Research on the impostor phenomenon has demonstrated that feelings of intellectual fraudulence are shaped by a complex interaction of psychological, social, and contextual factors. Early conceptualisations emphasised developmental and interpersonal antecedents, while later scholarship foregrounded the role of gender norms, racialised identities, and organisational contexts. Within the South African context, recent research has highlighted the relevance of racial and transformation dynamics in shaping professional self-concept (Bowen & Botha, 2023; Folabit & Jita 2024; Ojetimi, 2025), although studies specific to HR remain scarce. This section reviews the literature thematically, focusing on psychological antecedents, gender dynamics, racial identity and stereotype threat, and organisational context, to establish the conceptual foundation for examining the impostor phenomenon among South African HR professionals.

This study adopted a positivist paradigm and a quantitative research design. The positivist approach was appropriate given the study’s objective to measure and compare observable patterns of the impostor phenomenon across demographic groups, consistent with established quantitative research on psychological constructs (Creswell & Creswell, 2017). A structured questionnaire was used to collect primary data, enabling the statistical analysis required to test the study’s hypotheses.

A voluntary sampling approach was implemented because of the absence of a comprehensive, accessible national database of HR professionals in South Africa. This method facilitated access to a geographically dispersed professional population through the South African Board of People Practitioners (SABPP) network. While practical and consistent with sampling methods used in prior impostor phenomenon research (e.g. Bravata et al., 2020), this approach introduces potential self-selection bias. Professionals who are members of SABPP and choose to respond may differ systematically from non-members or non-respondents in terms of professional engagement, self-awareness, or psychological vulnerability, which may affect the generalisability of the findings. This limitation is acknowledged.

The target sample size was approximately 300 participants, a threshold commonly deemed adequate for detecting medium effect sizes in group comparison studies using t-tests and analysis of variance (ANOVA), while allowing for demographic subgroup analysis (Cohen, 1988). The final sample of 317 respondents met this target.

The Clance Impostor Phenomenon Scale (CIPS) was selected as the measurement instrument because of its extensive validation and widespread use in professional and cross-cultural contexts (Chrisman et al., 1995; Mak et al., 2019). The scale’s 20-item structure, measuring three subscales (Discount, Luck, Fake), aligns well with the theoretical components of the impostor phenomenon. However, the cultural validity of the CIPS in the unique South African context, with its distinct socio-historical and linguistic landscape, has not been formally established. While the scale was deemed suitable for this exploratory study, the potential for cultural nuance in the interpretation of items related to ‘success’, ‘intelligence’, and ‘praise’ is recognised as a consideration for interpreting results and a recommendation for future validation studies. Table 2 shows the score criteria put forward by Clance (1985):

The professional registration body for the HR profession in South Africa, the SABPP, disseminated the invitation to participate in the study via email in their weekly mailers to all registered HR professionals. The requests for participation were published along with an intent statement, consisting of the study’s purpose and objectives. After approximately 6 weeks, the study received feedback from 320 respondents from different industry sectors in South Africa. Of the 320 questionnaires received, only 317 were regarded as usable. Ethical considerations to conduct this study were obtained through the University of the Western Cape Humanities and Social Sciences Research Ethics Committee with reference no. HS23/6/85. The study was guided by confidentiality, and all data remained solely with the researcher and were safely secured with a password-protected encryption known only by the researcher to avoid any data breaches. Informed consent to participate in the study was obtained by explaining the nature of the research to participants before displaying their willingness to participate. Respondents were provided with the researcher’s contact information in the informed consent form and encouraged to contact them with any questions, concerns, and clarifications during the completion of the questionnaire. In addition, respondents were allowed to withdraw from the study at any time, for any reason, without facing consequences. This aim was achieved by including a mandatory statement on the front page of the questionnaire, which respondents had to initial and agree to. The statement read as follows:

‘I understand that my participation is voluntary and that I am free to withdraw at any time without giving any reason and without there being any negative consequences. In addition, should I wish to withdraw, I may contact the lead researcher at any time to do so.’

All collected data were solely used for the purposes of this study.

The data analysis followed a structured approach, beginning with reliability assessment, followed by descriptive and inferential statistics aligned with the study’s hypotheses. An exploratory multiple regression analysis was also conducted to address the study’s broader aim of understanding potential predictors of the impostor phenomenon.

The reliability of the three subscales of the CIPS was evaluated by using Cronbach’s alpha. As reflected in Table 3, Table 4 and Table 5, the Luck (α = 0.792) and Fake (α = 0.871) subscales demonstrated satisfactory to excellent internal consistency. In contrast, the Discount subscale initially produced a weak reliability coefficient (α = 0.544). Further examination of item-total correlations revealed that items S1 and S19 showed poor alignment with the overall subscale. Once these items were removed, the reliability of the Discount subscale improved to an acceptable level (α = 0.740), leading to their exclusion from subsequent analyses. This refinement highlights the influence of contextual considerations on the interpretation of ‘discounting’ behaviours among South African HR professionals.

An exploratory factor analysis was performed on the revised Discount subscale, following the removal of the problematic items. The data met the requirements for factor analysis, as evidenced by a Kaiser–Meyer–Olkin (KMO) value of 0.500 and a statistically significant Bartlett’s test of sphericity (χ² = 134.659, df = 1, p < 0.001). In line with Kaiser’s criterion (as cited in Field, 2005), these results indicate that the correlation matrix was suitable for factor extraction. The analysis identified a single factor with an eigenvalue exceeding 1, explaining 79.43% of the total variance. All retained items loaded strongly on this factor (loadings > 0.50), thereby supporting the unidimensional structure of the refined Discount subscale.

The reliability analysis output for the discount subscale results is displayed in Table 3.

The Cronbach’s alpha coefficient for the Discount subscale (α = 0.544) fell below the conventional threshold of 0.70, indicating inadequate internal consistency. Item-total correlation analysis revealed that items S1 and S19 had correlations below the acceptable 0.30 threshold, suggesting that they did not align well with the overall subscale construct. Their removal increased reliability to α = 0.740, and thus, they were excluded from subsequent analyses to enhance measurement integrity. This refinement highlights the role of both contextual and language-related influences in shaping how discounting behaviours are understood within the South African HR professional environment.

After the exclusion of the identified items, an exploratory factor analysis was performed on the remaining indicators of the Discount subscale. The suitability of the data for factor analysis was supported by a KMO measure of 0.500, together with a statistically significant Bartlett’s test of sphericity (χ² = 134.659, df = 1, p < 0.001). In accordance with Kaiser’s guideline for factor extraction (Kaiser, 1974; as referenced in Field, 2005), these statistics confirm that the correlation matrix was appropriate for further analysis. The factor solution produced a single dominant component with an eigenvalue exceeding one, explaining 79.43% of the total variance. All retained items demonstrated strong associations with this factor, with loadings above 0.50, thereby supporting the single-factor structure of the revised subscale. The corresponding results are summarised in Table 6.

The intended meaning is clear: the Luck subscale achieved a reliability score of 0.792, which exceeds the acceptable threshold of 0.70. Therefore, the subscale demonstrates satisfactory internal consistency.

An exploratory factor analysis was conducted to examine the underlying structure of the Luck subscale. The adequacy of the data for factor analysis was confirmed by a KMO value of 0.782 and a statistically significant Bartlett’s test of sphericity (χ² = 399.793, df = 6, p = 0.001). In line with Kaiser’s guideline for acceptable sampling adequacy (as referenced in Field, 2005), these findings indicate that the correlation matrix was suitable for factor extraction. The analysis identified a single underlying factor with an eigenvalue exceeding one, explaining 62.91% of the total variance. All items loaded strongly onto this factor, with loadings greater than 0.50, supporting the unidimensional nature of the Luck subscale. The detailed factor analysis results are presented in Table 7.

The Fake subscale yielded a Cronbach’s alpha coefficient of 0.871, reflecting strong reliability. Therefore, the Fake subscale met the critical cutoff value of 0.70 for an acceptable internal consistency coefficient. Table 5 presents Cronbach’s alpha coefficient for the Fake subscale.

An exploratory factor analysis was undertaken to assess the construct structure of the Fake subscale. The suitability of the dataset for factor extraction was confirmed by a KMO measure of 0.898, alongside a statistically significant Bartlett’s test of sphericity (χ² = 927.226, df = 28, p = 0.001). In accordance with Kaiser’s recommended criteria for sampling adequacy (as cited in Field, 2005), these statistics demonstrate that the correlation matrix was appropriate for factor analysis. The results indicated a single-factor solution, with one component yielding an eigenvalue greater than one and explaining 51.41% of the total variance. All retained items showed strong associations with the extracted factor, with loadings exceeding 0.50, thereby confirming the unidimensional structure of the Fake subscale. The detailed outcomes of this analysis are presented in Table 8 (Omar, 2025).

Analysis of the 317 completed CIPS questionnaires reflected a notably high level of impostor feelings among South African HR professionals. Specifically, 83% of participants reported moderate or higher levels of the impostor phenomenon, encompassing 55% with moderate feelings, 25% with frequent feelings, and 3% with intense impostor experiences. The remaining 17% reported few impostor feelings. This widespread experience aligns with the study’s conceptual framework, suggesting that the high-stakes, advisory nature of HR work may create a context ripe for self-efficacy doubts and attributional biases, irrespective of objective competence.

In Table 9, similar response categories were combined for clarity: ‘not at all’ and ‘rarely’ form ‘NOT TRUE’, while ‘often’ and ‘very true’ form ‘TRUE’. Table 9 shows how participants responded to statements across the three CIPS subscales (Discount, Luck, and Fake).

As shown in Table 9, a notable proportion of respondents endorsed impostor-related beliefs, particularly within the Discount subscale, in which items related to downplaying success and avoiding recognition were frequently acknowledged. This pattern suggests that even among experienced HR professionals, internalisation of achievement remains a challenge. The Luck and Fake subscales also revealed meaningful levels of self-doubt and external attribution of success, reinforcing the multifaceted nature of the impostor phenomenon in professional settings.

Table 10 displays the results of an independent-samples t-test conducted to examine potential differences in the impostor phenomenon scores between self-identified male and female participants. The analysis indicated that scores did not differ significantly between males (M = 43.17, s.d. = 12.79) and females (M = 46.33, standard deviation [s.d.] = 13.82), t(317) = −1.85, p = 0.07 (two-tailed). The observed mean difference of 3.16 (95% confidence interval [CI]: −6.51 to 0.195) was minimal, with a corresponding effect size of η² = 0.001. The magnitude of this effect was evaluated by using Cohen’s (1988) guidelines for interpreting eta-squared values, where 0.01 represents a small effect, 0.06 a moderate effect, and 0.14 a large effect.

A one-way between-groups ANOVA was conducted to explore the impact of race on the impostor phenomenon. The results of the ANOVA are presented in Table 11. The race variable was divided into five categories, namely, black people, coloured people, Indian people, white people, and Chinese people. There was a statistically significant difference at the p < 0.05 level in the impostor phenomenon scores for the five race categories: F(7, 310) = 3.56, p = 0.001. While the analysis indicated statistical significance, the practical difference in mean scores across groups was moderate. The effect size, calculated by using eta-squared, was 0.08 (Cohen, 1988).

This result provides empirical support for the relevance of social identity theory and stereotype threat within the South African context. It suggests that historically hypothesised groups may face additional identity-related stressors that amplify feelings of intellectual fraudulence, as hypothesised in the conceptual framework.

An exploratory standard multiple regression was conducted to move beyond bivariate comparisons and explore potential predictors of the impostor phenomenon. The total CIPS score was entered as the dependent variable. Predictor variables included: gender (dummy-coded), race (dummy-coded with ‘white [person]’ as reference), age, and years of HR experience.

The regression model as a whole was significant, F(9, 307) = 3.92, p < 0.001, accounting for approximately 10.3% of the variance in the impostor phenomenon scores (Adjusted R² = 0.103). The detailed results are presented in Table 12, in which age emerged as a significant negative predictor (β = −0.22, p < 0.001), indicating that older HR professionals reported lower levels of IP. Furthermore, identifying as ‘black [person]’ (β = 0.18, p = 0.012) and ‘coloured [person]’ (β = 0.16, p = 0.022) were significant positive predictors compared to the ‘white [person]’ reference group, even after controlling for age and experience. Gender and years of experience were not significant predictors in this multivariate model.

Table 12 shows the results of multiple regression analysis predicting the impostor phenomenon scores.

The regression analysis substantiates and extends the ANOVA findings. The persistent significance of race, even when accounting for age and experience, strongly reinforces the social identity perspective, suggesting that racial identity itself, potentially through mechanisms of stereotype threat or belongingness uncertainty, is a key factor in IP experiences. The negative relationship with age may reflect the development of stronger self-efficacy beliefs or more stable attributional styles over time, as postulated by Bandura (1997). The non-significance of gender in the multivariate model further consolidates the finding that within this professional cohort, demographic identity factors related to race and age are more salient predictors of impostor feelings than gender is.

The finding that 83% of HR professionals experience at least moderate impostor feelings suggests that the impostor phenomenon is widespread within this occupational group. This prevalence aligns with the conceptual framework, which integrates self-efficacy and attribution theories to explain how professional contexts may trigger self-doubt. According to self-efficacy theory, individuals’ beliefs about their competence influence motivation, emotional regulation, and performance (Bandura, 1997). In HR roles characterised by advisory authority, ethical responsibility, and transformation mandates, professionals may perceive their responsibilities as challenging and evaluative, reducing self-efficacy despite objective achievements.

Attribution theory further explains the persistence of these feelings: HR professionals may discount success as luck or external circumstances while internalising failures, reinforcing impostor experiences over time. Responses on the Discount and Luck subscale exemplify this maladaptive attribution pattern, illustrating how perceived incompetence persists even in high-achieving individuals. These findings underscore the relevance of theoretically grounded constructs in explaining not only the prevalence but also the psychological mechanisms underpinning the impostor phenomenon in South African HR contexts.

Contrary to early feminist formulations of the impostor phenomenon (Clance & Imes, 1978), this study found no significant gender differences in impostor scores. This finding aligns with more recent research suggesting that impostor experiences are comparable across genders (Bravata et al., 2020; Naser et al., 2022).

From a theoretical perspective, self-efficacy theory provides an explanatory lens: professional socialisation and role expectations in HR may homogenise experiences of self-doubt across genders, particularly in high-stakes advisory positions where performance demands are equally salient. Attribution patterns may operate similarly across males and female, with both genders tending to internalise failure and discount success in line with impostor tendencies. Methodologically, it is also possible that social desirability or gendered expression of self-doubt influenced self-report responses, as suggested by Mann (2019). Overall, these findings support the conceptualisation of the impostor phenomenon as context-driven rather than inherently gendered.

Significant racial differences in impostor scores, with moderate effect sizes, highlight the relevance of social identity theory in the South African context. Human resource professionals from historically marginalised groups reported higher impostor feelings, reflecting the enduring impact of structural inequality and identity-based stressors on self-perception. Social identity theory posits that individuals’ sense of belonging and legitimacy is shaped by group membership (Turner et al., 1979). In post-apartheid South African workplaces undergoing transformation, professionals from marginalised racial groups may experience heightened self-doubt because of stereotype threat (Steele, 1997) and the ongoing negotiation of professional legitimacy.

Furthermore, self-efficacy may be undermined by structural inequities, with external barriers and subtle biases amplifying internalised feelings of inadequacy. These findings align with prior research linking racial discrimination to elevated impostor experiences (Bernard & Neblett, 2018; Cokley et al., 2013) and emphasise that demographic identity interacts with psychological mechanisms to shape the impostor phenomenon.

In interpreting findings through the combined lenses of self-efficacy, attribution, and social identity theories, this study provides a nuanced understanding of the impostor phenomenon among South African HR professionals. The high prevalence supports self-efficacy theory (Bandura, 1997) and attribution theories, showing that perceived competence and attributional tendencies are central to experiencing impostor feelings. The absence of gender differences suggests that context and role expectations, rather than demographic category alone, determine self-doubt, while racial disparities underscore how social identity and historical inequalities amplify impostor experiences (Steele, 1997). Together, these insights reinforce the study’s conceptualisation of the impostor phenomenon as a context-sensitive, psychologically mediated experience.

The high incidence of the impostor phenomenon among HR professionals has practical implications for organisational practice and employee well-being. Guided by theoretical insights, HR departments can implement interventions that target underlying psychological mechanisms.

Self-efficacy enhancement: Awareness workshops and coaching sessions to reinforce competence feedback and validate achievements.

Attributional reframing: Training to challenge maladaptive beliefs about luck and internalised failure, fostering realistic self-assessment.

Identity-sensitive support: Mentorship, sponsorship, and inclusive organisational policies to reduce identity-based stressors and stereotype threat.

Leadership development: Equipping managers to recognise impostor feelings and provide supportive feedback.

Through embedding these strategies within professional development frameworks, organisations can enhance resilience, psychological safety, and role efficacy, ultimately contributing to more inclusive workplaces.

This study demonstrates that the impostor phenomenon is highly prevalent among South African HR professionals, with 83% of participants reporting at least moderate feelings of self-doubt regarding their abilities and competencies (54 [17%] low, 174 [55%] moderate, 78 [25%] frequent, 11 [3%] intense). These findings highlight the need for HR departments to implement interventions that address both the psychological and the organisational aspects of impostor experiences.

Practical strategies include:

The study found no significant gender differences, suggesting that interventions can be applied broadly across men and women employees. However, racial disparities indicate that targeted support may be particularly beneficial for individuals from underrepresented groups, consistent with social identity theory and research linking structural inequality to heightened impostor experiences. Through integrating these strategies into professional development frameworks, organisations can foster resilience, self-efficacy, and psychological safety across HR teams (Omar, 2025).

The interpretation of the findings should be considered in light of several limitations. Firstly, the cross-sectional design constrains causal inferences regarding the causes and consequences of impostor feelings. Secondly, the quantitative approach identifies prevalence patterns and demographic associations but does not capture the nuanced lived experiences or organisational processes through which impostor feelings manifest, highlighting the need for complementary qualitative research.

Thirdly, the use of voluntary sampling introduces potential self-selection bias, as individuals with stronger impostor experiences or greater interest in the topic may have been more likely to participate. Fourthly, reliance on self-reported measures may result in common method bias, potentially inflating observed relationships between variables. Fifthly, although the CIPS has been used in prior South African studies (e.g. Bowen & Botha, 2023), its cultural and linguistic validity requires further verification through qualitative or mixed-methods investigations.

Finally, the unequal gender distribution in the sample may have limited the statistical sensitivity to detect subtle gender-based differences, so gender-related findings should be interpreted cautiously. Despite these limitations, the study provides valuable empirical insights into the impostor phenomenon among South African HR professionals and lays the foundation for future research using complementary methods and more diverse samples.

Future research should address the methodological limitations of this study. Longitudinal designs are recommended to examine the trajectories of the impostor phenomenon over time and its causal relationships with outcomes such as burnout, job satisfaction, and career progression (Folabit & Jita 2024). Incorporating multi-rater assessments, including supervisor or peer ratings, could reduce self-report bias and provide a more objective assessment of professional competence and performance.

Research should also investigate organisational moderators, such as leadership style, team climate, and diversity and inclusion policies, to understand their influence on impostor feelings (Fisher, 2019). Beyond methodology, future studies should adopt a context-specific approach, exploring the impostor phenomenon across diverse South African professional sectors, using mixed-methods designs to capture nuanced narratives of identity, belonging, and self-efficacy in workplaces undergoing transformation. Intervention studies evaluating the effectiveness of impostor-focused training, coaching programmes, or mentorship schemes are also needed to translate research into practical support mechanisms (Ojetimi, 2025).