained from the Ethics Committee of the Central University of Technology, Free State (reference number CUT/HREC 2024/000273). Standard ethical protocols were adhered to, including obtaining informed consent, ensuring voluntary participation, maintaining confidentiality and anonymity, and upholding principles of beneficence. Data protection was ensured by securely storing all data on a password-protected laptop.

6 Results

6.1 Measuring instrument psychometric properties and model fit

Item consistency, reliability, and validity were assessed using Cronbach's alpha coefficients and composite reliability (CR). Average variance extracted (AVE) values were used to assess convergent validity, while heterotrait-monotrait (HTMT) ratios were used to evaluate discriminant validity (see Table 1).

Table 1. Convergent validity and heterotrait-monotrait discriminant validity.

CA, Cronbach's alpha; CR, composite reliability; AVE, average variance extracted.

As shown in Table 1, Cronbach's alpha (CA) and composite reliability (CR) values exceeded the recommended threshold of 0.70, indicating satisfactory internal consistency (Van Tonder et al., 2026). Specifically, DTT yielded a CA of 0.91 and a CR value of 0.96, while subjective career success reported a CA of 0.94 and a CR value of 0.94. EQ demonstrated a CA of 0.97 and a CR of 0.96.

Convergent validity was assessed using AVE values. dos Santos and Cirillo (2021) suggest that latent constructs should explain at least 50% of the variance in their indicators, implying an AVE threshold of 0.50. As shown in Table 1, SCS reported a slightly lower AVE value (0.49), while DTT and EQ exceeded the recommended threshold. However, convergent validity may still be considered adequate when CR exceeds 0.60, even if AVE is slightly below 0.50.

Discriminant validity was evaluated using HTMT ratios. HTMT values below 0.85 indicate adequate discriminant validity (Ali et al., 2020). As all HTMT values were below this threshold, discriminant validity was confirmed. These results indicate that each construct is empirically distinct and measures a unique concept (Van Tonder et al., 2026).

To assess whether the structural model was suitable for hypothesis testing, model fit indices were calculated using structural equation modeling (SEM) with maximum likelihood estimation in SPSS Amos version 30. The results indicated the following fit indices: χ² = 188.773; p < 0.001; degrees of freedom (df) = 6; χ²/df = 31.462; normed fit index (NFI) = 0.90; comparative fit index (CFI) = 1.00; and root mean square error of approximation (RMSEA) = 0.247.

The CFI exceeded the recommended threshold of 0.90, indicating acceptable model fit (Van Zyl and Ten Klooster, 2022). However, the RMSEA value suggests poor fitness. McNeish et al. (2018) noted that high RMSEA values (e.g., 0.20) might not always indicate poor fit if measurement quality is high. Additionally, Kenny et al. (2015) noted that RMSEA often falsely indicates a poor fit with models characterized by a small degree of freedom (df). Hence, computing the RMSEA value for small degree of freedom models are not recommended (Kenny et al., 2015). RMSEA values should be used alongside other indices like CFI (Shi et al., 2022). Additionally, the high χ2/df ratio indicates potential model misspecification. On the other hand, the size of the model, based on the number of observed variables, could inflate the chi-square value (Shi et al., 2018). In the current study the number of observed variables might have inflated the chi-square value. Additionally, χ2/df ratio is sensitivity to sample size, in large samples, even a minor deviation can result in an inflated value, causing the model to be rejected even if it is practically useful (Berg, 2015). Considering the fore mentioned, it is acknowledged that the model shows poor fit according to these indices and that the structural findings should be interpreted as preliminary with considerable caution.

6.2 Method variance

Given that Likert-type items were used to measure both predictor and outcome variables, common method bias (CMB) was identified as a possible internal threat that required statistical control. Measurement error may affect the accuracy of estimated relationships between observed and latent variables, potentially leading to biased inferences (Podsakoff et al., 2024).

To assess CMB, an unrotated exploratory factor analysis (EFA) was conducted (Jonck and Manamela, 2025). Kock et al. (2021) suggest that if a single factor accounts for more than 50% of the total variance, common method bias may be present. Preliminary diagnostics indicated that the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.938, and Bartlett's test of sphericity was statistically significant (χ² = 19882.436; df = 2145; p < 0.001), supporting the suitability of the data for factor analysis.

The unrotated EFA revealed that 11 factors accounted for 68.703% of the total variance, with the first factor explaining 33.342% of the variance. As this value is below the 50% threshold, common method bias was not considered a significant threat in this study.

6.3 Descriptive results

Table 2 presents the descriptive statistics for the measured variables.

Table 2. Descriptive statistics for study variables.

As shown in Table 2, the mean scores for the DTT exceeded the median values, suggesting that participants tended to endorse higher levels of these traits. Specifically, Machiavellianism had a mean of 5.33 (median = 4.00), narcissism had a mean of 7.30 (median = 6.00), and psychopathy had a mean of 6.20 (median = 6.00).

Regarding the dimensions of SCS, most variables demonstrated mean values that were equal to or slightly above their respective medians, indicating generally positive perceptions of career success. The exception was personal life, which had a slightly lower mean (12.86) compared to the median (13.00), suggesting that participants may perceive some imbalance between their professional and personal lives.

For emotional intelligence, the mean score (183.34) was lower than the median (191.00), indicating a slight negative skew in responses.

Skewness values were within acceptable limits, while most kurtosis values fell within recommended thresholds. However, Machiavellianism and growth and development exhibited kurtosis values exceeding recommended limits, indicating deviations from normality for these variables.

6.4 Correlation analysis

To determine the strength and direction of the bivariate relationships, Pearson product-moment correlations were computed. One of the underlying assumptions of correlation analysis is that the sample size is sufficiently large to support the applicability of the central limit theorem (Field, 2024). In this study, the sample size (n = 396) was deemed adequate to meet this assumption. The results of the correlation analysis are presented in Table 3.

Table 3. Bivariate correlations for the measured variables.

^*p ≤ 0.05; ^**p ≤ 0.01. MaC, Machiavellianism; Psy, psychopathy; Nar, narcissism; Rec, recognition; Q, quality work; MW, meaningful work; Inf, influence; Aut, authenticity; PL, personal life; GAD, growth and development; Sat, satisfaction; EQ, emotional intelligence.

As shown in Table 3, psychopathy demonstrated a statistically significant positive association with Machiavellianism at the 99% confidence level (r = 0.516 large effect, p < 0.01). Narcissism was significantly positively associated with both Machiavellianism (r = 0.252, small effect, p < 0.01) and psychopathy (r = 0.704, large effect, p < 0.01). Recognition exhibited a small but statistically significant negative association with narcissism (r = −0.108; p < 0.05).

Quality work demonstrated both positive and negative associations, including a small positive relationship with Machiavellianism (r = 0.107, p < 0.05), and small negative relationships with psychopathy (r = −0.142, p < 0.01) and narcissism (r = −0.169, p < 0.01). A strong positive association was observed between quality work and recognition (r = 0.648, p < 0.01).

Meaningful work showed statistically significant negative associations with psychopathy (r = −0.217, small effect, p < 0.01) and narcissism (r = −0.204, small effect, p < 0.01), as well as strong positive associations with recognition (r = 0.587, p < 0.01) and quality work (r = 0.644, p < 0.01).

Influence exhibited several statistically significant relationships, including negative associations with psychopathy (r = −0.156, small effect, p < 0.01) and narcissism (r = −0.159, small effect, p < 0.01), and strong positive associations with recognition (r = 0.610, p < 0.01), quality work (r = 0.618, p < 0.01), and meaningful work (r = 0.771, p < 0.01).

Authenticity did not show statistically significant associations with the DTT; however, it demonstrated strong positive relationships with recognition (r = 0.572, p < 0.01), quality work (r = 0.545, p < 0.01), meaningful work (r = 0.682, p < 0.01), and influence (r = 0.658, p < 0.01).

Growth and development exhibited several statistically significant associations, including negative relationships with psychopathy (r = −0.131, small effect, p < 0.01) and narcissism (r = −0.152, small effect, p < 0.01), and positive relationships with recognition (r = 0.421, medium effect, p < 0.01), quality work (r = 0.439, medium effect, p < 0.01), meaningful work (r = 0.572, large effect, p < 0.01), influence (r = 0.512, large effect, p < 0.01), authenticity (r = 0.456, medium effect, p < 0.01), and personal life (r = 0.723, large effect, p < 0.01).

Regarding satisfaction, several statistically significant relationships were observed, including a small positive association with Machiavellianism (r = 0.128, p < 0.05) and a small negative association with narcissism (r = −0.152, p < 0.01). Satisfaction was also positively associate with recognition (r = 0.401, medium effect, p < 0.01), quality work (r = 0.444, medium effect, p < 0.01), meaningful work (r = 0.522, large effect, p < 0.01), influence (r = 0.503, large effect, p < 0.01), authenticity (r = 0.502, large effect, p < 0.01), personal life (r = 0.653, large effect, p < 0.01), and growth and development (r = 0.658, large effect, p < 0.01).

Finally, EQ demonstrated statistically significant relationships with most variables, except Machiavellianism. Specifically, EQ was negatively associated with psychopathy (r = −0.375, medium effect, p < 0.01) and narcissism (r = −0.437, medium effect, p < 0.01), and positively associated with recognition (r = 0.386, medium effect, p < 0.01), quality work (r = 0.432, medium effect, p < 0.01), meaningful work (r = 0.605, large effect, p < 0.01), influence (r = 0.541, large effect, p < 0.01), authenticity (r = 0.414, medium effect, p < 0.01), personal life (r = 0.442, medium effect, p < 0.01), growth and development (r = 0.462, medium effect, p < 0.01), and satisfaction (r = 0.395, medium effect, p < 0.01).

6.5 Hypothesis testing

To determine the direct relationship between the measured variables, regression analysis was conducted. The results are presented in Table 4.

Table 4. Hypothesis testing and direct associations.

^*p ≤ 0.05; ^**p ≤ 0.01; DTT, Dark Triad traits; EQ, emotional intelligence; SCS, subjective career success.

The results in Table 4 indicate that DTT have a statistically significant negative effect on EQ (β = −0.370; t = −7.874; p = 0.000^**), supporting Hypothesis 1. This finding suggests that higher levels of Machiavellianism, narcissism and psychopathy are associated with lower levels of emotional intelligence.

Furthermore, EQ had a positive and statistically significant effect on SCS (β = 0.523; t = 12.165; p = 0.000^**), supporting Hypothesis 2. This indicates that higher levels of EQ are associated with higher levels of SCS.

Lastly, DTT had a small but statistically significant negative effect on SCS (β = −0.104; t = −2.084; p = 0.038^*), supporting Hypothesis 3. This suggests that higher levels of DTT are associated with lower levels of SCS.

To investigate these relationships further, mediation analysis was conducted, with Machiavellianism, psychopathy, and narcissism as predictor variables, SCS dimensions as outcome variables, and EQ as the mediator (see results presented in Table 5). The reason being is that even though the dark triad traits are interrelated in terms of social aversion (viz. emotional coldness, exploitation and lack of empathy) forementioned are distinct personality traits (Brewer et al., 2019). Additionally, the SCSI measures seven dimensions, notably recognition, quality work, meaningful work, influence, authenticity, personal life, and growth and development. EQ as measured by the REQI was treated as a unidimensional construct (Nel et al., 2015). Hence, it is assumed that since the dark triad traits are distinct, there would be varying interactions with the different SCS dimensions and EQ would mediate the mention interactions in a different manner. Furthermore, though several hypotheses were tested, a basic mediation model was used with a single mediator. Moreover, a bias-corrected bootstrap method tends to accurately estimate power and type 1 error in complex models (Williams and Mackinnon, 2008).

Table 5. Supplementary mediation analyses results.

Mac, Machiavellianism; Psy, psychopathy; Nar, narcissism; Rec, recognition; QW, quality work; MW, meaningful work; Inf, influence; Aut, authenticity; PL, personal life; GAD, growth and development; Sat, satisfaction; EQ, emotional intelligence.

The results indicate that EQ did not mediate the association between Machiavellianism and any of the SCS dimensions.

In contrast, EQ significantly mediated the relationships between psychopathy and all dimensions of SCS. The mediation effects were predominantly full, as the direct effects became non-significant after controlling for emotional intelligence.

Specifically, EQ reduced the influence of psychopathy on recognition (β = −0.0945; 95% CI [−0.1510, −0.0504]), indicating that approximately 9.45% of the variance in this relationship is attributable to emotional intelligence. A similar pattern was observed for quality work (β = −0.0929; 95% CI [−0.1419, −0.0503]) and meaningful work (β = −0.1524; 95% CI [−0.2341, −0.0830]), according for approximately 9.29% and 15.24% of the variance, respectively.

EQ also fully mediated the relationships between psychopathy and influence (β = −0.125; 95% CI [−0.1936, −0.0702]), personal life (β = −0.1031; 95% CI [−0.1640, −0.0550]) growth and development (β = −0.1200; 95% CI [−0.1847, −0.0679]), and satisfaction (β = −0.0926; 95% CI [−0.1471, −0.0507]). These mediation effects accounted for approximately 12.56%, 10.31%, 12.00% and 9.26% of the variance, respectively. In all cases, the direct effects became non-significant after controlling for emotional intelligence.

Partial mediation was observed only for authenticity (β = −0.0953; 95% CI [−0.1545, −0.0493]), where the direct effect of psychopathy remained statistically significant (β = −0.0560; p < 0.05). This indicates that emotional intelligence reduces, but does not fully eliminate, the influence of psychopathy on this dimension.

A similar pattern emerged for narcissism. EQ significantly mediated the relationship between narcissism and all dimensions of SCS. Most relationships reflected full mediation, while partial mediation was observed for influence and authenticity.

Specifically, EQ fully mediated the relationships between narcissism and recognition (β = −0.0708; 95% CI [−0.1098, −0.0403]), quality work (β = −0.0705; 95% CI [−0.1044, −0.0430]), and meaningful work (β = −0.1204; 95% CI [−0.1747, −0.0740]), accounting for approximately 7.08%, 7.05%, and 12.04% of the variance, respectively.

EQ further fully mediated the relationships between narcissism and personal life (β = −0.0810; 95% CI [−0.1219, −0.0468]), growth and development (β = −0.0922; 95% CI [−0.1357, −0.0563]), and satisfaction (β = −0.0671; 95% CI [−0.0993, −0.0416]), explaining approximately 8.10%, 9.22%, and 6.71% of the variance, respectively.

Partial mediation was observed for influence (β = −0.0986; 95% CI [−0.1431, −0.0605]); p < 0.05) and authenticity (β = −0.0772; 95% CI [−0.1177, −0.0441], p < 0.01), indicating that EQ reduces, but does not fully eliminate, the influence of narcissism on these dimensions.

Collectively, these findings highlight the central role of EQ as a mediating mechanism linking DTT, particularly psychopathy and narcissism, to SCS. The negative indirect effects indicate that higher levels of EQ mitigate the detrimental influence of these traits, resulting in more favorable career-related outcomes.

7 Discussion

The purpose of this study was to investigate the relationship between DTT and SCS in the South African SMME sector, with a specific focus on the mediating role of EQ.

The findings indicate that DTT have a statistically significant negative influence on EQ, suggesting that individuals who exhibit higher levels of Machiavellianism, narcissism, and psychopathy are less likely to demonstrate EQ. This finding is consistent with prior research (Michels and Schulze, 2021; Walker et al., 2021) and aligns with Trait Activation Theory, which posits that personality traits influence behavior through their interaction with situational demands (Tett and Burnett, 2003). Furthermore, EQ was found to have a moderately strong positive influence on SCS. This supports prior findings that emotionally intelligent individuals are more likely to experience positive career success outcomes (Sharma and Tiwari, 2024). Given that SCS is inherently evaluative and relational, EQ plays a central role in shaping how individuals interpret and experience their careers. Although DTT demonstrated a statistically significant negative relationship with SCS, the relatively small effect size suggests limited practical significance. This aligns with prior reported mixed findings (Spurk et al., 2016; Volmer et al., 2016; Paleczek et al., 2018; Furnham and Treglown, 2021; Oldewage and Jonck, 2025).

Importantly, the mediation results provide deeper insight into these relationships. EQ emerged as a key explanatory mechanism, particularly for psychopathy and narcissism. The findings suggest that EQ buffers the negative interpersonal and emotional tendencies associated with these traits, thereby enhancing subjective career outcomes. However, the absence of mediation effects for Machiavellianism suggests that this trait may operate differently, possibly due to its more strategic and context-dependent nature. This finding highlights the importance of treating DTT as distinct constructs rather than a unified framework. Nevertheless, given the RMSEA value and the chi-square degrees of freedom ration, the findings presented are preliminary and should be interpreted with caution.

From a theoretical perspective, this study contributes to the literature by integrating Trait Activation Theory with EQ theory to explain how personality traits influence SCS. From a practical perspective, the findings suggest that SMMEs should prioritize the development of EQ to mitigate the negative effects of maladaptive personality traits and promote more sustainable career outcomes.

Practically, the findings of this study are particularly relevant within the South African SMME sector, which is characterized by resource constraints, high levels of uncertainty, interpersonal dependence, and dynamic work environments. Moreover, a notable characteristic of SMMEs is that these organizations often rely on close interpersonal interactions, flexible role structures, and collaborative relationships to maintain organizational performance and staff development. Therefore, within this setting, personality traits and emotional competencies may exert a stronger influence on workplace relationships and career experiences. As a result, the negative relationship between DT and EQ may be more significant in the SMME context, as reduced emotional awareness and interpersonal effectiveness may undermine teamwork, trust, and organizational cohesion within smaller organizational settings. Furthermore, the positive influence of EQ on SCS highlights the importance of emotional regulation and adaptive interpersonal skills in navigating the complex and often unstable nature of South African SMMEs. Consequently, the findings provide context-specific insight into the psychological factors shaping career experiences within emerging economy SMME environments. Ascribed to the Dark Triad traits being deemed condition-dependent coping mechanisms to solve difficulties in unpredictable, resource constrained and volatile conditions such as those indicative of the SMME sector findings could be uniquely attributable to SMME environments.

8 Limitations and future research

Several limitations should be considered when interpreting the findings of this study. First, the cross-sectional research design limits the ability to draw causal inferences and may negatively influence the external validity of the findings. As data were collected at a single point in time, causal relationships between DTT, EQ, and SCS cannot be definitively established. Future research should consider longitudinal designs to better capture the dynamic nature of these relationships and provide stronger evidence of causality.

Second, the use of convenience sampling may limit the generalizability of the findings beyond the specific sample and geographical context. Although the sample size was adequate for statistical analysis, future studies should employ probability sampling techniques and include more diverse samples across different regions and industries. This would enhance validity and allow for meaningful comparisons across various work contexts.

Third, reliance on self-report measures introduces the potential for common method bias and social desirability effects, which may have influenced participants' responses. Although statistical procedures were applied to assess common method variance, future research could incorporate multi-source data, such as supervisor or peer evaluations, to strengthen the robustness of the findings.

Furthermore, future studies could conceptualize EQ as an ability perspective rather than a trait, offering an alternative perspective on its role in career outcomes. In addition, while EQ was examined as a mediating variable, other psychological and organizational factors may also influence the relationship between DTT and SCS. Future research should explore additional mediators and moderators, such as organizational culture, leadership style, job demands, and psychological wellbeing, to develop a more comprehensive understanding of these relationships. Lastly, given the RMSEA value and the Chi-square degree of freedom ratio, the findings presented are preliminary and should be verified in future research endeavors.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Flavio Urbini, Libera Università Maria SS. Assunta, Italy

Reviewed by: Sadika Ismail, University of South Africa, South Africa

Mpho D. Magau, Stellenbosch University, South Africa