Original Research

Measuring ethical leadership as a governance capability in South African government institutions: An exploratory study

Mariette Coetzee, Dorothee Burger-Snyman

Received: 09 Mar. 2026; Accepted: 23 June 2026; Published: 12 Aug. 2026

Copyright: © 2026. The Authors. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Persistent governance failures in South African public institutions have intensified scholarly and policy interest in ethical leadership as a governance capability associated with accountability, institutional integrity and organisational performance. However, dominant ethical leadership measures were primarily developed in private-sector environments and may not adequately reflect the realities of bureaucratic governance within South African government institutions.

Aim: This study aimed to develop a contextually grounded ethical leadership measurement scale for South African government institutions and to explore its underlying dimensional structure.

Setting: The study was conducted within two national government institutions in South Africa.

Methods: A quantitative cross-sectional research design was adopted. An Ethical Leadership Questionnaire was developed from established ethical leadership and governance theories and administered to permanently employed public-sector staff members. An exploratory factor analysis (EFA) was employed to examine the latent structure of ethical leadership in a public-sector governance context.

Results: The findings supported a multidimensional ethical leadership structure comprising Character, Influence and Morality. The dimensions reflected dispositional ethical attributes, behavioural enactment of ethical standards and normative ethical orientation, respectively.

Conclusion: The study provides preliminary evidence that ethical leadership in government institutions may comprise distinguishable but interrelated governance capabilities.

Contribution: This study contributes to governance and leadership scholarship by developing a contextually grounded, exploratory measure of ethical leadership for South African government institutions.

Keywords: ethical leadership; governance capability; morality; character; influence; South African government institutions.

Introduction

South Africa continues to confront persistent governance failures characterised by procurement irregularities, audit regressions, weak internal controls and declining public trust in state institutions (Auditor-General South Africa [AGSA] 2024; Corruption Watch 2024). These governance challenges have economic and managerial implications, including fiscal inefficiencies, service delivery failures, resource misallocation and increased compliance costs (AGSA 2024; Organisation for Economic Co-operation and Development [OECD] 2024). Within organisational and public administration scholarship, such failures are increasingly linked to leadership deficits affecting accountability systems, ethical climate formation and organisational effectiveness (Abdi, Hashi & Latif 2024; Aunin et al. 2024; Mahohoma & Sihlangu 2024). Ethical leadership has therefore become central to discussions concerning governance quality, institutional accountability and public-sector management effectiveness.

Despite growing recognition of ethical leadership as a governance mechanism, uncertainty persists about how it should be conceptualised and measured within South African government institutions. Existing ethical leadership instruments were largely developed in private-sector or Western organisational environments and may insufficiently reflect highly regulated bureaucratic contexts characterised by legislative compliance, political oversight and formal accountability structures (Brown, Treviño & Harrison 2005; Kalshoven, Den Hartog & De Hoogh 2011; School of Government 2022). This contextual mismatch presents an important measurement gap warranting further investigation.

Ethical leadership is commonly understood as leadership grounded in moral principles and demonstrated through responsible conduct, ethical decision-making, communication and behavioural influence (Brown et al. 2005; Mayer, Kuenzi & Greenbaum 2010). Although often conceptualised through normative ethics, ethical leadership also incorporates applied ethics, which concerns the enactment of ethical principles within organisational settings, and descriptive ethics, which examines how ethical behaviour is perceived in practice (Lewis & Gilman 2012; Rossouw & Van Vuuren 2017). Within government institutions, ethical leadership extends beyond interpersonal conduct to include governance stewardship, accountability enforcement and the formation of an ethical climate.

Dominant ethical leadership instruments, such as the Ethical Leadership Scale (Brown et al. 2005) and Ethical Leadership at Work Questionnaire (Kalshoven et al. 2011), were developed primarily outside South African public-sector governance environments. While psychometrically robust, these instruments may insufficiently differentiate between ethical character, moral orientation and ethical influence within bureaucratic systems where accountability, compliance and consequence management are central. Given South Africa’s governance mandates and accountability pressures, an exploratory scale development approach was considered appropriate to examine whether ethical leadership dimensions manifest distinctly within this context.

Theoretical perspectives provide a basis for differentiating ethical leadership dimensions. The moral person–moral manager framework distinguishes between leaders’ personal moral attributes and active ethical management (Mayer et al. 2010), while governance-oriented approaches emphasise accountability, stewardship and ethical culture formation (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017). Across scholarship, recurring themes include ethical character, moral judgement and ethical influence processes (Brown & Treviño 2006; Mihelič, Lipičnik & Tekavčič 2010). Within South African public institutions, governance failures may therefore reflect not only personal immorality but also weak ethical influence, inconsistent enforcement and limited managerial follow-through (AGSA 2024; Corruption Watch 2024).

The primary objective of this study was to develop a contextually grounded Ethical Leadership measurement scale for South African government institutions and to explore its dimensional structure using EFA. Secondary objectives included examining whether Character, Morality and Influence emerge as distinguishable leadership dimensions within governance settings, assessing internal consistency and contributing a measurement foundation for future studies examining ethical leadership, governance effectiveness and institutional performance outcomes.

The study contributes to ethical leadership scholarship by exploring ethical leadership as a multidimensional governance capability within South African public institutions and providing a foundation for future examination of ethical leadership, accountability and institutional performance.

The article proceeds with a review of the literature and theoretical foundations, followed by the research methodology, findings and discussion and concluding sections on implications, limitations, recommendations and conclusions.

Literature review and theoretical framework

Ethical leadership in governance contexts

Ethical leadership has gained increasing attention in organisational and governance scholarship because of its links with accountability, trust, ethical workplace climates and effective institutional functioning (Abdi et al. 2024; Aunin et al. 2024; Mayer et al. 2010). Ethical leadership is commonly understood as leadership grounded in moral principles and demonstrated through ethical conduct, decision-making and behavioural influence (Brown et al. 2005; Lewis & Gilman 2012). Within governance settings, ethical leadership extends beyond interpersonal conduct to include accountability enforcement, stewardship, fairness and ethical oversight (Crews 2015; Rossouw & Van Vuuren 2017).

For purposes of this study, Character refers to leadership behaviours associated with integrity, trustworthiness, fairness and ethical consistency; morality refers to ethical judgement, responsibility and normative decision-making; and influence refers to behavioural enactment through communication, accountability, role modelling and ethical reinforcement. These dimensions informed the conceptualisation of ethical leadership behaviour within South African government institutions.

Within the South African public sector, constitutional principles emphasise accountability, transparency and professional ethics (School of Government 2022), while reforms such as the Department of Public Service and Administration (DPSA) ethics officer framework seek to operationalise these principles (DPSA 2024). However, recurring audit regressions and internal control failures suggest continuing implementation challenges (AGSA 2024; Corruption Watch 2024). These governance conditions raise an important question: Whether ethical leadership in public institutions requires not only moral character but also the ability to institutionalise ethical standards behaviourally.

Recent South African public-sector scholarship similarly suggests that ethical leadership is increasingly viewed as a determinant of governance effectiveness and service delivery outcomes. Empirical studies within local government contexts report associations between ethical leadership, accountability, organisational trust and service performance (Mahohoma & Sihlangu 2024; Mpontshane & Holtzhausen 2024). These findings reinforce the need for contextually grounded ethical leadership measurement within South African governance institutions, particularly where accountability failures continue to undermine public confidence.

Conceptual foundations of ethical leadership

Ethical leadership is rooted in broader ethical philosophy and leadership theory. Ethics concerns principles of right and wrong that guide moral judgement (Mihelič et al. 2010; Rossouw & Van Vuuren 2017), while integrity reflects consistency between espoused values and observable behaviour, reinforcing trust and ethical credibility within leadership relationships (Rossouw & Van Vuuren 2017; Treviño, Brown & Hartman 2003). Leadership scholarship integrates these moral foundations with influence processes that shape organisational behaviour.

The moral person–moral manager framework distinguishes between leaders’ personal moral attributes (e.g. honesty, fairness, integrity) and their active management of ethics through communication, modelling and reinforcement (Mayer et al. 2010). This distinction is foundational because ethical leadership involves both moral disposition and behavioural enactment, although many empirical studies still treat ethical leadership as a unified construct.

Similarly, value-based leadership perspectives emphasise prudence, moral persistence and value alignment (Mihelič et al. 2010). Governance-oriented scholarship conceptualises ethical leadership through accountability, relationship-centredness, stewardship and ethical congruence within institutional systems (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017).

Although these perspectives converge around integrity, fairness, accountability and ethical influence, they differ in emphasis and dimensional structure. Notably, they do not consistently distinguish between ethical character, moral orientation and ethical influence. This lack of dimensional precision limits theoretical clarity, particularly in governance environments where bureaucratic systems may weaken the translation of moral intent into institutional practice.

Ethical leadership as influence and institutional agency

Contemporary scholarship increasingly recognises ethical leadership as relational and institutional. Ethical leaders influence behaviour through role modelling, accountability systems and ethical reinforcement (Babalola, Stouten & Euwema 2016; Freeman & Stewart 2006; Lewis & Gilman 2012). However, governance research suggests that personal integrity alone may be insufficient for effective ethical leadership if leaders lack the authority or organisational capability to institutionalise ethical standards consistently (Al Halbusi et al. 2024; Mahohoma & Sihlangu 2024).

South African governance analyses continue to report weaknesses in oversight, consequence management and accountability (AGSA 2024; Corruption Watch 2024). Ethical leadership may therefore depend not only on moral disposition but also on leaders’ ability to operationalise ethical norms behaviourally within institutional systems.

Recent South African governance analyses (AGSA 2024; Corruption Watch 2024; National Corruption Risk Management and Prevention Framework [Draft] 2025) point to recurring weaknesses not necessarily attributable to individual immorality but to ineffective enforcement, weak oversight and inconsistent consequence management. This suggests that ethical influence – the behavioural and managerial capacity to embed ethical norms – may operate independently of moral character.

More recent scholarship further suggests that ethical leadership contributes to institutional effectiveness through both behavioural modelling and governance-oriented accountability processes. For example, Aunin et al. (2024) found that perceived ethical leadership positively influences public-sector outcomes, while Abdi et al. (2024) demonstrated relationships between ethical leadership, organisational politics and public-sector performance. Ethical leadership therefore increasingly appears not merely as a moral ideal but also as an operational governance capability influencing institutional effectiveness.

Empirical studies on ethical leadership in governance contexts

Empirical scholarship increasingly links ethical leadership to organisational trust, employee engagement, accountability and governance effectiveness. In organisational settings, ethical leadership has been associated with improved organisational citizenship behaviour, reduced misconduct and stronger ethical climates (Hassan, Wright & Yukl 2014; Mayer et al. 2010). More recent studies further demonstrate relationships between ethical leadership, employee trust, organisational resilience and performance-related outcomes (Abuzaid et al. 2024; Al Halbusi et al. 2024).

Within governance and public-sector environments, ethical leadership has similarly been linked to accountability, institutional legitimacy and service performance. For example, Mahohoma and Sihlangu (2024) found that ethical leadership positively influenced governance effectiveness within South African municipal environments, while Mpontshane and Holtzhausen (2024) reported associations between ethical leadership, institutional trust and public-sector accountability. Aunin et al. (2024) further demonstrated positive relationships between ethical leadership and organisational effectiveness in governance-oriented institutions.

Despite these contributions, prior empirical work has predominantly examined ethical leadership as a unified construct and has focused largely on behavioural outcomes rather than exploring whether underlying ethical leadership dimensions may operate differently within bureaucratic governance contexts. Limited scholarship has examined whether ethical character, moral orientation and ethical influence emerge as empirically distinguishable dimensions within South African government institutions. This gap informed the exploratory orientation of the present study.

Comparative dimensions of ethical leadership

Across the literature, three recurring clusters of ethical leadership dimensions can be synthesised:

• Character-based dimensions – integrity, honesty, courage and trustworthiness (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017).
• Influence or managerial dimensions – communicating ethical expectations, reinforcing standards, modelling behaviour and governance enforcement (Crews 2015; Mayer et al. 2010).
• Moral or normative dimensions – fairness, compassion, stakeholder orientation, commitment to the broader good (Freeman & Stewart 2006; Mihelič et al. 2010).

The present study integrates these clusters into a three-factor model:

• Character (dispositional ethical traits),
• Morality (normative ethical orientation and fairness),
• Influence (behavioural enactment and reinforcement of ethical standards).

This structure builds upon existing theoretical foundations but introduces an explicit analytical distinction between internal moral disposition and external ethical enactment. By empirically examining whether these dimensions emerge as distinct latent constructs within government institutions, the study addresses a key theoretical gap: Whether ethical leadership should be conceptualised as a unified moral construct or as a multidimensional governance capability.

Table 1 summarises key ethical leadership perspectives informing the conceptualisation of this study. Despite differences in emphasis, recurring themes include ethical character, moral judgement, accountability and behavioural influence, supporting the inclusion of Character, Morality and Influence in the measurement instrument.

Conceptual framework of the study

The conceptual framework integrates Character, Morality and Influence as interrelated dimensions of ethical leadership behaviour in governance settings. Drawing on social learning theory, moral person–moral manager theory and governance scholarship, the framework assumes that ethical leadership reflects both personal moral attributes and leaders’ ability to enact ethical standards behaviourally within institutional environments. Figure 1 presents the conceptual structure informing the exploratory measurement model.

FIGURE 1: Conceptual framework of ethical leadership as a governance capability.

Methods

Research paradigm

This study was positioned within a positivist research paradigm, which assumes that organisational phenomena may be measured objectively and analysed using empirical techniques to identify patterns and relationships (Bryman et al. 2014). A positivist orientation was considered appropriate because the study sought to develop and examine the dimensional structure of an Ethical Leadership measurement instrument using statistical procedures. Consistent with quantitative scale development research, the paradigm supported systematic measurement, factor identification and reliability assessment within a public-sector governance context.

Research approach and design

The study utilised a quantitative, cross-sectional design to examine the dimensional structure of an Ethical Leadership measurement instrument within two South African government institutions. The study was positioned as an exploratory measurement investigation, focusing on scale development and dimensional identification rather than confirmatory validation. The study was classified as exploratory because the dimensional structure of ethical leadership behaviour within South African government institutions had not previously been empirically established in the present form. Exploratory factor analysis (EFA) was therefore used to investigate latent dimensionality, identify emerging factor patterns and assess whether theoretically informed ethical leadership dimensions manifested empirically within a governance context (Hair et al. 2019; Kline 2016). A cross-sectional design was considered appropriate because data were collected at a single point in time to examine employee perceptions of ethical leadership within governance settings (Hair et al. 2019).

Item generation was informed by ethical leadership, governance, organisational ethics and leadership scholarship, including ethical leadership models (Brown et al. 2005; Kalshoven et al. 2011; Mayer et al. 2010), governance-oriented ethics scholarship (Lewis & Gilman 2012; Rossouw & Van Vuuren 2017), value-based leadership approaches (Mihelič et al. 2010) and public-sector governance perspectives (Crews 2015; ed. Lazenby 2024). Recent public-sector scholarship further informed contextual relevance (Abdi et al. 2024; Aunin et al. 2024; Mahohoma & Sihlangu 2024; Mpontshane & Holtzhausen 2024). Items were formulated to reflect observable ethical leadership behaviours relevant to governance settings rather than abstract ethical principles alone. Particular attention was given to contextual alignment with bureaucratic accountability systems and regulatory oversight structures.

A four-point Likert scale was used, with response options ranging from 1 (totally disagree) to 4 (totally agree). The forced-choice format was selected to reduce central tendency bias in governance research, where neutral responses may obscure evaluative perceptions. Given the cross-sectional design, no causal inferences are implied. The primary objective was to investigate the latent dimensional structure of ethical leadership through EFA.

Data collection procedures

Both secondary and primary data collection procedures informed the study. Secondary data comprised an extensive review of ethical leadership, governance and leadership theory literature, which informed item generation and theoretical positioning of the Ethical Leadership questionnaire (Crews 2015; ed. Lazenby 2024; Mayer et al. 2010; Mihelič et al. 2010; Rossouw & Van Vuuren 2017).

Primary data were collected using a structured, self-administered questionnaire (see Table 5 for the Ethical Leadership items included in the questionnaire), distributed to permanently employed staff members across two national government institutions. Participants completed the instrument anonymously to encourage honest responses and minimise social desirability bias. Data collection occurred at a single point in time, consistent with the study’s cross-sectional design.

Instrument development, content and face validity

Prior to data collection, a structured expert review was conducted to assess content validity. Three leadership and governance specialists, together with one statistician, evaluated item clarity, theoretical alignment and contextual appropriateness relative to Character, Morality and Influence. Ambiguous or redundant items were revised or removed, strengthening theoretical coherence and contextual relevance before empirical analysis. The reviewers further evaluated whether items adequately represented observable ethical leadership behaviours relevant to governance contexts, including integrity, ethical communication, accountability, fairness and managerial ethical influence. This process strengthened conceptual alignment between the theoretical framework and questionnaire content. Although a formal Content Validity Index (CVI) was not computed, the structured expert review process enhanced theoretical coherence and contextual relevance prior to empirical analysis.

For measurement purposes, Character referred to observable leadership behaviours reflecting integrity, fairness, trustworthiness, supportiveness and ethical consistency. Morality was reflected in behaviours associated with ethical judgement, fairness, responsibility and a normative ethical orientation. Influence referred to behavioural enactment through ethical communication, accountability enforcement, role modelling and managerial reinforcement of ethical standards. These constructs informed item wording and subsequent exploratory dimensional assessment.

Population and sampling

The target population comprised permanently employed staff members across two national government institutions in South Africa. A total of 1296 employees were invited to participate using simple random sampling to enhance representativeness and reduce selection bias. The final sample consisted of 274 usable responses.

The inclusion of two national government institutions was considered appropriate for the exploratory purpose of the present study. The objective was not to produce a definitive standardised instrument for immediate application across the entire public sector but rather to conduct an initial examination of the dimensional structure of ethical leadership behaviour within governance-oriented institutional settings. National government institutions were considered suitable because of their formal accountability mandates, regulatory oversight responsibilities and bureaucratic governance structures, which align closely with the conceptual focus of ethical leadership as a governance capability (DPSA 2024; School of Government 2022).

The achieved sample satisfied recommended respondent-to-item ratios for EFA. With 30 initial items, the respondent-to-item ratio approximated 9:1, exceeding the commonly suggested minimum of 5:1 (Hair et al. 2019). The suitability of the sample for EFA was further assessed using the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy.

Although broad institutional invitations and assurances of anonymity were used to mitigate non-response bias, it could not be entirely excluded. Generalisation beyond similar national public-sector contexts should therefore be undertaken cautiously.

Measuring instrument

The measuring instrument was developed to capture observable ethical leadership behaviours and governance-related leadership practices relevant to South African public-sector institutions. Respondents were not asked to evaluate the validity, accuracy or completeness of questionnaire items. Rather, participants assessed their perceptions of observable ethical leadership behaviours demonstrated by leaders within their governance environments. The questionnaire comprised two sections. Section A captured participants’ demographic information, while Section B consisted of a 30-item Ethical Leadership scale.

Although institutional performance outcomes were not directly measured in the present study, the instrument was developed to establish a behavioural measurement foundation capable of supporting future examination of how ethical leadership behaviour may relate to governance effectiveness, accountability and public-sector performance outcomes. Reverse-scored items were included to reduce acquiescence bias and detect inattentive responding.

Following EFA, a three-factor structure emerged comprising:

• Character (18 items)
• Influence (9 items)
• Morality (3 items)

Although the Morality factor consists of three items, this is methodologically acceptable in exploratory research when items demonstrate conceptual coherence and adequate factor loadings. Research indicates that well-defined constructs can be represented by as few as three indicators, particularly in early-stage scale development (Hair et al. 2019; Kline 2016). Future research employing confirmatory factor analysis (CFA) may refine item representation and assess model fit more rigorously.

The unequal distribution of items across factors reflects empirical loading patterns rather than predetermined parity. Item retention was guided by statistical loading criteria and theoretical coherence.

Data analysis procedures

Data analysis was conducted using the Statistical Package for the Social Sciences (SPSS), Version 22. The analysis concentrated on EFA to examine the underlying dimensional structure and reliability of the Ethical Leadership scale. Given the exploratory nature of the study, a CFA was not conducted.

The analytical procedure followed established guidelines (Hair et al. 2019):

• Evaluation of factorability using the KMO measure together with Bartlett’s test of sphericity.
• Extraction using principal axis factoring (PAF), appropriate for identifying underlying latent constructs.
• Determination of factor retention using the Kaiser criterion (eigenvalues > 1.0) in conjunction with theoretical interpretability.
• Examination of the pattern matrix with a factor loading cut-off of 0.30, while prioritising loadings above 0.40 where possible.
• Reliability assessment using Cronbach’s alpha (threshold ≥ 0.70).

Although an orthogonal Varimax rotation was implemented to improve interpretability, the study acknowledges the theoretical interrelatedness of ethical leadership dimensions. As the study’s objective was exploratory dimensional identification rather than structural model testing, orthogonal rotation was deemed acceptable at this stage. Future research may employ oblique rotation and structural equation modelling to further examine construct interrelationships.

The initial extraction suggested four factors. However, the fourth factor (‘Sense of Responsibility’) lacked conceptual distinctiveness and demonstrated cross-loadings. Guided by theoretical coherence and interpretability, a parsimonious three-factor solution was retained. This decision reflects best practice in exploratory scale development, where statistical criteria are interpreted alongside substantive theory.

Internal consistency reliability coefficients ranged from 0.777 to 0.962, exceeding recommended thresholds (Hair et al. 2019). While high alpha values may suggest item homogeneity, potential redundancy within the Character dimension is acknowledged as an area for refinement in future confirmatory research.

Ethical considerations

Ethical clearance was obtained from the University of South Africa’s Human Resource Management (HRM) Ethics Review Committee (2020_HRM_010). Participation was voluntary and anonymous. Data were stored securely and accessed solely for research purposes.

Results

This section presents the findings of the exploratory investigation into the dimensional structure of ethical leadership within South African government institutions. As the study was exploratory, the emphasis was placed on dimensional identification rather than confirmatory model validation.

Descriptive statistics

Descriptive statistics were computed to examine central tendency, dispersion and distributional characteristics of the three ethical leadership dimensions. Table 2 summarises the descriptive statistics for each factor, including the mean, standard deviation, skewness and kurtosis.

The means were interpreted relative to the midpoint of the four-point Likert scale (2.5). Character (M = 2.764; SD = 0.734) and Morality (M = 2.923; SD = 0.819) exceeded the midpoint, indicating generally favourable perceptions of leaders’ ethical character and normative moral orientation. In contrast, Influence (M = 2.373; SD = 0.570) fell below the midpoint, suggesting comparatively weaker perceptions of leaders’ ability to actively model, enforce and institutionalise ethical standards.

The standard deviations indicate moderate variability across responses, particularly for Morality, suggesting perceptual differences across departments or managerial units. The skewness and kurtosis values for all dimensions were within the acceptable range of ± 1, suggesting approximate normality and supporting the suitability of the data for factor analysis.

Substantively, the descriptive pattern suggests that while leaders are perceived as possessing moral intent and integrity, their capacity to translate ethical values into organisational influence may be comparatively limited. This pattern is noteworthy because it suggests a potential tension between leaders being perceived as ethically well-intentioned and their ability to consistently translate ethical intentions into managerial influence, accountability enforcement and behavioural reinforcement within governance systems.

These descriptive findings provide preliminary support for the proposition that ethical leadership within government institutions may not operate as a singular construct but instead reflect distinguishable ethical orientations requiring further structural investigation.

Exploratory factor analysis

The suitability of the data for factor analysis was evaluated using the KMO measure of sampling adequacy and Bartlett’s test of sphericity. The results, presented in Table 3, indicate that the dataset was adequate for EFA.

The KMO value of 0.955 indicates excellent sampling adequacy (Kaiser 1970), suggesting that the correlations among variables are sufficiently compact to yield reliable factors. Bartlett’s test of sphericity yielded a statistically significant result (p < 0.001), indicating that the correlation matrix differed significantly from an identity matrix and supporting the suitability of the data for factor analysis.

Factor extraction and variance explained

Principal axis factoring was selected as the extraction technique because it is well suited to identifying latent constructs underlying observed variables. The Kaiser criterion (eigenvalues > 1.0) was applied to determine factor retention (Feinberg, Kinnear & Taylor 2013; Field 2013; Girden & Kabacoff 2011; Pallant 2011). Table 4 provides information on the eigenvalues and variances.

The initial factor extraction identified four factors with eigenvalues greater than 1.0, collectively explaining 62.878% of the total variance. However, inspection of the pattern matrix and theoretical alignment indicated that the fourth factor (‘Sense of Responsibility’) lacked conceptual distinctiveness and was defined by items exhibiting cross-loadings. Consequently, theoretical coherence and interpretability informed the decision to retain the more parsimonious three-factor structure. Some degree of conceptual overlap across ethical leadership dimensions was anticipated given the theoretically interrelated nature of leadership ethics, governance behaviour and accountability processes.

After evaluating both statistical criteria and conceptual coherence, a three-factor solution was retained. The refined three-factor structure accounted for 56.613% of cumulative variance after extraction. The retained factors were interpreted as:

• Factor 1: Character
• Factor 2: Influence
• Factor 3: Morality

Pattern matrix interpretation

The factor pattern matrix, which presents the regression coefficients for each item (Howard 2016) and how the items load onto the various Ethical Leadership factors, is presented in Table 5.

Factor loadings ≥ 0.30 were considered meaningful (Field 2013; Howard 2016). Most retained items demonstrated strong loadings above 0.50, indicating substantive relationships with their respective factors:

• Character (18 items) reflects integrity, fairness, empowerment, adaptability and relational trust.
• Influence (9 items) captures enforcement behaviours, managerial dominance, political conduct and result-driven pressure.
• Morality (3 items) reflects high moral standards, ethical responsibility and rule enforcement orientation.

Collectively, the factor structure provides preliminary support for ethical leadership as a multidimensional governance capability within public institutions. The findings suggest that respondents differentiate between dispositional ethical characteristics, behavioural enforcement processes and normative moral orientation, reinforcing the conceptual rationale underlying the study.

The reverse-scored Item 18 loaded negatively on Influence (−0.479), confirming conceptual alignment with that factor after score reversal. Some overlap across dimensions was expected, given the relational nature of ethical leadership behaviours and the interconnectedness of governance responsibilities.

Although Character contained more items than the other dimensions, its strong loading pattern suggests a broad but coherent dispositional construct. Morality, despite comprising only three items, demonstrated acceptable theoretical coherence and reliability (Hair et al. 2019; Kline 2016).

Reliability

The internal consistency reliability of the factors identified through EFA was assessed using Cronbach’s alpha coefficients. A threshold of 0.70 was adopted as the minimum acceptable reliability level for this study, although Bryman et al. (2014) regard a coefficient alpha of 0.6 or higher as acceptable. Table 6 presents Cronbach’s alpha values for the Ethical Leadership factors.

Cronbach’s alpha coefficients ranged from 0.777 to 0.962. Character demonstrated very high internal consistency (α = 0.962), indicating strong inter-item correlation. While this suggests homogeneity, potential redundancy within this dimension is acknowledged and may warrant future item reduction. Influence (α = 0.777) and Morality (α = 0.903) demonstrated satisfactory reliability given their respective item counts. Notably, the Morality factor achieved high reliability despite comprising only three items, supporting its internal coherence.

As the study was explicitly exploratory, these findings should be interpreted as an initial structural identification rather than as confirmatory validation. Future research employing independent samples and CFA is recommended to further test the stability and generalisability of the identified factor structure.

Overall, the exploratory findings provide preliminary evidence that ethical leadership within South African government institutions may comprise distinguishable but interrelated dimensions associated with ethical character, moral orientation and behavioural influence. These findings provide the basis for the discussion that follows, where the dimensions are interpreted relative to existing leadership and governance scholarship.

Discussion

This study aimed to develop and analyse the dimensional structure of an Ethical Leadership scale within two South African government institutions. Exploratory factor analysis identified three interrelated dimensions – Character, Morality and Influence – providing preliminary support for a multidimensional understanding of ethical leadership within a governance context. The findings further suggest that ethical leadership behaviour in public institutions may reflect distinguishable but interrelated dimensions associated with dispositional ethics, normative judgement and behavioural enactment.

Alignment with prior leadership scholarship

The prominence of Character as a broad dispositional factor aligns with virtue-based and character-focused leadership scholarship. Leadership character encompasses traits such as integrity, compassion, service and forgiveness (Belcher 2024). The strong loading of trust, fairness, empowerment and adaptability items onto Character suggests that respondents perceive ethical leadership as embedded in leaders’ value orientation and moral identity. This aligns with Treviño et al. (2003) findings that employees often evaluate ethical leadership first through perceived integrity and credibility. Empirical evidence further shows that consistent character-based behaviour enhances employee commitment and constructive conduct (Abuzaid et al. 2024; Serang et al. 2024). The prominence of Character within the exploratory structure further aligns with contemporary public-sector scholarship suggesting that integrity-based leadership is increasingly viewed as a prerequisite for institutional trust, governance legitimacy and organisational accountability (Aunin et al. 2024; Serang et al. 2024). Within governance environments characterised by accountability pressures, ethical leadership may therefore begin with leaders’ dispositional credibility before extending into institutional enforcement practices.

The emergence of Morality as a distinct normative dimension reflects justice-based and fairness-oriented leadership perspectives. Organisational justice research demonstrates that fairness, procedural consistency and ethical decision-making shape trust and organisational climate (Colquitt et al. 2001). Kalshoven et al. (2011) similarly identified fairness and ethical guidance as distinguishable ethical leadership behaviours. Qasim and Laghari (2025) highlight morality, fairness and trust-building as core drivers of ethical leadership effectiveness. Furthermore, Santiago-Torner, González-Carrasco and Miranda Ayala (2024) argue that moral intensity influences both follower ethical behaviour and emotional responses. The differentiation of Morality from Character in this exploratory model suggests that employees may distinguish between a leader’s internal virtues and their explicit commitment to normative fairness standards, supporting multidimensional conceptualisations advanced by Rossouw and Van Vuuren (2017). This distinction is particularly important in public institutions where procedural fairness, equitable treatment and accountability obligations are central to governance legitimacy and employee trust. However, the comparatively smaller item representation of Morality suggests that ethical judgement may be perceived as less behaviourally visible than broader leadership character within institutional settings.

The Influence dimension reflected behavioural enactment through communication, ethical reinforcement and accountability processes. Its emergence as a separate factor suggests that respondents distinguished between leaders’ personal morality and their ability to operationalise ethical standards behaviourally within institutional systems. This finding reinforces arguments that ethical leadership requires not only ethical intent but also visible ethical management and enforcement practices (Al Halbusi et al. 2024; Brown & Treviño 2006; Mayer et al. 2010).

The exploratory findings converge with virtue ethics (Peterson & Seligman 2004), authentic leadership (Walumbwa et al. 2008), organisational justice (Colquitt et al. 2001), social learning theory (Bandura 1986) and multidimensional ethical leadership models (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017). The structure identified in this study, therefore, resonates theoretically across multiple leadership traditions.

Collectively, the findings support multidimensional understandings of ethical leadership and suggest that ethical leadership in governance environments may depend on the interaction between ethical character, moral judgement and behavioural influence processes (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017).

Operationalising the constructs in a governance context

Given South Africa’s persistent governance challenges (AGSA 2024; Corruption Watch 2024), the practical relevance of these constructs lies in how they are operationalised within institutional systems.

Operationalising character

Character should be embedded in recruitment, promotion and succession planning processes that assess integrity, accountability, track records and value congruence. Behavioural integrity research emphasises alignment between espoused values and enacted behaviour as critical to credibility (Simons 2002). In public institutions, this may include integrity screening, transparent appointment criteria and ethics-focused leadership development. Reflective ethics training and scenario-based exercises can strengthen moral courage and consistency (Belcher 2024).

Operationalising morality

Morality can be institutionalised through structured ethical decision-making frameworks and procedural justice systems. Organisational justice scholarship demonstrates that consistent, transparent processes enhance legitimacy (Colquitt et al. 2001). In governance environments, this entails impartial procurement processes, transparent performance management systems and equitable grievance procedures. Leaders must communicate and consistently apply fairness standards, reinforcing trust and reducing perceptions of bias (Qasim & Laghari 2025).

Operationalising influence

Influence requires visible enforcement and consequence management. Social learning theory (Bandura 1986) suggests that employees internalise norms when behaviour is consistently rewarded or sanctioned. Brown and Treviño (2006) emphasise that ethical leadership credibility depends on the consistent reinforcement of standards. In public institutions, this translates into aligning performance management systems, disciplinary procedures and oversight mechanisms with ethical expectations. As governance analyses indicate (AGSA 2024; Corruption Watch 2024), weak enforcement undermines ethical intent; therefore, Influence must be operationalised through structured accountability mechanisms rather than symbolic commitments.

An integrated approach is essential. Character provides credibility, Morality defines normative standards, and Influence embeds these standards into organisational practice. Where these dimensions are misaligned, governance fragility may persist despite formal ethical frameworks.

Although exploratory, the study advances ethical leadership research by demonstrating that perceptions of leadership ethics within South African government institutions may cluster into three theoretically coherent domains. The convergence with established scholarship – including virtue ethics (Peterson & Seligman 2004), authentic leadership (Walumbwa et al. 2008), social learning theory (Bandura 1986), organisational justice (Colquitt et al. 2001) and governance-oriented models (Crews 2015; ed. Lazenby 2024; Rossouw & Van Vuuren 2017) strengthens theoretical plausibility.

The exploratory differentiation between Character, Morality and Influence underscores that ethical leadership in governance contexts cannot be reduced to personal virtue alone. In environments characterised by audit regressions and accountability failures (AGSA 2024; Corruption Watch 2024), ethical leadership must be conceptualised as a structured capability embedded in systems, procedures and observable behaviour. By providing a measurement foundation aligned with both theory and governance realities, this study contributes to moving ethical leadership from normative aspiration towards empirical and operational scrutiny.

Taken together, the findings suggest that ethical leadership in South African government institutions may require a broader conceptualisation than individual morality alone. Ethical governance may depend equally on leaders’ ability to translate ethical intentions into institutional practices, behavioural consistency and accountability systems. This insight has important implications for leadership development, governance reform and public-sector performance management.

Theoretical and practical implications

Theoretical contributions

The originality of this study lies in its exploratory examination of ethical leadership as a multidimensional governance capability within South African government institutions, rather than assuming a unified leadership construct derived predominantly from private-sector contexts.

The contribution lies not in validation but in measurement development. The study strengthens the measurement foundation of ethical leadership research in governance contexts and provides a platform for future CFA, measurement invariance testing and structural modelling linking ethical leadership to institutional performance outcomes.

By differentiating between Character, Morality and Influence, the study provides preliminary empirical evidence suggesting that ethical leadership in governance settings may comprise distinct yet interrelated leadership capabilities.

Practical contributions

The practical significance of this study lies in its potential contribution to public-sector governance and leadership development. By conceptualising ethical leadership as comprising Character, Morality and Influence, the study offers government institutions a more differentiated framework for leadership reflection, ethical capability development and governance strengthening.

Leadership development initiatives may benefit from differentiating between:

• Character development (integrity, authenticity, relational trust);
• Moral orientation (fairness, ethical responsibility);
• Ethical influence (communication, enforcement and modelling of standards).

The study further emphasises the need to reinforce behavioural and managerial enforcement practices in governance environments. While personal virtue remains foundational, ethical leadership effectiveness in public institutions may depend equally on the capacity to institutionalise standards through consistent oversight and accountability mechanisms (AGSA 2024; Corruption Watch 2024).

Limitations

Several limitations should be acknowledged. Firstly, the study was conducted within only two South African government institutions, limiting broader generalisability across the public sector. However, the exploratory nature of the study justified an initial focus on governance-oriented institutional settings characterised by formal accountability structures and bureaucratic oversight. Secondly, the study relied on self-report perceptions of ethical leadership behaviour, which may be influenced by respondent subjectivity. Thirdly, although the EFA provided promising evidence of dimensional structure, confirmatory validation was beyond the scope of the present study. Consequently, additional testing across broader institutional settings is necessary to establish dimensional stability and external validity. Nonetheless, these limitations should be interpreted considering the study’s exploratory purpose: To establish an initial measurement foundation rather than to produce a definitive, validated instrument.

Recommendations

Several targeted directions for future research and refinement are proposed. Firstly, future research should subject the three-factor structure of Character, Morality and Influence to CFA using independent samples drawn from a broader range of public-sector institutions. Confirmatory testing would allow assessment of model fit, dimensional stability and discriminant properties, thereby strengthening confidence in the scale’s structural robustness.

Secondly, further refinement of item composition is warranted. Regarding the Morality dimension, future research may expand the item set to enhance construct breadth. Conversely, the Character dimension may benefit from selective item reduction to improve parsimony while preserving reliability. Such refinement should be guided by confirmatory modelling rather than exploratory criteria alone.

Thirdly, subsequent studies should examine the interrelationships among the identified dimensions using oblique rotation and structural equation modelling. Exploring how Character, Morality and Influence interact may clarify whether ethical leadership operates as distinct but correlated capabilities within governance systems.

Fourthly, linking the scale to measurable management outcomes represents an important next step. Future research should investigate associations between ethical leadership dimensions and governance-relevant indicators such as audit performance, compliance effectiveness, ethical climate, organisational commitment and service delivery efficiency. Establishing predictive relationships would extend the instrument from structural identification to explanatory relevance within management science.

Fifthly, broader institutional replication across provinces, municipalities and state-owned enterprises would enhance generalisability and contextual robustness and thus contextual understanding of ethical leadership within South African governance systems.

From a practical perspective, public institutions should consider leadership development initiatives that strengthen ethical communication, procedural fairness, accountability enforcement and behavioural reinforcement processes. Emphasis should be placed on ensuring that ethical intent translates into consistent ethical practice within governance systems.

Conclusion

This study explored the dimensional structure of an Ethical Leadership scale within two South African government institutions and identified three interrelated dimensions: Character, Morality and Influence. The findings provide preliminary support for a multidimensional understanding of ethical leadership in governance settings and suggest that ethical leadership behaviour may extend beyond personal ethical orientation to include behavioural enactment, accountability and ethical influence processes.

The prominence of Character suggests that ethical leadership may initially be perceived through leaders’ integrity, fairness and moral credibility, while the separate emergence of Morality and Influence indicates that ethical judgement and behavioural enforcement may function as distinguishable leadership capabilities. These findings are particularly relevant within public institutions where governance legitimacy depends on accountability, procedural fairness and ethical oversight.

Although exploratory in nature, the study contributes a contextually grounded foundation for ethical leadership measurement within South African government institutions and provides a basis for future research examining ethical leadership behaviour in relation to governance effectiveness, accountability systems and organisational performance outcomes.

Acknowledgements

This article is based on research originally conducted as part of Dorothee Burger-Snyman’s doctoral thesis titled ‘An ethical leadership competency framework for managing the efficiency of internal controls and accountability to facilitate effective governance’, submitted to the College of Economic and Management Sciences, University of South Africa in 2026. The thesis was supervised by Mariette Coetzee. The thesis was reworked, revised and adapted into a journal article for publication. The authors affirm that this article complies with ethical standards for secondary publication, and appropriate acknowledgement has been made of the original work. The original thesis is publicly available at: https://ir.unisa.ac.za/handle/10500/32715.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

Mariette Coetzee: Conceptualisation, Supervision, Writing – original draft, Writing – review & editing. Dorothee Burger-Snyman: Data curation, Formal analysis, Methodology, Writing – review & editing. Both authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication and take responsibility for the integrity of its findings.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The data that support the findings of this study are available from the corresponding author, Mariette Coetzee, upon reasonable request.

The views and opinions expressed in this article are those of the authors and are the product of professional research. It does not necessarily reflect the official policy or position of any affiliated institution, funder, agency, or that of the publisher. The authors are responsible for this article’s results, findings, and content.

References