Generation X leaders from London, New York and Toronto

Inspired by scholarly calls to focus more intently on the influence of context on leaders’ construction and negotiation of identity, this paper draws on evidence from our Economic and Social Research Council (ESRC) project in London, New York City and Toronto. Throughout the paper, we strive to illuminate how the city-based context influences how race/ethnicity is experienced and described. We use social identity theory, organisational fit and in-group prototypes to frame school leaders’ explicit discuss race/ethnicity when reflecting on identity. We describe our data gathering process using our Professional Identity card-sort Tool, which guided leaders’ reflections on identity. The analysis details how we extracted and interpreted evidence from leaders who were explicit about the interrelationship between their own personal racial/ethnic identification and its alignment or misalignment with their school-level communities. We explore how different city contexts influence leader experience of in-groups and out-groups and the related leadership challenges and opportunities. In conclusion, we reflect on the influence that structures, policies and communities have on how leaders experience identity and the possible implications for their work. We also explore the value of attending to potential context-based identity-driven experiences for school leader development and support.


Introduction
Teacher disinterest in leadership (Gronn and Lacey, 2004), headteacher retirements (Howson, 2008) and rapid cycles of staff turnover (McKinney et al., 2007) continue to negatively influence recruitment and retention of school leaders (BBC, 2009). As school leaders exert the greatest influence after teachers on student achievement and development (Leithwood and Jantzi, 2008;Leithwood et al., 2004), identifying, nurturing, recruiting, developing and retaining leaders is increasingly one of the most pressing urban education issues.
Within this context of shifting principal demographics, a new generation of leaders has recently entered senior school leadership roles. It is often recognised that each new generation of leaders brings challenges to its organisations to evolve, build and adapt to its cohort's expectations and ways of working (Smola and Sutton, 2002). As such, the arrival of Generation X (GenX) leaders (born between 1960 and 1980) marks an important transition and handover as baby boomers retire (Edge et al., 2016a). As GenX leaders have grown up in the most rapidly progressive technological era, they are recognised as more globally minded, techno-savvy and informal (Zemke et al., 2000). GenXers also desire and actively seek out collaboration (Smola and Sutton, 2002), mobility (Duchscher and Cowin, 2004), diversity and more experimental organisational structures (Kunreuther, 2003). GenX work-related characteristics are markedly different from those of previous generations. However, despite the potentially seismic influence of GenX leadership on education systems globally, little research explores the work, experience, careers and aspirations of these leaders in education or beyond.
To address this knowledge deficit, we located our Economic and Social Research Council (ESRC)-funded , three-year study in three highly diverse, English-speaking, top-ten ranked (Foreign Policy, 2009) and educationally innovative Global Cities (Sassen, 1991): London, New York City and Toronto. We assumed that if GenX leaders were approaching their work and lives differently, they might require a new set of policy and practice strategies, incentives and supports to be recruited, developed and retained. We explore whether, and how, the generational attributes of GenX principals and vice-principals influence their career development, work/ life issues, organisational improvement and future aspirations. Cohorts of 20-30 under-40-year-old GenX vice-principals and principals participated in annual interviews and focus groups .
Our initial analysis within and across cities indicated distinct patterns in how our small sample of GenX leaders view the advantages and disadvantages associated with being a young leader (Edge et al., 2017). Advantages included: being open to new ideas, relating to younger staff/ students, energy, tech-savviness and greater commitment to collaboration. Conversely, leaders share the disadvantages of being a young leader including: having to prove yourself, perceived lack of credibility, negative and inaccurate perceptions, perceived lack of experience, looking too young for the role and managing older staff. Building credibility emerged as one of the greatest challenges for GenX leaders, who perceive their age adds additional challenges to the process. The challenge of being perceived as credible mounts when their age predisposes their followers and communities to hold concerns about their experience and ability. To address this, leaders focus on building relationships, formalising their commitment to the school and managing expectations (Edge et al., 2016b). Within these findings, we noted several city-based differences, specifically in how age, gender and race dovetailed to create different perceived challenges in building credibility amongst their staff and community colleagues.
In our second year, we sought to understand more about GenX leaders and how they conceptualise their person and professional identity and its influence on their leadership. To this end, we developed our Global City Leaders (GCL) Professional Identity Tool (PIT) and employed it to facilitate discussions about identity with 60þ leaders. Our within-and between-city analysis of the data identified two strikingly different contextual patterns for women leaders. First, London-based women leaders report vastly different experiences from their Toronto and NYC peers . More London women discuss delaying or foregoing children -or multiple children -due to school leadership pressures and accountability. Second -and the focus of this paper -leaders in each city discuss race/ethnicity within their school leadership roles in very different ways. Our evidence illuminates how, within our small group of participants, race/ethnicity is constructed and perceived differently in each city and school system.

This paper
Echoing scholarly calls to recognise the importance of context on leaders' construction and negotiation of identity (Crow et al., 2017), this paper illuminates how a small cohort of GenX leaders experience and describe the influence of race/ethnicity within their own city-based leadership practice. We review leadership-related research and theory exploring social identity theory, organisational fit and in-group prototypes. Focusing on evidence generated by our PIT card-sort tool, we describe our data gathering and analysis process. We concentrate our analysis to frame how school leaders explicitly discuss the interrelationship between their own personal racial/ethnic identification and its mis/alignment with their school-level communities. In turn, we explore how different city contexts may influence leader experience of in-groups and out-groups and the related leadership challenges and opportunities. In conclusion, the paper reflects on what counts as diverse identities in each city and posits suggestions on how leaders can navigate the perceived social and structural challenges.
In this paper, we strive to make three unique contributions. First, our evidence is gathered from a new generation of leaders who are underrepresented in the research literature but important to the future of global school systems. Second, our data look exclusively at leaders in urban contexts across three very different education systems. The analytical merit of the international comparative perspective provides interesting reflections on what counts as diversity in identity reflections in each city. Third, through this comparative lens we consider a more clearly and well-defined image of how identity may be evolving for different groups of leaders. Our policy/practice studies (Armstrong et al., 2013a;Armstrong et al.., 2013b;Edge et al., 2013;Mejias et al., 2013) note significant differences in the role of leaders and the structures, supports and levels of accountability in each system. The generational and international comparative elements of the study create a departure point for considering how leaders are approaching their roles, attempting to build credibility and coming up against the historical and political contexts of the wider education systems, cities and countries. These inspire discussion of the potential implications for education systems in the future.
Our intention remains to outline how context may influence leader experience and identity. Although important, this paper does not discuss our additional evidence examining the intersection of race/ethnicity with other aspects of intersectional identity and its influence on leaders. To simplify the presentation of our early thinking on the topic, we focus on here on race/ethnicity and do not stratify our analysis by leaders' gender, background and years of experience.
What a difference a city makes: setting out city-based contexts Although London, NYC and Toronto share many commonalities, each is nested within its own greater historical, educational and political context. We will not outline the historical evolution of race/ethnicity within each city and country here but will do so in the findings section as appropriate. Instead, in this section, we focus strictly on what Crow et al. (2017: 271) describe as the institutional context: 'the people in positions of power who control process and product, and the structural regularities of schools and districts that govern the actions of school principals'. In each city, we highlight leader certification, hiring, allocation practices and roles. We describe accountability practices and the degree of decentralisation. These factors will, in turn, be examined in the light of their possible influence on leaders' identity. The following summaries are drawn from our original aforementioned GCL policy synopses of London (Armstrong et al., 2013a), New York City (Mejias et al., 2013) and Toronto . As each jurisdiction has legally and politically acceptable language to describe race/ethnicity, we adopt the local city-based language as appropriate.

London
London school leaders work in the highly decentralised and pressurised English education system that has experienced decades of rapidly evolving policy (Armstrong et al., 2013a). Schools are inspected and graded by Ofsted (the national inspectorate), and national student testing provides frequent and public 'accounts' of school-level success, often using student scores as proxies. Teacher and leader recruitment challenges persist and the government sponsors fast-track teacher and leader programmes. New types of schools (academies and free schools) are granted increasing autonomy from local authorities (LA). The National Professional Qualification for Headteachers, once mandatory for school principals, is no longer a legally required certification for school leaders. Academies and free schools can also hire unqualified teachers. Principals are hired locally by school-level governing bodies that may comprise teachers, parents, community members and, if applicable, LA members. Principals are responsible, with governors, for running the school, including caretaking, budgeting, planning, hiring all staff members, teaching and learning and professional development, among other tasks. There is persistent recognition of the excessively high number of white leaders in the system, possible exacerbated by local school-level hiring practices which may be more influenced by unconscious bias (Edge et al., 2016b).
Across inner London's 14 LAs, 34% of primary and 36% of secondary students receive free school meals; 49% of secondary students speak English as a second language; and 81% are described as non-white British (The Guardian, 2012). The 19 Outer London authorities report that 20% of primary and 18% of secondary students receive free school meals; 32% of secondary students speak English as a second language; 59% are reported as non-white British (The Guardian, 2012). Across both Inner and Outer London Authorities, 25% of primary and 23% of secondary students receive free school meals; 37% of secondary students speak English as a second language; and 66% are reported as non-white British (The Guardian, 2012). In 2015, the The Guardian (2015) also reported that 6.7% of the English teaching workforce are from a minority ethnic group compared to 13% of the overall population. Only 2.8% of headteachers are from minority ethnic groups.

New York City
The New York City Department for Education (NYC DOE) guides the city's schools, within national and state-level education policy infrastructures. The DOE includes a series of networks, support organisations and districts to support schools. Published school-level quality assessments and student testing are used to encapsulate and publically share school and leader performance. Parents previously provided anonymous feedback on schools and principals via an online survey; however, this has recently been removed. A growing number of charter schools have greater freedoms from DOE regulation and structures. Well-established government-sponsored fast-track programmes exist for teachers and leaders. All principals are required to have a School Building Leader certificate. Leaders are hired by the DOE and allocated to schools via networks and districts. Owing to shortages, fast track teacher and leader programmes are present in the city. In most schools, leaders do not have complete freedom to hire teachers but are responsible for the quality of teaching and for implementing relatively rigorous teacher assessment. With the expansion of school-level autonomy, principals have more freedom over teacher hiring and professional development.
While demographic data on NYC-based school leaders are not public and are difficult to access, public school student demographic information indicates that NYC DOE educates 1.1 million students at over 1700 schools with over 75,000 teachers. Across primary and secondary schools, 76.5% of students are recorded as meeting the poverty threshold and 40% speak English as a second language. Across all schools, DOE reports students as: Asian (15.5%); Black (27.1%), Hispanic (40.5%) and White (14.8%). We remain unable to find accurate or even proxy diversity demographic data for NYC school leaders as the NYC Workforce Profile Report (2015) provides race/ethnicity data for DOE employees as an overall statistic (48% White: 52% Minority). The Independent Budget Office report (2014) does highlight teacher demographics (58.5% White: 19.6% Black: 14.4% Hispanic). However, as both sets of statistics are not disaggregated by job category, it is not possible to present leader demographic data.

Toronto
In Canada, provinces/territories hold responsibility for education policy and management. Ontario represents, in comparison to London and NYC, a highly centralised, low-stakes accountability system. Toronto has four state-funded school districts nested within the overall provincial Ministry of Education (MOE). MOE policy shifts often represent tweaks in training and curriculum that appear slight when compared to the structural overhauls characterised by new schools and fasttracking found in London and NYC. Leader and teacher recruitment and retention remain unproblematic. The Ontario Principal Qualification Programme Certificate is required of all school leaders. Districts hire and allocate leaders to schools. Principals have little control over teacher hiring and work within strict union-driven regulations to assess teachers.
To understand more about the demographic composition of the education system in Toronto, we use the published evidence of the Toronto District School Board (TDSB) as a proxy for the city. However, the TDSB as the largest and only English language non-denominational district, is by far the most diverse. In 2015, TDSB educated 246,000 students in 584 schools. In 2012, student demographic background was reported as 29% White and 71% Visible Minority, with language ability described as English only (44%), English and another language (22%) and no English (34%). In total, 115 languages are spoken by TDSB students. While historically, national Canadian teacher demographic data included race/ethnicity data (Ryan et al., 2009), this practice appears to have ended with no publically available national datasets currently available. Similarly, demographic data on school leaders is also not public and is difficult to access in Ontario both provincially and locally.

Literature review: Social identity and education leadership
Conceptions of professional identity abound within education research, with much focus on teacher identity (Bloom et al., 2015;Blum, 2015;Collay, 2006). The ways in which the personal and professional identities and experiences of teachers 'interact and relate to each other' (White, 2009: 861) provide much of the groundwork for examining how the 'school leadership identities' (Crow et al., 2017: 265) are constructed and change over time. Crow et al. (2017: 268): Identity is the way we make sense of ourselves to ourselves and the image of ourselves that we present to others. It is culturally embedded but subject to change. There is unavoidable interrelationship, also, between the professional and the personal.
However, recognition of the ambiguity of leadership as a personal identity, for example, adds complexity which merits detailed investigation (Day and Harrison, 2007;DeRue et al., 2009). Crow et al. (2017: 274) highlight the gaps in school leadership identity research including the tensions, conflicts, similarities and differences between principals' 'educational values, emotions and beliefs' and the expectations held for them.
To situate the overall design and analysis strategies of the GenX leaders' professional identity reflections, we briefly summarise research emerging from sociology, psychology and business on identity. In turn, we focus more intently on social identity theory and in-and out-groups, which frames our analysis of how leaders experience and discuss race/ethnicity within different citybased contexts. We also highlight evidence pertaining to educational organisational fit.

Professional identity research
There are many ongoing debates within professional identity research. For some, multiple identities are inherently in conflict and can only be 'resolved' by 'ordering, separating, or buffering them' (Ashforth and Mael, 1989: 35). Conversely, identities can also be viewed as intersecting dimensions which can be integrated or engaged simultaneously (Jones and McEwen, 2000). Identities have also been viewed as actively formed through individual agency or whether they are 'socially "given"' (Coldron and Smith, 1999: 714). For example, identities may be forced on people by oppression, marginalisation, difference or privilege (Jones and McEwen, 2000) and/or shaped by institutional environments (Chreim et al., 2007). Leadership then becomes a particularly interesting phenomenon since, in most contexts, it is regarded as a 'positive personal identity and social designation' (DeRue et al., 2009: 219). This complicates the closely related notions of power and empowerment, blurring the intersections between social identities that are oppressive or discriminatory and that might enable the individual to challenge, subvert or vanquish such oppressions. Haslam et al. (2009) also challenge classifications of socio-demographic groups, such as gender and age, reminding us that social identities are not only strongly relative to context but also dynamic and fluid. Accordingly, in this paper we consider narratives of identity salience as metaphorical snapshots of an ever-moving scene, negotiated between the 'inner' and 'outer' self and group norms, and hugely affected by (and in turn influencing) contexts, policies and practices.
For the purposes of this paper, we rely on Ashforth and Mael's (1989) conception of personal identity as the personal traits or characteristics associated with a person's actions or other attributions. These can include abilities, interest and bodily attributes (Ashforth and Mael, 1989). We have adopted Ibarra's (1999: 764), Schein-inspired (1978), definition of professional identity as 'the relatively stable and enduring constellation of attributes, beliefs, values, motives, and experiences in terms of which people define themselves in a professional role'. Ashforth and Mael (1989) posit that social identification with a particular identity is based on various categories of prototypical characteristics or social categories including gender, age, race/ ethnicity. Individuals can 'locate or define' themselves within different environments. Social identity, in turn, is how the individual or the community ascribes membership within different groups sharing the same characteristics. Perceived or self-ascribed membership is shifting and often relationally linked to group identification (Tolman, 1943).

Social identity theory: In and out-group prototypes
Looking beyond internally constructed personal and social identities, Reicher et al., (2005: 555) posit the idea of leadership being dependent upon the notion of a 'shared social identity' through which leaders and followers are 'bound together'. Hernandez et al. (2014Hernandez et al. ( : 1884 suggest leaders spend more time 'crafting and maintaining positive leader-follower relationships' than 'projecting their own personal characteristics or transmitting the organisational mission' in order to 'enhance follower trust'. Such relationship-building between leaders and followers is highly influenced by the expectations of followers that leaders will have an 'ethic of care', a moral compass guided by the specific needs of followers (Gabriel, 2015: 328).
Shared social identity involves a process of depersonalisation, which entails differentiating the group from the 'other' by enhancing stereotypes of common 'perceptions, attitudes, feelings, and behaviours' (Hogg, 2001: 187) and thus constructing group norms beyond individual identities. These shared norms form the groundwork for 'in-group prototypes' (Hogg, 2001: 187) -similarities shared by the group that distinguish the in-group from out-groups. Oakes et al. (1991: 141) illustrate the importance of in-group prototypicality, indicating that a shared internalised group norm is more influential on a group member's actions than external group pressure. Platow (2001: 1471) argue that leader success is likely to depend on 'the capacity of the leader to act in a way that affirms and advances the in-group's position relative to salient outgroups'. More specifically, they also assert that 'leadership centers around the process of creating, coordinating, and controlling a social self-categorical relationship that defines what leader and follower have in common and that makes them "special"'. Accordingly, to lead a cohesive group, high levels of ingroup prototypicality are needed (Hogg, 2001;Reicher et al., 2005).
The degree to which in-group prototypes are stable or fluid, strong or weak, depends in part upon the context but also upon the degree to which the leader maintains and identifies with a consensual prototype (Hogg, 2001;Haslam and Platow, 2001). For example, Platow (2001: 1477) posited that if a leader 'represented and affirmed the values of the group (and did not merely "sit on the fence")', followers were more likely to 'go the extra mile' for their leader. Hogg (2001: 187) identifies a possible 'prototype-based social attraction gradient' within each in-group. Despite the flexibility of individual and social identities described above, validation of leadership is strongly influenced by the extent to which the leader 'fits' within the in-group prototypes and represents the 'collective' aspirations of group members Platow (2001: 1478).
While powerful and offering potential insight into the selection, development and support of school leaders, in-group protoypicality has rarely, if ever, been explored in the educational leadership literature. In this paper, protypicality and in-groups are important concepts for understanding how leaders experience and discuss race in their schools and leadership. Hogg (2001: 195) states that 'if social minorities intrinsically less prototypical than majorities, then minorities will find it more difficult to achieve and maintain an effective leadership role' (Hogg and Terry, 2000). For these reasons, we have chosen to explore how in-group theory can shed light on the experiences of participating GenX Global City leaders.

Organisational fit
'Organisational fit' is an underexplored but important concept within educational leadership. Alongside the usual job selection criteria, recruiters may be looking to identify if the candidate may 'fit' within the organisation's culture. Tooms et al. (2010: 108) examine fit as a 'condensation symbol', a word that 'stir[s] vivid impressions and a listener's most basic values' but that may mask underlying complexities: 'The ease and accessibility of this word blinds us from the interplay of social constructionism, identity theory and hegemony that reveals the complexity of its meaning' (Tooms et al., 2010: 118). Tooms et al. (2010: 123) conclude that school leader selection and support is 'not always an egalitarian process' with assumptions that if a leader fits within one element of an expected identity role, he or she will meet the other expectations of the role, and vice versa (Tooms et al., 2010: 109). This may be entirely unintended and may continue beyond the recruitment process and into the interactions between the school leader and the school community. Therefore, even where school leaders are deemed to fit within the organisation during the selection process, certain identity constructs and behaviours may distance them from the in-group prototype and cause challenges and conflicts throughout their working life. Tooms et al. (2010: 116) suggest that 'the administrator needs to constantly read the community's spoken and unspoken rules involving fit and then adjust his or her everyday actions to maintain community support'. This demanding process is, according to Fine andHallett (2014: 1774), enacted through 'shared awareness and memory' which recognises the importance of the community's 'cultures, traditions and histories'. Again, this will become important in the context of fit or in-group linked to invariable components of a leader's identity, including age, gender and race.
While Tooms et al. (2010: 121) note that 'fit is not just about racism or sexism' (rather, it is about values), there are certain characteristics that make people more likely to be affected by the hegemonic processes described above. For example, Hogg (2001: 195) suggests that 'social minorities (e.g. based on race, ethnicity, gender, disability) may find it difficult to assume leadership roles in some contexts'. Lumby (2014: 39) describes one response in which individuals may distance themselves from a disadvantaged group, for example, by denying that the disadvantage exists.
However, it may be that experience provides insight and ability to more successfully anticipate and navigate identity-related challenges. Lord and Hall (2005) argue that, as leaders become more experienced and more skilled, their identities become more complex. In turn, leaders focus more upon relational, collective and value-based identities than individual ones. As Day and Harrison (2007: 367) note, 'A leader with a sophisticated and complex identity is able to draw from individual, relational, and collective levels of self-concept'. The research described in this paper provides an interesting snapshot of a small group of GenX leaders' perception of professional identity across three diverse city-based contexts. The evidence enables us to examine the relative salience of leadership identities and manifestations of in-groups, prototyping and fit through a comparative lens.

Our overall GCL and identity-driven research strategy
The Global City Leaders project set out to understand the experiences, lives and aspirations of the new generation of school leaders via a three-year mixed qualitative methods design involving three city-based policy studies, 12 GenX leader networking events, 125 individual interviews and nine school-based studies. We recruited city-based cohorts of 20 to 25 GenX principals and viceprincipals via invitation emails through our district, organisational and advisory group contacts . Two annual interviews with each leader examined career choices and experiences, professional identities, future ambitions and possible emerging GenX leadership model(s). All 120 interviews were fully transcribed. Using Dedoose (an online Cloud-based encrypted qualitative analysis program), we applied a multi-staged coding process and structure Miles and Huberman, 1994). Transcripts were analysed using full-grounded theory approach (Strauss and Corbin, 1994), which resulted in, for example, an application of 25,000 codes across the 65 interviews conducted in the first year . The first-and second-year findings have been reported in our research reports (Edge et al., 2013c;2013d).

Identity-related data gathering and analysis strategies
The data collection within the study focused on gathering leaders' perceptions and experience of their own personal and professional identity. For the purpose of the paper, we re-examined the evidence gathered during the study employing a social identity theory lens. Identity-related evidence was gathered during second-year interviews using our PIT tool . The PIT prompts leaders' conversations and reflections on identity via a card-sorting activity. This ensures breadth of the aspects of identity discussed while maintaining consistent language. It also encourages participants to share the narratives of their experience. To understand GenX leaders' development and perspectives on professional identity, we asked leaders to identify up to three leadership roles that influenced their sense of self-and/or professional identity. For each selfidentified role, leaders outlined their age, role, title and details of the experience. In turn, we asked leaders to reflect on how they were perceived when they 'walked in the door' and to select the cards from the PIT pack that best described their experience.
Our PIT card pack contained 15 business-card-sized cards, each containing one identity-related word or short phrase generated during the first year of the research. The final set of PIT cards included gender, race/ethnicity, nationality, language, age, experience, sexuality, physicality/ height, relationship status, parental status, family history/background, image, work ethic and work job/role. We did not ask leaders to rank the importance of the identity cards but merely use them as a springboard to discuss their experience. Leaders shared examples of how they knew people perceived them for each card and how the experience influenced their sense of professional identity. We also asked leaders to share strategies they adopted to address their positioning and, if applicable, how the experience had influenced their approach to leading.
To ensure the validity of the use of the PIT across the study, each leader was given the same pack of cards and provided with the verbatim list of instructions. This enabled all leaders to consider the same set of dimensions of identity, bringing consistency to examination of this complex concept. It presented each leader with the same possibilities while allowing time and opportunity for reflection on their personal experiences. This also provided consistency across interviewers. We did ask each leader if they wanted to add any additional cards to the list however, none were added. The cards narrowed the scope of the narratives in terms of language and terminology used, which allowed for easier coding of interviews and analysis. Thus, this method can be used for replication in other contexts and populations.
We designed a specific strategy for extracting and analysing identity-related transcriptions. For each city, we prepared an Excel sheet template with one record per leader which included the following demographic information: participant number, role, age, gender, relationship status, sexuality, relationship status, parental status, children (number and age), years in the role, years in the school, phase and fast-track status. We used two approaches to extract identity-related data. First, the above-mentioned critical roles were numbered, extracted and imported into the datasheet. Second, all additional discussions of identity-related experience were extracted and imported. However, these were not the explicit link to specific critical roles. Each unit of data was placed in its own individual line of the Excel sheet.
The first tranche of analysis involved assigning identity-related incidents/roles to one of four categories: current role (job-related), current role (life-related), past educational role and other role. The subsequent strands of analysis involved tagging each unit of evidence with relevant identity codes, extracting discussion of strategies employed as a result of their identity-related experience, and isolating mentions of identify-related experience on their own development as leaders. Final rounds of analysis also applied in-/out-group tags to the coding structure.
For this paper, we reviewed each race/ethnicity-related role/event mentioned by participating leaders during the PIT discussions. We then conducted a more refined analysis by coding all examples as being positive or negative. We also examined instances where leaders described professional identity related to race/ethnicity. Most specifically, these instances almost always highlight alignment, or a lack thereof, of the leader's race/ethnicity with that of their school community. Our reanalysis for this paper focuses on instances where leaders articulate their identity in concert or in conflict with that of their teachers, parents or community members. In doing so, we are able to illuminate how professional identity, race/ethnicity, leadership and local context may be intertwined.

Findings
In this paper, we report on instances of leaders discussing their own race/ethnicity in alignment or misalignment with their school community using representative quotes from leaders' own narrative descriptions. To do so, we present within-city mini-cases of the findings, beginning with an introduction to the number and description of participants, followed by a description of how they verbalised race/ethnicity. Implications of the findings are considered in the discussion and conclusions.
Distinct differences emerged when examining how leaders in each city voiced their experience and views of professional identity. While we provide more thorough and robust analyses of withinand between-city patterns in our aforementioned reports, here we focus on two patterns of discussion that emerged across all three cities: alignment of personal and community identity and proving professional ability. All quotes are anonymised, but include role (i.e. headteacher/principal ¼ HT/P), gender (male/female ¼ M or F) and self-identified race/ethnicity (white ¼ W, South Asian ¼ SA, black ¼ B, mixed/bi-racial ¼ M/B). We are mindful that each city has its own, nationally embedded, nomenclature to describe elements of race/ethnicity and we have tried to reflect these throughout.

London
Twenty young London leaders participated: 15 female and five male leaders serving as headteachers (8) and deputies (12). One leader identifies as South Asian, 19 as white. Seven leaders discuss race/ethnicity during their reflections on identity and its influence on experience and work including the one South Asian leader. For most leaders, the focus of their discussion almost exclusively rests on in-and out-group alignment and building professional credibility. The relatively low number of race/ethnicity references may be due to the homogeneity of participating leaders in our study and the overall leadership population in the country. There was less discussion of race/ethnicity among white leaders in London compared to New York City and Toronto. This lack of overt discussion of race may be attributed to a lack of willingness to discuss challenging issues, including difference and race in the UK, which tends to be less common than in other participating countries. The overall homogeneity of leaders in the country may be due to the fact that London school leaders are hired directly by the schools and their governing bodies themselves, rather than districts, as is the case with the majority of Toronto and New York City participants which may influence an overall increase in diversity (Edge, 2016b). This may, in turn, lead to more harmonious relationships between leaders and schools and communities, with London leaders experiencing less resistance due to the fact they were chosen by the community members themselves.
Alignment of personal and community identity. London leaders in our study rarely discuss race/ethnicity. On the limited occasions when they do, leaders share reflections on misalignment between their own personal race/ethnicity and that of their school communities. Most white leaders express awareness of differences between themselves and their students and see their responsibility as being role models. The one South Asian leader discusses deliberately seeking a role in a school that reflects his personal racial/ethnic community. This leader also expresses the challenges of seeking a race/ethnicity match as there remain substantial gaps related to language, socioeconomic status and educational background that create misalignments with his school.
Proving professional ability. Proving professional ability or developing credibility was only mentioned by two leaders. One white leader shares how governors challenged her to examine how the community could find the school's leadership credible when they did not reflect the race/ethnicity composition of the school community, People make assumptions about all people. I've had it quite a lot in my current role. Governors, stakeholders or people who are involved in the . . . school have said, 'You need to hire people who don't look like you, because you're very white and very middle-class. When you stand in front of people and you say, "We understand what it's like to look after a child', it's very hard to believe you"'. (F/HT/W) The South Asian leader also reports slight pressure to prove his competence to offset any perceptions of his selection as a 'token' representative minority leader, explaining, When you're appointed as a member of the minority, where there's a higher proportion of ethnic minority students, you have to be clear that you can do the job and you can do it really well. There is always just that little bit in the background, 'Were you [hired] because you fit a profile?' (M/DH/SA)

Toronto
In Toronto, 10 men and 10 women participated with equal numbers serving as principals and viceprincipals. Leaders self-identify as white (14), East Asian (3), black (2) and mixed race (1). Twelve leaders, of which half self-identify as Asian, black or mixed race share a more nuanced commentary on race/ethnicity and professional identity. They share a much more nuanced view of identity and leadership than their London counterparts. This may directly relate to the more diverse composition of participating leaders and a greater willingness to openly talk about race/ethnicity within the language of Toronto's explicit commitment to multiculturalism and pride in its diversity.
Alignment of personal and community identity. Asian and black leaders consistently believe their ingroup membership makes positive contributions, providing cultural understanding, credibility and community representation. One leader states: My race . . . it can go both ways probably. To the students, it's a huge plus . . . There are benefits of alignment between leader and community identity and also the ability to cross cultural boundaries, as one mixed-race leader says, Proving professional ability. There appears to be variation in leader experience in building professional credibility. Asian leaders consistently discuss the built-in credibility they gain from working in schools with highly aligned in-group community populations and the positive stereotyping of Asians within Toronto. However, one black leader experienced challenges in building professional credibility with peers that the leader explicitly linked to perceptions of readiness for leadership and the associated challenges with gaining promotion. A white leader describes the desire to be culturally sensitive to counter the drawbacks of not being representative of the community. However, even with significant experience, her credibility has occasionally been questioned explicitly due to her whiteness: I had to get some buy-in. [I am] a white, woman, European. I don't have the social identity of being a young black male who might have a special needs background. So they look at me and [think]: 'What qualifies you to be up there talking to us about kids you don't even know and you've probably never taught?' (F/VP/W) Addressing the credibility-related influence of mismatch between leader and school community race/ethnicity, one leader describes benefit from social bias advantages afforded white and male leaders, I was in a South Asian [community]. In some respects, the [community] expects to see a white man running the show and that gives them confidence. There were some fathers [who] would come in that would not speak to her (the principal), that would wait for [me]. (M/VP/W)

New York City
In New York City, 20 leaders (12 female, eight male) shared their insights, comprising principals (16) and vice-principals (4). Leaders self-identified as white (14), black (3), Hispanic/Latino/a (2) and bi-racial (1). Sixteen leaders raised race/ethnicity and its influence on personal and professional work and life. Six of these leaders self-identity as black, Hispanic/latino/a and/or bi-racial. Discussions of race/ethnicity were by far the most prevalent and passionate in New York City, which could reflect the more diverse cohort of leaders and/or their willingness to talk about race/ ethnicity. Leaders from all backgrounds appear much more reflective of the influence of their race/ ethnicity on their work. However, white male leaders, while reflective and aware, appear to be the least negatively influenced by the misalignment of leader and community culture.
Alignment of personal and community identity. Several leaders explain the importance, in NYC, of alignment between leader and community racial/ethnic composition. While leaders suggest there is not a formal policy of matching leader and school community race/ethnicity, several believe there are district-level attempts to align leaders with communities if possible. One leader explains: '[My training programme leader] . . . said: "There's going to be some districts that you're going to apply to, and . . . they're not going to let you work there"' (F/P/B).
Several leaders also share that geographical areas within the city demand a more nuanced need for alignment that includes, but extends beyond, race/ethnicity and the challenges faced by leaders on occasion if they do not personally reflect the dominant demographic of the community. In NYC communities are different even within the same community. What makes leaders viable in one community [varies]. People may look at you and say: 'You may be from the community -but not from the community at all. You don't "get it."' (F/P/B) Another black leader who did not culturally represent the demographic of her school faced community scepticism from members who said the school culture had changed under her leadership: The [previous] principal was Hispanic and [the school] has a Hispanic student population. I was asked if I speak Spanish. I started speaking to another principal and I said: 'What's different? How come [parents] say the culture's changing? How come they're saying it's not a family feel anymore?' He just said: 'Are you that naïve? You're not the right colour'. I didn't want to believe that was true. He said: '[Parents] aren't sure how a black leader, or someone who identifies as black, is going to run a school that was [historically] Hispanic'. (F/P/B) However, a leader experiencing high levels of in-group alignment along race/ethnicity and economic lines speaks about the benefits for students, 'Kids are more focused on nationality, race -and ethnicity is a BIG thing for them. Because of [my background], I AM so approachable to them'. (F/P/L) Some white leaders acknowledge the challenge of misalignment but see their function as advocates of the communities and use their 'whiteness' and inherent social capital as a leverage to resources and access they are afforded: There are things that I need to advocate for. I know how to make people really uncomfortable, too. I really worked to build allies in the building. I know my limitations when it comes to who I am, and how these parts affect my identity. I can say some things that other people can't. There are some things I can say that black principals can't. I can talk about things like, in terms of how the system works. (F/P/W) Proving professional ability. Unique to NYC was a universal discussion about the relationship between in-group alignment of leaders and communities and the credibility challenges that misalignment provokes. Those leaders who are not seen as representing their school communities often get overt backlash and face many challenges. One leader states: 'I had to overcome [being different]. I still always have to overcome the idea that there [are] some people that truly believe that the leader of the school should come from the community of the school'. (M/P/L). Two white leaders who did not represent the school community describe the challenges and having to forge other strategies to connect with students and communities: I was a villain for the first seven months of the job . . . then we got our test data back and the school did so much better than it had in previous years. We also had a quality review that [improved] then all of a sudden it [improved]. (F/P/W) The second leader shares, I've been called the Devil. I've been called [names]. But [parents] will go talk to my co-ordinator, they will go talk to my [teacher]. You've got to figure out your leverage. I connected with the children and it doesn't seem to be the colour, the pigmentation of my skin rarely causes a problem or gets in the way of the relationship-building (with kids). (F/P/W) In New York City, unlike other cities, white leaders often discuss needing to prove their credibility, as they are not immediately trusted by their school communities. One white leader describes how the leader following her in post is Latina and the difference in their experience, Being a Latina, she's been able to make strides with the parent community that I was unable to. When I was really hard on people, or had really high standards, or intervened in parent-child stuff, I was just being somebody who was 'doing' something to the [community]. Sometimes, not all the time, that was the perception. [The new principal] is perceived as their partner automatically because she 'cares' about them. Being white was a really big deal. Huge deal. There's a reverse racism that happens here. (F/P/W) Although some leaders are not trusted or do not have credibility as they are not representative of the school community, their whiteness is still seen as beneficial to the school and students: People see that I'm white. I'm different. I don't belong here. That's something that's working against me in some ways. The [community] like that I'm white because this means that I have access to things that they don't. Our first year here, parents were coming in and requesting white teachers [saying], 'Can my kid be in a white class?' There's this idea that if you're in a white class, with a white teacher, and have a white principal, it's better. I think it's an ingrained area of racism that exists. That people see white kids go to college, and white kids go to good schools, and if you're in a school with a white principal and white teachers you get these things. (F/P/W) Similarly, leaders also talk about recognising their own limitations and explicitly trying to ensure that the staff complement is diverse if the leadership team is not: My hirings are very diverse. I'm not. I'm diversifying so my staff reflects the community. I also leverage that where I can't make that connection. There are some parents who do not want to hear what I have to say. (M/P/L)

Cross-city reflections: Leader/community in-grouping
The importance of a racial/ethnic match between leaders and communities was most frequently discussed in NYC by both white leaders and leaders of colour. Even throughout our relatively small sample, comments about mismatch rarely emerged in schools with mixed community composition and appear more likely in schools with students from predominantly one racial/ethnic community. While we did not specifically examine the school community composition in each leader's school, we observed more multi-ethnic schools in London and Toronto than we did in New York City. This may be the result of the overall diversity of the population of each city, the citybased geographical distribution of members of specific ethnic and cultural groups and the history of urban planning and social assistance in each city. The increased diversity within schools in Toronto and London may also be one reason why there was less discussion of the importance of leader-community in-grouping in these cities.
Based on our overall observations, London leaders are predominantly white. Combined with the overall shortage of school leaders in London, there appears to be little pressure on leaders to reflect their communities. Again, this may be due to shortages of leaders, but even more so of leaders who represent black and minority ethnic (BME) groups. Another contextual influence that may contribute to a quiet or more subtle conversation about race/ethnicity in London could be the very localised nature of school-leader hiring. As local schools, with their governing bodies, hire the school leaders, this may nurture a certain level of confidence among the community that the best leader has been chosen for the school. However, the often-negative tone taken by the English media and government about schools and school leaders may also influence leader-community relations in juxtaposition with patterns of conflict-avoidance within English communities (Meyer, 2014).
Conversely, in Toronto, participating leaders openly talk about the benefits of matching leaders to school communities to create in-group relations. However, this was primarily when Asian leaders worked in Asian communities. Participating black leaders did not feel as confident that race/ethnicity was always viewed as a positive. Toronto's explicit multicultural history and shared narrative was highlighted by leaders as a reason why this may be the case. Schools are assigned leaders; schools have very little choice regarding the leader they are given. At the same time, there is little, if any, public negative discussion by government about schools and school leaders. While this serves to build public confidence in schools and the education profession, the by-product may also be a smoothing of community-leader relationships.
Finally, New York City leaders proffered the most examples of leadership challenges due to mismatches between leader and school race/ethnicity. New York City has experienced a more sustained negative discussion around education and education provision and a greater publically discussed discrepancy about the quality of schools in high-vs low-income neighbourhoods. Rooted in historical traditions and contemporary conflict, there appears to be a greater willingness on the part of school communities to express explicit distrust in leaders, especially those who are not representative of the community. This neighbourhood-based variation may be due to the historical relationships between different race/ethnic groups as well as the overall hierarchical nature of various elements of socio-economic status and gender, for example, within the school system and beyond. The apparent distrust of leaders may also be rooted in the pervasive shifts in school organisation and design across the city as well as the introduction of small schools, which led to the very public appointments of fast-track leaders in schools. The tolerance for leader 'fit' appears to be lower in NYC. Thresholds for tolerating or working with out-group leaders appears even higher in some neighbourhoods and for some leaders. As we have seen, for young white women leaders in predominantly black neighbourhoods, the experience can be difficult. Young black women leaders in Hispanic neighbourhoods also experience challenges. Interestingly, men leaders do not talk about these challenges as often. Men were more likely to talk about how their own position as white male leaders creates smoother relationships with communities regardless of in-group alignment.
Cross-city reflections: Proving professional ability and building credibility Riecher et al. (2005: 563) argue that the 'possibility of leadership is dependent upon the existence of a shared social identity. Without such an identity, there is nothing to bind leaders and followers together, there is no consensus for a leader to represent and therefore leadership is impossible'. In this light, examining school leader experience related to racial/ethnic in-group experience sheds light on an important area of leadership work and development.
Participating leaders express distinct differences in how their own racial/ethnic in-groupness influences their leadership and relationships with others. However, the strategies they adopt for overcoming these challenges and bridging the gaps created by their in-group alignment, remain very similar across all cities and all leaders. Our evidence demonstrates differences within each city, and even neighbourhood in the case of NYC, in how in-groups membership influences leader experience and interpretation of their own professional identity and their role. However, we observe less variation between cities in how leaders feel they need to prove their professional ability and develop credibility within their schools and school communities. While there is variation in the racial/ethnic composition of our participating cohorts in each city, they share membership in the same generational cohort. For these leaders, race/ethnicity also intersects with age and experience. Therefore, it is difficult to disconnect their experience as young leaders with their racial/ethnic group membership and mis/alignment with their communities.
Most leaders across all three cities experience resistance when appointed to a new school. They discuss resistance almost always coming from staff and community members that represent differences in age, experience and race/ethnicity. Leaders are not daunted by these challenges and share similar strategies across all three cities in their approaches to building credibility by prioritising: (a) explicitly working on relationships that build support for their school-level goals for students; (b) demonstrating their own skill and knowledge in the classroom; and (c) establishing their commitment to the school and remaining in post.

Conclusion
The overall number of leaders discussing race/ethnicity and the importance of in-group relationships between leaders and communities varied considerably in each city. The nature of how race/ ethnicity and in-group relations were discussed also varied. Our GCL project explored how leaders conceptualise the influence of race/ethnicity within their jurisdictions. The findings provide a helpful example of how context can influence leader perception and experience. Similarly, our findings illustrate the potential relative importance of leader and community in-group alignment between and, even, within cities. Participating leaders' experience articulates how 'gaining legitimacy is crucial and authority can never be taken for granted' (Crow et al., 2017: 9). Discussions of context in educational leadership and school improvement research often focus on country-or school-based contextual influences. However, our evidence illustrates that, even within the same city, leaders can experience challenges to their professional/personal identity in different ways. Collay (2006: 134) found that leaders experience challenges as group members' race, class and gender can 'add a level of complexity to each interaction'. In parallel, our evidence demonstrates how, on a small scale, city and local context may strongly influence a leader's ability to lead from the start. Research, policy and educational leaders need to recognise how contextually nuanced the experience of race/ethnicity may be and the level of increased complexity that leaders from dissimilar in-groups may face. Although many jurisdictions have initiated training and support programmes to build this awareness and practice, many still have a long way to travel to ensure diversity and its positive and negative influences are reflected.
It is important to understanding the challenge that community in-group alignment and support / belief in their leader can have on the school, the leader and the leader's career. Beyond challenges to a leader's ability to leading their school toward improvement, leaders facing school-level ingroup related challenges may also experience negative psychological consequences when their 'sense of social identity is compromised in some way' (Haslam et al. (2009: 5). For all leaders, understanding how nuanced the influence of context may be on their experience of leading is essential. As demonstrated in NYC, leaders would benefit from very local knowledge of their school communities and the shared wisdom of other leaders. More specifically, learning between leaders who had also experienced the different thresholds for tolerance facing leaders who are not from the racial/ethnic in-group could be incredibly helpful. Relatedly, Hogg (2001: 196) offers advice to leaders in relation to their own in or out-group positioning suggesting, 'prototypical leaders would do well to raise group solidarity and cohesion while accentuating their prototypicality, whereas nonprototypical leaders should lower solidarity and cohesion while accentuating how well they match leader schemas'. Building on the experience of leaders in our study would make a substantial contribution to academic and professional discussions.

Training and development
As policy and educational leaders strive to retain this new generation of leaders, preparing them for these challenges and opportunities will become increasingly important. Reicher et al. (2005: 563) argue that 'leaders are not passive onlookers when it comes to identity processes. They actively intervene in creating and redefining identities and thereby in creating and transforming their followers'. They continue and state that 'creativity of leaders is not limited to words and ideas but also extends to the initiation of structures which can make those ideas manifest'. Based on our evidence, we strongly believe in the important role leadership training plays in addressing the relational and structural strategies leaders can adopt to address challenges brought on by group misalignment. Assisting leaders in understanding how they may be perceived as leaders and how best to identify and overcome the influence of these interpretations will serve to support leaders who, for the most part, are already often considered to be too young for their roles. Building on thinking about in-group prototypes and leadership, several key ideas emerge that would be helpful for the development and training of school leaders. First, outlining the importance of leaderfollower relationships, Platow (2001: 1471) posit that leaders' success 'hinges on an ability to turn "me" and "you" into "us" and to define a social project that gives that sense of "us-ness" meaning and purpose'. In this light, developing leader capacity to 'act in a way that affirms and advances the ingroup's position relative to salient outgroups' will further enhance their ability to succeed'.

Prioritising continued diversification of aspiring and serving leadership cohorts
While training and development remain essential to support those leaders currently entering and serving in diverse educational settings, policy and practice leader must remain committed to recruiting an increasingly diverse cohort of leaders. No amount of training and development can replace the overall merit of ensuring a diverse cadre of leaders are hired and supported to serve diverse city-based schools and their communities. Talent-spotting, recruitment and development of diverse leaders that represent all community groups should remain the key priority for education systems.
The new pressures and demands placed on the role and education systems by GenX and, in turn, millennial leaders, may compound current leadership recruitment and retention challenges being experienced in most urban jurisdictions. City schools are demanding and, as Day and Harrison (2007: 367) share, a leader 'with a sophisticated and complex identity that is able to draw from individual, relational, and collective levels of self-concept therefore may have a strong advantage given the complexity of situations leaders face'. As many jurisdictions strive to evaluate the work and leadership of principals, our findings suggest that leaders may experience different levels of initial success and buy-in based on in-group relationships with school communities. To support, develop and retain leaders, these factors need to be considered in order to provide a level playing field upon which all leaders can thrive and deliver on their mission.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
This study was funded by the Economic and Social Research Council (RES-061-25-0532) in the United Kingdom.