Constructing a leader’s identity through a leadership development programme

Abstract

This article explores the notion of leadership identity construction as it happens through a leadership development programme. Influenced by a conception that leadership development is essentially about facilitating an identity transition, it uses an intersectional approach to explore school leaders’ identity construction as it was shaped and influenced by experiences on the leadership development programme. The article draws data from a mixed-methods study that evaluated the impact of the leadership training programme offered to practising school leaders in South Africa. In order to examine the process of leadership identity construction, the article draws from data where identity work was visible. It argues that categories of identity – gender, race and social class – interacted simultaneously with the contexts and backgrounds of participants to shape and influence the outcome of the leadership development programme. This complex intersection enabled unexpected outcomes where women appeared to benefit more from the programme despite their less privileged entry status. The article calls for more work that asks direct questions on leaders’ construction of identity in order to inform leadership development programmes more meaningfully.

Keywords

identity construction intersectionality leadership development

Introduction

The purpose of this article is to explore school leaders’ identity construction through a leadership development programme. The article draws data from a study that evaluated the impact of the leadership training programme offered to practising school leaders, comprising principals, deputy principals and heads of department, between 2007 and 2009 in South Africa. Leadership development is used as an overarching term to refer to the programme as it was piloted on both practising and aspiring school principals. An impact evaluation of the programme showed that the candidates found it beneficial (Bush et al., 2011), with varying and differential experiences from various features of the programme such as mentoring (Moorosi, 2012) and networking (Kiggundu and Moorosi, 2012). While these publications and other studies have examined leadership development in Africa and other developing areas (Bush and Oduro, 2006; Lumby et al., 2008), the differential impact such training has on women, and other people of minority representation, and how these diversity factors interact remain unexplored. In this article, an intersectional analysis framework is used to explore how experiences of the leadership development programme shaped and influenced the constructions of leadership identity.

The rationale for using an intersectional approach in the analysis of school leadership development in South Africa is both pragmatic and theoretical. First, there is the long-standing problem of gender inequity in school leadership that has implications for leadership preparation and development. Coleman and Fitzgerald (2008) identified a gap in research that examines gender and educational leadership preparation and development. In particular, they problematized the exclusive focus on gender as a single category of identity, since it ignores the complexity of multiple identities that influence individual leadership learning and outcomes. Coleman and Fitzgerald (2008: 121) argued that, ‘individuals derive from a range of loyalties and socialisation dependent on their culture, language, religion and history’. All these diversity attributes, and others, constitute participants’ core identities and influence the outcomes of leadership development programmes.

Second, intersectionality is regarded as particularly useful in measuring leadership diversity since it helps to ‘reveal the multiple identities and personas of social actors, exposing the connection between those points’ (Richardson and Loubier, 2008: 143), thereby facilitating the understanding of distinct experiences from the effects of inextricably connected roles and situations. Burman (2003) argued that focusing on a single dimension of difference is essentially inadequate, as various dimensions of a person’s identity impact on each other and have a combined influence on the individual (Booysen and Nkomo, 2010). Booysen and Nkomo further argue that, in the case of South Africa, it is highly likely that certain racial and gender hierarchies are still influential on people’s experiences and material outcomes, which necessitates the use of a framework that allows analysis of multiple factors on the leadership development phenomenon.

Against this backdrop, an intersectional approach enabled the analysis of a leadership development programme, identifying issues that affected female and male participants as social groups and how the differences amongst them intersected to facilitate leadership identity development.

Leader and leadership development

A distinction is made between leadership development and leader development (Day, 2001; O’Toole, 2001); the former emphasizes development of every member of the organization through the creation of social capital, while the latter is focused on enhancing individuals in leadership roles without direct connection to organizational context. However, by positioning leader development as human capital investment, both Day (2001) and McCullum and O’Connell (2009) imply an overarching organizational development strategy that builds intrapersonal competence needed for effective performance of organizational roles. In this way, the individual in leader development is not divorced from the organizational context. Day (2001: 584) emphasizes the significance of intrapersonal competence associated with leader identity development through initiatives of ‘self-awareness’, which includes emotional awareness and self-confidence; ‘self-regulation’ (with self-control and trustworthiness), and ‘self-motivation’, encompassing commitment, initiative and optimism. Day (2001: 584) recognizes these capabilities as important for the enhancement of ‘individual knowledge, trust and personal power’, which is essential for the social interaction entailed in the leadership practice.

In recognition of these individual capabilities and their significance, Dalakoura (2010) argues for a systemic and integrated view of leadership that acknowledges its complexity in encompassing interactions between the leader and the social organizational environment. In this view of leadership, leadership development would be seen as a ‘social process’ (Dalakoura, 2010: 433) that expands the collective capacity of organizational members to enable groups to work together within relational commitments and mutual respect (Day, 2001). This collective approach to capacity-building invests more in social capital, putting emphasis on interpersonal exchange and interpersonal competence through networked relationships among individuals to enhance cooperation and resource exchange (Day, 2001). This collective approach to leadership essentially makes it a relational process that is supported by mutual trust and respect (Dalakoura, 2010), highlighting the interdependence between leader and leadership development in the pursuit of leadership effectiveness in organizations. Thus, the distinction between leader and leadership development is rather blurred as the concepts are sometimes used interchangeably. Leader development should be viewed as a starting point for leadership development as the two are mutually inclusive.

Leadership development as identity construction

When it happens at the beginning of career, leadership training is often construed as the initiation of identity construction rather than just an acquisition of managerial skills (Lumby and English, 2009). Petriglieri (2011) conceptualizes leadership development programmes as ‘identity workspaces’¹ that help meet the demand for effective leadership by benefiting the individual, the organization and society. A leadership identity model proposed by DeRue and Ashford (2010) suggests that leadership identity should be constructed at three different levels: the personal identity level, relational identity level and the collective identity level. These levels are also advanced by other theorists (e.g. Lord and Hall, 2005; Ibarra et al., 2010; Debebe and Reinert, 2012), who are all driven by the conception that sees leadership identity construction akin to a relationship construction (DeRue and Ashford, 2010). In this sense construction and internalization of a leader identity is perceived to be central to the process of becoming a leader (Ely et al., 2011).

DeRue and Ashford (2010: 629) explain the personal identity level as ‘individual internalisation’ which involves the incorporation of the identities of the leader as part of their self-concept or ‘persona’ (Richardson and Loubier, 2008: 148). This involves personal attributes of a leader that are embedded in ‘specific contexts where an identity is asserted and ascertained in the course of social interaction’ (DeRue and Ashford, 2010: 629). Richardson and Loubier (2008) add that leaders’ identities change as they go through leadership development programmes, but that personal identity precedes leadership development programmes. However, how the would-be leaders craft this personal identity or ‘knowing thyself’ (Bennis 2009: 50) becomes a significant process of leadership development. Bennis (2009: 50) states that knowing yourself means ‘separating who you are, and who you want to be, from what the world thinks you are and wants you to be’ and sees this as a lifetime self-invention process that happens at different stages for different people. This level involves actions that assert one’s leadership that would be affirmed or disaffirmed by others (such as followers) within a particular context, leading to a more relational aspect of leadership identity construction (Ely et al. 2011). When these actions are affirmed, one’s self-identity as a leader is strengthened (DeRue and Ashford, 2010), pushing one to seek further opportunities.

The realization of new opportunities leads to leadership growth and capabilities that get embedded in how a leader relates to others in the broader social or organizational context. According to DeRue and Ashford (2010), leadership identity internalization is closely linked to relational recognition or role-related identity (Debebe and Reinert, 2012), a second level of leadership identity construction which suggests that, in addition to leader identity internalization, the leader’s identity will be stronger if it is relationally recognized. This means that leadership is expressed through an interactive process of roles that are recognized and reciprocated by others (such as followers). DeRue and Ashford (2010) argue that the construction of leadership, and that of a leadership identity, are essentially about constructing a relationship – a reciprocal relationship that has to be affirmed by the parties concerned – in this case leaders and followers.

The third level of identity construction is collective endorsement, which involves being seen within the environment as part of a broader social group. Debebe and Reinert (2012: 11) identify this as a ‘socio-political context’ that ‘influences patterns within the person’s immediate contexts and defines and assigns individuals into social groups, thereby shaping social identity’. This is where individuals are collectively endorsed as a part of the group, such as women or men (DeRue and Ashford 2010). It is within this collectivism that relationships are built through networks that enrich shared leadership identity that embodies ‘shared representations and collective meanings’ (Day, 2001: 585) amongst leaders and those who are led.

There is a degree of overlap in the leader identity construction levels presented above which should not be interpreted to suggest a linear process of identity construction, but one that is iterative and co-constructed. Although Lord and Hall’s (2005) conceptions, represented as novice, intermediate, and expert respectively, can be construed to suggest linear progression, leadership identity construction cannot be conceived of as linear as that would condone a static and hierarchical process of leadership. The conception of leadership assumed in this article is one of a process that is dynamic, relational, and contextual, and changes over time, as posited by DeRue and Ashford (2010).

Ely et al. (2011) argue, however, that men and women experience different impediments to leader identity development. This is largely because leadership is still construed as a masculine entity in which women face particular challenges. Coleman and Fitzgerald (2008) reiterated the point made by Blackmore (1999), that leadership development programmes reinforce gendered barriers because they pay little attention to women’s ways of knowing and leading. These criticisms led to the suggestion of single identity leadership development programmes (Coleman and Campbell-Stephens, 2010; Ogunbawo, 2012; Debebe, 2011), on the premise that women would benefit more from leadership development programmes that are directed at their needs. However, single identity leadership development approaches are criticized for privileging one identity over others, for example, gender over race when women are singled from black people, as posited byDebebe and Reinert 2012). Coleman (2010) also highlighted problems facing women-only networks which risk not being taken seriously by the very nature of their composition (Ibarra, 1993). However, it must be noted that while customized leadership development programmes for Black leaders were found ‘patronizing, inappropriate’, and ‘likely to cause further resentment within the profession’ (Bush et al., 2006: 299), others continue to argue for bespoke programmes as a way of addressing diversity in leadership development (Coleman and Campbell-Stephens, 2010; Ogunbawo, 2012). The stance taken in this article argues that preserving differences oversimplifies the construction of leadership identity. As argued by others, single identity programmes of leadership development understate the complexity and intricate nature of the problems preventing equity in organizational leadership (Ely and Meyerson 2000). As Lumby and English (2009) suggest, if leader identity construction is oversimplified, leadership development programmes may not always be able to bring about equity.

Thus, leadership development is a complex process that involves more than the advancement of one individual but occurs through the interaction of the individual with the social and organizational environments. As has been argued, the development of leadership skills is intricately linked to the development of one’s self-concept as a leader (Ibarra et al., 2010). Leadership development is best understood as a process through which leadership identity is co-constructed through transformative practices of (amongst others) mentoring, networking and shared learning. However, Ely et al. (2011) argue that this construction is fraught with difficulties as it is more challenging for women to find the most powerful mentors and sponsors. This is due to lack of role models women identify with, gendered career paths that still favour men and gendered organizations that reflect men’s lives and men’s situations. Leadership development programmes, therefore, would be most effective if they enabled individuals to understand themselves within their contexts so that they are able to address issues that are unique to them (Debebe, 2011).

An intersectional approach to leadership development

Several authors identify two models (additive and intersectional) of identity development in conceptualizing the interaction of multiple identities. Holvino (2012) goes on to propose a third model called ‘simultaneity’, which due to limited scope is not reviewed in this article. Additive approaches are based on the assumption that demographic categories of identity are distinct and have additive effects on one another (Hancock, 2007). It is through such models that black women are often said to suffer a ‘double jeopardy’ of racism and sexism or triple jeopardy when they are additionally poor (Glen, 2002; Hancock, 2007). Debebe and Reinert (2012) argue that this additive relationship is hierarchical and assumes one category of identity is more critical than the other, when it is actually the interaction of categories that is more important than the hierarchical relationship. Additive approaches have therefore been criticized on their failure to recognize complexity of interactions between multiple categories of identity, by reducing them to mathematical metaphors (Holvino, 2012).

Intersectional approaches to leadership identity development, on the other hand, recognize the complexity of interactions between multiple categories. Yuval-Davis (2006) states that ‘intersectionality’ emphasizes the complexity of processes of individual identification as social inequity cannot be captured by arithmetical frameworks suggested by additive approaches that view identities as ‘independent axes’ (Glenn, 2002: 7). Yuval-Davis argues that,

Categories like gender, ethnicity and class co-construct each other, and they do so in myriad ways, dependent on social, historical and symbolic factors. (Yuval-Davis, 2006: 279)

Intersectionality theory is therefore premised on the notion that people live multiple layered identities derived from social relations, history and the operation of structured power (Crenshaw, 1991). Prins (2006) argues that intersectionality constitutes a critical alternative to identity politics by taking into account differences between and within groups. To this, Richardson and Loubier (2008: 143) add that an intersectional approach neither constructs individual categories as autonomous categories of analysis, nor adds one category on top of another, but rather it helps the analyst to examine the interaction of categories towards understanding leadership identity and differences in leadership behaviour and efficacy.

Intersectionality pays attention to the interplay of macro societal structural inequities and the interacting daily social practices that perpetuate inequality (Holvino 2012). The simultaneous and interactive relationship of these ‘various systems of social identity and social location’ (Mishna, 2012: 55) provides a more accurate picture of inequality in school leadership than focusing on a single dimension of difference. While it presents a more accurate picture of inequality, Holvino (2012: 170) adds an interesting dimension that intersectionality ‘stresses the qualitative impact of categories of identity, which intersect in a variety of ways, creating at the same time both oppression and opportunity’. The creation of both oppression and opportunity (also Smooth, 2010) is perhaps the ‘unintended consequence’ of intersectionality that McCall (2005) alludes to when emphasizing the complexity of an intersectional analysis. How these categories of difference intersect in a leadership development programme and the extent to which this intersection shapes the construction of leadership identity become the focus in later discussions.

Methodology

This article draws on data from a longitudinal study conducted between 2007 and 2010, evaluating the impact of a leadership development programme. The evaluation was carried out in six of the nine South African provinces that participated in the first cohort of the national pilot of the ACE² leadership development programme. The evaluation comprised a mixed-methods approach involving a survey, with quantitative data, and case studies. A population survey of 287 participants was conducted at the beginning of the programme, as a baseline study, and at the end of the programme, as an impact study. Both surveys guaranteed a successful close to 100 percent return rate. Case study methodology involved shadowing the participants, and interviewing them, and at least five members of their role sets. Data were also gathered from observations of the networking and mentoring sessions, which were significant components of the programme. There were 27 case study participants in the baseline phase, but sample erosion reduced this to 25 in the mid-term phase and 24 in the impact study. Most (17) of the case study participants were school principals, with five deputies and two heads of department. The final 24 case study candidates provided a gender balance almost in line with the aims of the ACE programme³ (13 women and 11 men). Eleven were from primary schools, 10 from secondary, two from combined schools and one from a training centre.

Although the article employs an intersectional analysis (often construed as a feminist tool) of the leadership development programme, it analyses data from all participants, including men. A critical point to note is that the study was not conceived with an intersectional analysis but the framework is imposed on the already existing data. This is an attempt to offer an alternative analysis of the data, which is hoped to invite more critical analyses of the leadership development programme that has thus far been accepted as a significant step towards leadership effectiveness in South African schools.

Perhaps central to the critique are the author’s own biases and values influenced by the political insider status which informs this analysis in unpacking subtle gendered and racialized factors as they intersect with the social context. It is acknowledged that the author’s familiarity with the context, as a Black South African female academic, carries some significance in the current interpretation of the findings.⁴ On the other hand, the social and professional ‘outsider status’ of the author as a university researcher has limitations that may lead to the ignorance of more subtle nuances and create barriers to interpreting some of the experiences (Beoku-Betts, 1994). Beuko-Betts problematized her position in interviewing participants similar to her in colour and questioned whether her blackness was enough to claim insider status. These nuances are revisited towards the end of the article.

In terms of data analysis, the author used themes developed from the literature on identity construction. These themes were then used to code the data to identify areas where identity work was visible. Secondary data analysis (although often used in large quantitative studies) is a valid method of analysis that looks into examining already existing data, using a different lens or framework. In this case, the use of multiple lenses on the evaluation findings from the leadership development programme yields significant results that may be useful for advancing the understanding of issues concerning women and other minority or less privileged groups within leadership development programmes. However, limitations of imposing both an intersectional approach and leader identity development frameworks on the existing data are fully acknowledged and some reflections are made towards the end of the article.

Participants’ profile

Participants in the leadership development programme comprised practising school principals, deputy principals and heads of department, who will all be referred to as leaders in this article. One province recruited school management teams instead of a single leader (e.g. the principal or deputy principal) as was the case in all other provinces. The overall pattern of recruitment resulted in more principals (77.5 percent) on the programme, with the remaining 22.5 percent comprising deputy principals and heads of department. Women made up 44 percent of the principals, 29 percent of the deputy principals and 66 percent of heads of department. A further breakdown by school type showed that 64.4 percent of participants were from primary schools and 23.2 percent from secondary schools. The remaining 12.2 percent were from special and combined schools. The majority of women principals and deputies came from primary schools, while the heads of department were all from secondary schools. It must be noted that, even though the heads of department were all from one province, they do reflect gender representation nationally, as more women are represented at the middle levels of school leadership. This imbalance of gender representation in school leadership shows that participants had different levels of pre-knowledge, experience and skills, in spite of the programme being standardized. Significantly, the majority of men were at a more advantaged (principalship) level.

The racial profile comprised a majority of Africans (64.8 percent), followed by Coloured (22.9 percent). Indian and White leaders appeared to be under-represented as they shared the remaining 13.3 percent, but the overall representation is not very different to the national demographic profile. While teachers and school leaders from different racial groups may have received different levels of training previously, the leadership development programme in question was the first of its kind in the country to standardize training requirements for all leaders regardless of their background.

Findings

As the focus of the evaluation from which data are drawn was not leadership identity construction, this analysis focuses on selected sections of the data where identity issues were integral. The analysis therefore identified three themes reflecting factors that intersected and shaped the construction of leadership identity through the leadership development programme. These themes are identified as: the identification and development of personal attributes; the interaction of gender, race, background and context; and the interaction with mentoring and networking as external factors that interactively shaped the programme’s outcomes.

Developing personal attributes

Leaders mostly identified attributes such as approachability, openness, ability to work in a team and being consultative, as their strengths at the beginning of the programme. These attributes were identified mostly by female leaders and largely confirmed by their role sets. Male leaders, on the other hand, identified different attributes including: lead by example, clear vision, etc. However, at the end of the programme, both male and female participants’ reported strengths included better time management, improved self-confidence, enhanced self-control and better relationships with staff and community. One female principal of a township primary school identified lack of self-control as her main weakness at the beginning of the programme. Commenting on her behaviour before the programme, a member of her role set said:

The principal was moved straight from [a junior to a senior post] and was never exposed to any gradual managerial exposure. As a result, she would sometimes shout and be rude to other educators, and would not listen even when we would advise her.

However, at the end of the programme, the principal reported to have changed her behaviour and attitude towards other educators, was more able to control her temper and even accepted criticism. In a different case, a female principal of a township school, who was described as rigid in the baseline phase, was subsequently described as ‘not so one directional, is more democratic and more flexible’, by one of her teachers. One female principal of a primary school was reported to be ‘always firm and fair, but her attitude to staff has improved and stabilised’. Another female principal, who was perceived to be ‘too strict’, ‘too harsh’ and a ‘perfectionist’, by her role set, added that the programme gave her the confidence to reach out to staff to help identify her weaknesses which she was working on even after the programme had ended.

Other attributes included leadership styles of participants such as being an autocratic leader. A male principal was described as ‘autocratic’ in the baseline study, but more ‘democratic’ at the impact phase. Another one said that he used to be a ‘dictator’, but is now more ‘approachable’.

A further attribute identified was that of self-confidence. In one case, a female principal of a poor rural school did not identify confidence as a weakness at the baseline stage, but this was identified by her role set. At the end of the programme, the female principal herself remarked on how her self-confidence had improved to the extent that she could approach other people outside the school for things that the school needed. Her governing body chairperson said that ‘she is now able to suggest changes for improvement of the school’, adding that the school has acquired a computer, a photocopier and built a storeroom, since the principal began the leadership training programme. While this may not have been a direct result of the training programme, it is as a result of the confidence that the principal has built – the latter being a direct benefit from the programme. Another role set member (a senior educator) added that the principal has become ‘firmer in asking for evidence for the things she needs’, has become more open and wants to learn from others.

Although the majority of participants who reported on improved self-confidence were female principals, male principals also reported improved self-confidence even though it was not reported as a weakness at the beginning of the programme. For example one male principal said,

The ACE course taught me to be more tolerant with my colleagues and learners. I am now more confident and I am able to accept other people’s opinion/view and not looking at it as undermining my authority.

Another male principal of a township secondary school said,

After the course, I was able to transfer the knowledge I acquired during the course into my practice at school and the course turned me into a confident principal.

This suggests a level of awareness about who the participants were as leaders at the beginning of the programme and how this awareness changes and improves their confidence during the programme, strengthening their identity as leaders. It is perhaps important to note that not all participants improved on the attributes identified as weaknesses during the baseline study. For example, one male principal admitted to be struggling with people skills and this was still a problem even at the end of the programme, according to his role set. However, it can be argued that admission of weaknesses happens as confidence levels improve, which would prompt the leader to seek help.

It is significant that, while women tended to identify with the traditionally perceived female attributes and men identified more instrumental attributes at the beginning of the programme, at the end of the programme, male participants appeared to have adopted more ‘feminine’ attributes such as being a good listener, being more consultative, communicating more openly, which some authors have referred to as the female advantage (Eagly and Carli, 2003). It is also significant that self-confidence was not identified as a weakness by participants at the beginning of the programme, but came out when they reflected on their improvement during the programme.

Thus, as they gained more knowledge from the leadership development programme, and, they improved their self-confidence and self-control as well as changing their leadership styles leading to improved leadership performance, they developed a clearer identity of themselves as leaders. The role played by role sets in affirming the leaders’ advancement appeared to be crucial in how they were being perceived. DeRue and Ashford (2010) argue that being a leader is not a static process, but one that is influenced by various individuals collectively, as they construct their respective identities of leaders and followers. This is a more dynamic, social and relational conception of the leadership development process.

The intersection of gender, race, context and background

Responses to the survey question where participants were asked to rank their main benefits from the programme indicated higher gains by women who worked in disadvantaged rural primary schools. These results suggested that women were 10 percent more likely than men to have benefited in leadership, management and managing people. These three areas suggest basic skills of leadership and management and were indicated as the top three gains by participants. The case study data confirmed that participants from the former DET⁵ (African) schools who comprised 58 percent of the participants, gained a great deal from the training programme. These results suggest that African women, as they were mostly located in the rural primary schools compared to their Indian, Coloured, White and all male counterparts, received more gains from the training programme, depicting the gender, race and class intersection.

Male participants on the other hand, who constituted the majority of principals on the programme, entered the training with a slightly more advanced skill base, which might have been expected to put them at an advantage in terms of how they absorbed new learning from the programme. One Indian male principal of a city secondary school, who had a postgraduate qualification in leadership, had some misgivings about the programme, arguing that it was more like a ‘refresher course’, suggesting limited benefits because candidates ‘were not at the same level’. Although one female principal of a township secondary school also said ‘the first year was not challenging as it covered things I had already done before’, it was mostly men who were serving principals and those who had previous qualifications in leadership and management who reported limited benefits from the programme. Thus, while findings indicated differences by gender and race, greater differences were seen in terms of background. However, the apparent interaction of gender, race and background, highlights a significant outcome that made African women the majority benefactors from the programme. Their previous disadvantage becomes their source of empowerment that influenced their development as leaders through a leadership training programme.

In addition to background, greater variations were also seen by sector (primary or secondary) and by location (rural, township, urban). These variations signalled important contextual issues that impacted significantly on the outcomes of the leadership development programme. Although there were variations by province, overall 41.9 percent of participants were rural based, 24.3 percent were township based; 16.2 percent were urban schools; 11.3 percent were farm schools (which are mostly rural) and 3.2 percent were schools in informal settlements.⁶ These variations highlight the overall gains made by leaders in schools serving deprived communities.

The baseline survey conducted in 2007 revealed that many school leaders were experiencing significant problems with teachers and parents, but many of the participants reported much improvement in leading and managing educators and parents in the impact survey in 2009. The significant gains were made by participants from schools located in the townships and rural areas, with improved relationships with parents reported by all participants in rural locations. Survey respondents claimed significant improvements in relationships with a wide range of stakeholders. For example, 84 percent claim that ‘working with other educators’ was ‘much improved’, while most case study participants reported improved relationships with parents and some, but not all, attributing this to the ACE programme. Thus, leaders became aware of, and acknowledged the value of community links and stronger parental involvement as they became more confident in their role as school leaders.

Interactive learning: mentoring and networking

Mentoring and networking were significant features of the leadership development programme that exhibited leader identity construction to some extent. Mentors were instrumental in helping participants to internalize leadership learning, thereby nurturing their identity development. While the survey data revealed that the majority of participants generally found mentoring helpful (59 percent said mentoring was ‘of great help’ and 27 percent said mentoring was ‘of some help’), a closer analysis of the qualitative data showed that some men had some misgivings about the mentoring aspect of the programme. These male mentees appeared to devalue social support from their mentors and regarded it as inadequate (Moorosi, 2012). Female participants on the contrary appeared to value both social and professional support, suggesting fuller benefit from the mentoring programme. There is room to argue that mentors whose characteristics, such as experience of similar school contexts, matched those of mentees had more potential for leadership identity development as they provided realistic role models for participants. Although there were fewer role models for female participants, there does not seem to have been an issue with male mentors mentoring women as the literature suggests is possible (Clutterbuck and Ragins, 2002; Karsten 2006; Chatman, 2010).

Evidence from the case studies suggests that, although network groups in most provinces did not meet regularly, arguably due to lack of time and geographical proximity reasons, when they did meet their sessions were often informal and voluntary, suggesting a drive towards taking initiative. A networking session observed by one of the researchers was a typical assignment discussion meeting. In this meeting, a female principal of a city primary school was encouraged to lead the discussion with other leaders, who were all male. In other observations participants had their own networking meetings without the mentor. It can be argued that where participants took turns to lead the discussions this showed signs of a transition to developing the leadership identity. Mentoring and networking thus appeared to have facilitated leadership identity development, albeit with unequal gains for participants.

Discussion: constructing a leadership identity through leadership development

Scholars of identity in leadership (Lord and Hall, 2005; Ibarra et al., 2010; Ely et al., 2011; DeRue and Ashford, 2010) agree that developing leadership skills and knowledge is similar to developing one’s identity as a leader because of the strong relationship between the self and leadership development. Ely et al. (2011) view the relationship between leadership development and leader identity construction as ‘recursive and mutually-reinforcing’. In this sense, going through a leadership development programme entails some kind of identity transformation, not only on the part of the leader but on that of the followers as well. For Ely et al. identity work is a developmental process that suggests progression and acquisition of some basic skills, and a clear development of self-understanding, before a certain level of expertise (Lord and Hall, 2005) can be achieved. Although a leadership preparation programme for aspiring principals is an ‘initiation into identity construction and subsequent performance’ (Lumby and English, 2009: 97) and arguably, the first level of professional socialization, acquiring leadership expertise is not confined to a leadership development programme. Acquiring leadership expertise is a career-long process, with expertise being built over time. This is visible in how participants developed their identity over time but also in how, by the end of the programme, their achievement of leadership expertise still needed development.

At a personal identity level, a novice leader seeks to improve their individual knowledge and the focus is very much on the individual and their own leadership self-awareness. For Day (2001), this constitutes leader development which is essential for an individual leader’s preparation for interaction at the social organizational level. At this stage, the leader has an identity that confirms who they are, differentiating them from the others. In this study, participants were asked to identify their strengths and weaknesses at the beginning and the end of the programme. It is through these responses that identity transition becomes visible, with more transformation evident in female leaders who had been less privileged. These responses portray leaders’ reflections on their own self-awareness, self-confidence and self-regulation. As many leaders were only able to reflect on the enhancement of their self-confidence and self-control at the end of the programme, this suggests the non-linearity of the identity construction process, and that identity construction happens through reflection. Indeed, Bennis (2009) confirms that true understanding of oneself comes after reflection on experience.

At the relational level, leaders are still seeking affirmation and acceptance as leaders and this is where some participants in the study were beginning to claim leader identity, and being granted that leadership identity by their role set. By so doing, both leaders and their role set were embarking on what DeRue and Ashford (2010) call a ‘claiming-granting’ process where leaders assert themselves and their followers grant them the status of leadership by confirming their role. Mentorship roles, where leaders are spurred towards a leadership role by their mentors, contributing to the interactive learning process, also became significant processes of granting leadership by mentors. The intersection of gender and race in mentoring is evident in this analysis but even more visible and significant is the intersection with professional background. Moorosi (2012) concluded that racial and gender diversity cannot be ignored in mentoring, particularly in a context with historical baggage, such as in South Africa. However, it is evident that professional background is perhaps more significant in the constitution of mentoring relationships, and arguably important in facilitating leader identity development.

The collective nature of leadership development is similar to what Day (2001) describes as leadership rather than leader development and what Lord and Hall (2005) regard as expert knowledge. That is, leaders develop expertise as they integrate their development towards broader achievement of organizational goals, interacting more with others. In Lord and Hall’s (2005) words, ‘the more expert one is in their identity, the more inextricably integrated leadership skills and knowledge are with one’s self concept’ (592). Petriglieri (2011) emphasizes that leadership development programmes realize their full potential when they add value to these three levels of identity development. It is arguable that participants in the study had not reached the expert level of identity construction. Perhaps the duration of a leadership development programme is not long enough to exhibit the leaders’ expert or collective identity, as this is not yet solidified by the end. Moreover, participants entered the programme with different sets of skills and knowledge that interacted differently with how they internalized leadership learning and how they implemented new knowledge thereafter. This provides one explanation for the development of leadership identity at different times.

A further point to observe is that while participation in the leadership development programme guaranteed inclusion, subsequent leadership performance, which would facilitate identity development, was not a given. Ibarra et al. (2010) argue that leadership development is intrinsically connected to identity transition. However, these findings suggest that, while the identity development model is helpful in connecting the self to the process of leadership development, it is perhaps questionable that identity transformation will happen to all those involved in a leadership development programme and that identity transformation is essential for a leadership development programme to be successful. Perhaps notable from this analysis is the high level of interaction between leadership learning, context and background of participants.

Identity transition was shaped and influenced by context and background, giving the less privileged women leaders in rural contexts an empowerment opportunity through leadership learning. Holvino (2010) argues that this position of disadvantage creates opportunities particularly for Black women (who suffer multiple oppressions) and specific relationship to knowledge and knowledge production. There is an evident intersection of gender, race and class here, although this claim is made with great caution. Holvino (2010, 2012) and Smooth (2010) identify the coexistence of both privilege and marginalization in intersectionality work. Smooth (2010), in particular, links this conundrum to women in leadership where, on the one hand, women are powerful because they hold powerful positions as leaders while, on the other hand, they are less powerful as they operate within discourses laden with preferences that benefit the dominant race, gender and class. The current analysis reflects less privileged women leaders, turning the situation (albeit subconsciously) to their advantage by benefiting the most from the leadership development programme, thereby developing a stronger leadership identity. This privilege is not normative, but perhaps intersectionality in this analysis challenges normativity by allowing the possibility of unpredictable connections.

Thus, using the leader identity construction model, it is visible that leadership identity can be constructed at different levels, and through different ways, but the use of intersectionality makes all connection points of benefit and loss even more visible. Leaders developed an improved sense of self at different stages throughout the various aspects of the programme, leading to changed personal attributes which suggest an awareness of personal development that made them better leaders by the end of the programme. Participants’ leadership identity development seems to be interactive, with development occurring at different stages of the programme, rather than linearly. As Bennis (2009) suggested, leaders achieve different milestones of leadership identity development at different times in their careers.

Implications and conclusions

In using the leadership identity construction framework to interpret these results, there were some challenges that may have implications for identity transformation and the success of leadership development programmes in the future, as well as having implications for further research: First, the study was not conceived of as identity development work and it is possible that the author’s interpretation of identity construction in this programme may not be universal. However, there is a degree of certainty about the knowledge and skills developed by the leaders from the programme, with the interaction of others, which confirms leadership development as an individual, relational and collective enterprise. Although the leader identity development model suggests that the success of a leadership development programme is achieved when the leader’s identity is embedded in these three identified levels, it is argued here that, if we are to consider the definition of leadership development as a continual process, constantly shaped by context, then it is not possible to achieve an end point. Indeed some scholars (Lumby and English, 2009) have argued that leadership preparation, which the programme was for aspiring principals, should be treated as initiation into identity construction rather than full acquisition of managerial and technical knowledge.

Second, identity transformation through a leadership development programme cannot be achieved by all participants. As Ely et al. (2011) noted, men and women experience identity construction differently, and these authors are adamant that subtle and pervasive forms of gender bias impede women’s progression by obstructing identity work that is necessary for them to take up leadership roles. There were obvious ‘first generation’ forms of gender bias underpinning the leadership development programme that hint at the pervasive presence of the more subtle biases. For example, that the programme did not recruit equal numbers of men and women in one province signals a strong presence of ‘second generation’ forms of bias – the most powerful yet invisible barriers to women’s progression that arise from cultural practices, workplace practices and forms of interaction that indirectly favour men (Ely et al., 2011). While previous work may suggest a ‘self-sealing’ (Ely et al., 2011) nature of the problem where conditions surrounding leadership development continue to favour men, thereby making it difficult to confront the gendered challenges, the link between disadvantage and opportunity that is made possible through an intersectional perspective offers hope. Women leaders with their less privileged background and the gendered contexts in which they continue to work have an opportunity to turn their situation around through leadership development programmes. This has direct implications for the nature of such programmes, who delivers them and how. But perhaps a more directed study, that asks direct and relevant questions about the participation of women in leadership development, would uncover more subtle nuances that inform the conceptualization of leadership development programmes. Additionally, such directed questions would enhance our understanding of the intersection between gender, race and social class and how these interactively shape the construction of leadership identity. Having said that, an intersectional analysis employed in this article highlighted some interactions with background and context which historically put women in a disadvantaged situation and those that presently put African women at an advantage in terms of developing their leadership identity. Burman (2003) suggests that intersectionality is a helpful starting point for critical thinking and practice. However, the limitations of imposing intersectionality on a non-feminist secondary analysis are fully acknowledged.

Third, perhaps as a university based researcher, one would always be at a disadvantage in reading and understanding participants’ meanings of their identity development. Researching a multi-racial terrain, where the blackness of the researcher identifies with the majority (rather than the minority) of participants, and therefore makes the researcher claim the insider status, may appear as a position of advantage on the surface. However, this position must not be taken for granted as it by no means suggests automatic access into the participants’ world. Holvino (2010) confirms that university-based researchers are less powerful and usually least trusted or credited by the very communities they seek to represent. The combination of the three factors brought up in this section problematizes the researcher’s position and interpretation of these findings, but significantly, these reflections are helpful in considering ways forward in research underpinned by social justice and equity in educational leadership development.

The article concludes that leadership development experiences were characterized by a complex intersection of first-level identities (race, gender, class), backgrounds (experience and level of education) and contexts (location, socioeconomic status and type of school) of participants that influenced the outcomes from the leadership development programme. On the one hand, while traditionally the intersectionality framework has been used to explain oppression and injustice, in this analysis it additionally explains benefits and gains from the leadership development programme, suggesting female advantage in leader identity construction. Higher gains from the programme are undoubtedly made by African women working in rural contexts, thereby turning their disadvantage status into a source of empowerment. The intersection of contextual factors influenced and shaped leadership identity construction, which, in turn, influenced the outcomes of the leadership development programme.

Footnotes

Acknowledgements

Other members of the research team whose involvement in the research is acknowledged were: Tony Bush, Stunky Duku, Derek Glover, Edith Kiggundu, Soraya Kola and Vuyisile Msila. I am also grateful to colleagues who read earlier versions of this article.

Funding

The research was funded by the ZENEX Foundation (South Africa).

Notes

References

Bennis

(2009) On Becoming a Leader. New York: Basic Books.

Beoku-Betts

(1994) When black is not enough: Doing field research among Gullah Women. NWSA Journal 6(3): 413–33.

Blackmore

(1999) Troubling Women: Feminism, Leadership and Educational Change. Buckingham: Open University Press.

Booysen

LAE

Nkomo

(2010) Gender role stereotypes and requisite management characteristics: The case of South Africa. Gender in Management: An International Journal 25(4): 1754–2413.

Burman

(2003) From difference to intersectionality: Challenges and resources. European Journal of Psychotherapy, Counselling and Health 6(4): 293–308.

Bush

Oduro

(2006) New principals in Africa: Preparation, induction and practice. Journal of Educational Administration 44(4): 359–75.

Bush

Glover

Sood

(2006) Black and ethnic minority leaders in England: A portrait. School Leadership and Management 26(3): 289–305.

Bush

Kiggundu

Moorosi

(2011) Preparing new principals in South Africa: The ACE: School Leadership programme, South African Journal of Education 311: 31–43.

Chatman

(2010) Norms in mixed sex and mixed race work groups. Academy of Management Annals 4(1): 447–84.

10.

Clutterbuck

Ragins

(2002) Mentoring and Diversity: An International Perspective. Oxford: Butterworth-Heinemann.

11.

Coleman

(2010) Women-only (homophilous) networks supporting women leaders in education. Journal of Educational Administration 48(6): 769–81.

12.

Coleman

Campbell-Stephens

(2010) Perceptions of career progress: The experience of black and minority ethnic school leaders. School Leadership and Management 30(1): 35–49.

13.

Coleman

Fitzgerald

(2008) ‘Gender and leadership development’. In: Lumby

Crow

Pashiardis

(eds) International Handbook on the Preparation and Development of School Leaders. New York: Routledge, pp. 119–36.

14.

Crenshaw

(1991) Mapping the margins: Intersectionality, identity politics and violence against women of color. Stanford Law Review 43: 1241–99.

15.

Dalakoura

(2010) Differentiating leader and leadership development: A collective framework for leadership development. Journal of Management Development 29(5): 432–41.

16.

Day

(2001) Leadership development: A review in context. Leadership Quarterly 11(4): 581–613.

17.

Debebe

(2011) Creating a safe environment for women’s leadership transformation. Journal of Management Education 35(5): 679–712.

18.

Debebe

Reinert

(2012) Leading with our whole selves: A multiple identity approach to leadership development. In: Miville

Ferguson

(eds) The Handbook of Race-Ethnicity and Gender in Psychology. New York: Springer.

19.

DeRue

Ashford

(2010) Who will lead and who will follow: A social process of leadership identity construction in organizations. Academy of Management Review 35(4): 627–47.

20.

Eagly

Carli

(2003) The female advantage: an evaluation of evidence. Leadership Quarterly 14: 807–34.

21.

Ely

Meyerson

(2000) Theories of gender in organisations: A new approach to organisational analysis and change. Research in Organisational Behaviour 22: 103–51.

22.

Ely

Ibarra

Kolb

(2011) Taking gender into account: Theory and design for women’s leadership development programs. Academy of Management Learning and Education 10(3): 374–93.

23.

Glenn

(2002) How Race and Gender Shaped American Citizenship and Labor. Cambridge, MA: Harvard University Press.

24.

Hancock

(2007) When multiplication doesn’t equal quick addition: Examining intersectionality as a research paradigm. Perspectives on Politics 5(1): 63–79.

25.

Holvino

(2010) Intersections: The simultaneity of race, gender and class in organization studies. Gender, Work and Organization 17(3): 248–77.

26.

Holvino

(2012) The ‘simultaneity’ of identities: Models and skills for the 21st century. In: Wijeyesinghe

Jackson

(eds) New Perspectives on Racial Identity Development. New York: New York University press, pp. 161–91.

27.

Ibarra

(1993) Personal networks of women and minorities in management: A conceptual framework. Academy of Management Review 18(1): 56–87.

28.

Ibarra

Snook

Guillen Ramo

(2010) Identity-based leader development. In Nohria

Khuruma

(eds) Handbook of Leadership Theory and Practice. Boston, MA: Harvard Business School Press, pp. 657–78.

29.

Karsten

(2006) Gender, Race and Ethnicity in the Workplace: Issues and Challenges of Today’s Organisations, Westport, CN: Praeger Publishers.

30.

Kiggundu

Moorosi

(2012) Networking for school principalship in South Africa: Perceptions and realities. School Leadership and Management 32(3): 215–32.

31.

Lord

Hall

(2005) Identity, deep structure and the development of leadership skill. Leadership Quarterly 16: 591–615.

32.

Lumby

English

(2009) From simplicism to complexity in leadership identity and preparation: Exploring the lineage and dark secrets. International Journal of Leadership in Education 12 (2): 95–114.

33.

Lumby

Crow

Pashiardis

(2008) International Handbook on the Preparation and Development of School Leaders. New York: Routledge.

34.

McCall

(2005) The complexity of intersectionality. Signs 30(3): 1771–1800.

35.

McCullum

O’Connell

(2009) Social capital and leadership development: Building stronger leadership through enhanced relational skills. Leadership and Organisational Development Journal 30(2): 152–66.

36.

Mishna

(2012) Bullying: A Guide to Research, Intervention, and Prevention. New York: Oxford University Press.

37.

Moorosi

(2012) Mentoring for school leadership in South Africa: Diversity, dissimilarity and disadvantage. Professional Development in Education 38(3): 487–503.

38.

Ogunbawo

(2012) Developing Black and minority ethnic leaders: The case for customized programmes. Educational Management Administration Leadership 40(2): 158–74.

39.

O’Toole

(2001) When leadership is an organizational trait. In: Bennis

Spreitzer

Cummings

(eds) The Future of Leadership. San Francisco: Jossey-Bass, pp. 158–74.

40.

Petriglieri

(2011) Identity workspaces for leadership development. In: Snook

Nohria

Khurana

(eds) The Handbook for Teaching Leadership. Thousand Oaks, CA: SAGE, pp. 295–312.

41.

Prins

(2006) Narrative accounts of origins: A blind spot in the intersectional approach? European Journal of Women’s Studies 13: 277–90.

42.

Richardson

Loubier

(2008) Intersectionality and leadership. International Journal of Leadership Studies 3(2): 142–61.

43.

Smooth

(2010) Intersectionalities of race and gender and leadership. In: O’Connor

(ed.) Gender and Women’s Leadership: A Reference Handbook, Thousand Oaks, CA: SAGE, pp. 31–40.

44.

Yuval-Davis

(2006) Intersectionality and feminist politics. European Journal of Women’s Studies 13(3): 193–209.