Principals’ emotional identity – the Swedish case

ABSTRACT

This article explores Swedish principals’ emotional identity by employing an analysis instrument built up of three dimensions identified in the study by Crow, Day, and Møller (2017): “the professional,” “the situated or socially located,” and “the personal dimension.” The article builds on data obtained from a survey given to Swedish principals, completed in January 2019 (n = 1,286). The findings show that all three dimensions constitute important parts for Swedish principals’ emotional identity. The analysis also detected significant variances within the three dimensions regarding gender, age, and type of principal. Aside from these empirical findings, the article also offers a theoretical contribution because the analysis instrument succeeds in capturing the three dimensions. Thus, the article bridges data to theory in a way not previously done. The findings disclose the need for further studies in various contexts for a wider and more in-depth understanding of principals’ emotional identity.

KEYWORDS:

• Principals’ identity
• Emotions
• Sweden
• Leadership Theory

Introduction

The concept of principals’ identity has gained increased attention in educational research in recent decades (e.g. Beatty 2000; Blackmore 2004; Crow and Scribner 2014; Møller 2005; Murakami et al. 2016; Ritacco and Boliviar 2019; Tubin 2017). One overall conclusion revealed in the current body of research emphasised by Crow, Day, and Møller (2017) is that the construction of principals’ identity is situated in time, space, and place. Their comprehensive work also offers a synthesis of existing works, providing a framework for further research on principals’ identity construction and development. With the starting point in this valuable guide for future research, this article focuses on the ‘emotional dimension’ of principals’ identity and the ways in which leadership is formed according to the three dimensions described by Crow, Day, and Møller, namely, ‘the professional dimension,’ ‘the situated or socially located dimension,’ and ‘the personal dimension’ (see also Day et al. 2007; 2011).

There are strong arguments for focusing on the emotional dimensions of principals’ identity. As emphasised by Crawford (2007), emotions are intrinsic to how principals perform their roles and not separated from it. Crawford also stressed that all organising actions of principals are inseparable from and influenced by emotions. Similarly, Crow, Day, and Møller (2017, 269) underscored that ‘teacher and principals’ identities are constructed from and affected by the emotional influences of the self, role, and work context’. As mentioned above, they identified that principals’ identity and emotions are formed by three different dimensions. The concept of identity is also understood as something unfixed and adaptable, sometimes leading to ‘tensions’ between the three dimensions, which principals in turn have to handle.

Against this backdrop, the Swedish case stands out as very relevant. Sweden, for instance, is often portrayed as a noticeable example of decentralisation and market adoption (see e.g. Lundahl 2002a, 2002b), but the system has also gained increased interest in other countries, the United Kingdom, for example (see Allen 2010; Rönnberg 2015). Besides the intention to broaden the understanding of Swedish principal’s identity in a wider sense, this article approaches the concept from three different areas, namely, ‘gender,’ ‘age and length of time in the profession,’ and ‘type of principal.’ We believe that there are strong arguments to focus on these perspectives. Regarding gender, previous research on female managers in the public sector (Corin and Björk 2016; Corin and Björk 2017) showed that the managerial work includes a broad spectrum of demanding responsibilities and duties, and also that the managers’ work has expanded over time. In addition, female managers take great responsibility for many areas and tasks, thus, often performing duties that are not their responsibility. Concerning age and length of time in the profession, Swedish principals are less experienced, both as principals and teachers, compared with principals within the Organisation for Economic Co-operation and Development. Swedish principals also change jobs more often than principals from other countries in the OECD (2016). For instance, fifty percent of the principals in Sweden worked less than three years at their current school, making the turnover very high. To this end, what does this say abort principals’ identity and does identity change with increasing time in the profession? Regarding different types of principals, Sweden is at an important stage from a historical perspective, e.g. because preschool leaders are entitled to be principals from the 1st of July 2019 and they also have to attend the National School Leadership Training Programme (Bill 2017/18, 194). However, we still have limited knowledge on whether (and in that case how) emotional identity varies between principals in preschools and compulsory school and thus how they experience different elements of their leadership.

To enable detailed analyses of Swedish principals’ emotional identity, a former article took a point of departure from the work of Crow, Day, and Møller (2017) by creating an analysis instrument, structured with the three dimensions of their work (Nordholm et al. forthcoming). The current article employs this analysis instrument to explore Swedish principals’ emotional identity as well as what similarities and variances exist between different groups of principals. Empirically, the article builds on data obtained from a survey given to Swedish principals, completed in January 2019 (subsequently detailed). The following research questions directed the analytical work:

1. How is Swedish principals’ identity formed by professional, situated or socially located, and personal dimensions?

2. What variances exist between the different groups of principals, in terms of identity, and how can these differences be explained?

The article is structured as follows. First, the theoretical point of departure is outlined. Next, methods and data are described. Thereafter, the main results are presented. The article ends with a discussion and conclusion section.

Theoretical perspectives and the Swedish case

Regarding the Swedish school system, there are several systemic-structural aspects with relevance for principals’ identity to consider. As signposted, Sweden is often portrayed as a striking example of de-centralisation and marketisation (e.g. Blossing et al. 2014; Lundahl 2002a, 2002b; Rönnberg 2015; Seashore Louis 2013). During the 1990s, the local municipalities in Sweden attained and increased responsibility for the local schools as Local Education Authorities (LEAs) and overtook the ownership of their schools from the government. Thus, LEAs and local school principals received increased responsibility for issues related to school leadership and school improvement compared to the former system in which they had obtained detailed instructions from the government and its agencies. In addition, the school system was also opened up for ‘independent’¹ school organisers which together with public organisers form a school market in Sweden.

However, the challenges for school leaders and LEAs of navigating within and capitalising on the local freedom became evident over time (Swedish National Agency for Education 2000; Swedish National Audit Office 2004, 2011); therefore, a number of re-centralisation and re-regulation programmes have been launched and implemented since the early 2000s. The founding of the Swedish Schools Inspectorate in 2008, which regularly inspects and controls public and independent school organisers, is an important example of this ‘juridification’ and re-centralisation process (cf. Novak 2018). Recent research have also indicated that Swedish principals tend to support these re-centralisation reforms, confirming (institutionalised) ideas of a ‘strong state’ (Nordholm and Andersson 2019). It is, however, important to recognise that the de-centralised and marked-adopted school system appears to be strong from an international perspective. For example, ‘the Swedish School Commission’ (SOU 2017, 35) suggested in 2017 that further steps should be taken on the state level to prevent school segregation and to limit the possibilities for school companies to make a profit. Despite similar recommendation from the Organisation for Economic Cooperation and Development (OECD 2015), these proposals have received modest impact so far.

Thus, there are certain characteristics of the current forms of school leadership and management in Sweden to consider. A bit simplified, the division of responsibility for education is divided between the central government, which is responsible for ensuring the overall quality of education, and the LEAs or its independent counterpart who are responsible for providing education in preschools and schools. The majority of the public schools are organised by the municipality, with a school board and one or several superintendents. The Swedish school boards consist of appointed politicians, and they represent the political parties of the municipality council in such way that the school boards have the same political majority as the municipal council. However, it is important to note that Sweden has 290 local municipalities and despite the overall characteristics described, there is also a variety regarding local school governance and the municipal school organisation (Johansson et al. 2016). Both public and independent school organisers are, as noted, also evaluated regularly by the Swedish Schools Inspectorate. Worth highlighting, this increased and direct state regulation appears to weaken the relation between the local school principals and the LEAs of the municipality (Nihlfors and Johansson 2013).

Against this background and the current forms of leadership and school governance in Sweden, the choice of theoretical starting points becomes important. For this article’s relevance, previous research have shown that the concept of principal’s identity enables analyses, both in terms of the role(s) and activities of principals but also in terms of the impact that principals could have on schools’ results. Ritacco and Boliviar (2019), for instance, showed that successful leadership in schools largely depend on principals’ identities (see also Ritacco and Boliviar 2018). In the current article, the work of Crow, Day, and Møller (2017) on principals’ emotional identity formed by three dimensions (professional, situated or socially located, and personal) stood out as one potential theoretical starting-point relevant to explore Swedish principals’ identity. By relating to former works (e.g. Day and Lee 2011; Hargreaves 1998), Crow, Day, and Møller stressed that identities are neither intrinsically stable nor intrinsically fragmented – instead, they are more or less stable, conditioned by the capacities of teachers (and principals) to manage the three constituent dimensions of identity.

Outlining this theory, Crow, Day, and Møller detailed that the professional dimension reflects policy and social expectations of what a good principal is and leadership ideals. Further, this dimension is more open to influence from social trends and long-term policy. The professional dimension, however, can also contain competing and conflicting elements, for instance, regarding local and national policy, professional roles, research, leadership expectations, and so forth. Previous research have also shown that the professional dimension becomes important for successful school leadership – for example, principals with a strong identity have a positive impact on improvement results (Ritacco and Boliviar 2019). Regarding Sweden, the professional dimension becomes relevant to explore in the light of the previous and current forms of school governance and other contextual factors highlighted above. For example, it is relevant to consider the previous strong state governance and the re-centralisation reforms in recent years. From such a perspective, the two major school authorities in Sweden, the National Agency for Education and the Swedish Schools Inspectorate, are important because they reflect both formal directives but also normative expectations on Swedish principals.

Regarding the situated or socially located dimension, this dimension is linked to local conditions and a specific context, i.e. a school district, a local municipality, a specific school, or even a certain classroom. Thus, this dimension encompasses demographics, school culture, socio-economic conditions, principals’ support, and feedback. Therefore, this dimension is also connected to principals’ and teachers’ long-term identity. Understanding and adapting leadership in relation to the local context has also been identified as a key factor for a successful school leadership and school improvement (see e.g. Hallinger 2011; Hallinger and Heck 1996; Seashore 2015). Here again, the Swedish case and prevailing forms of school governance and school leadership are noteworthy from the situated or socially located dimension. For instance, it seems relevant to consider some of the arguments behind the far-reaching de-centralisation reforms in the 1990s and early 2000s proclaiming that local professionals should be given the opportunity and mandate to develop their schools based on competencies and needs at the local level. At the same time, it also becomes relevant to take into account the previous school system in Sweden, characterised by detailed regulations and directives to local school principals and teachers, as well as the re-centralisation reforms from later years. In such forms of governance, it might be less important to explore and consider the local school culture, the expectations that exist in the local school organisation, and principals’ colleagues, and so forth.

The third and final dimension, the personal dimension, is more linked to personal beliefs and values and the life ‘outside’ of schools, addressing characteristics closely related to family and social roles. This dimension, therefore, involves competing elements, for instance, family and friends’ relations. Feedback from friends and family could also become sources of tension as they might collide with the other two dimensions. The importance of this dimension is also confirmed in previous leadership research, where the importance of principals’ and other school leaderships’ personal commitment, values, authenticity, and curiosity has proven to be important for successful leadership (cf. Angelle 2017; Bezzina and Bufalino 2019; Bird et al. 2013). Linked to the Swedish case and the systemic-structural aspects described above, there are relevant characteristics related to this third identity dimension as well. Seen from an international perspective, Swedish principals are generally younger and less experienced compared to their international colleagues (OECD 2016). Another factor that seems relevant to highlight is the tradition of strong state governance in Sweden – notions that seem to remain despite the de-centralisation reforms in the 1990s and early 2000s (cf. Nordholm and Andersson 2019). In such school systems, it might be less natural for principals to embrace and integrate personal values, ideas, and considerations into their leadership – especially young and unexperienced principals.

In sum, the theoretical starting points elaborated in this article are to be found in theory on principals’ emotional identity and the professional, situated or socially located, and personal dimensions of this concept. In the analytical work, an analysis instrument formed by the three dimensions was applied to obtain a more detailed image of Swedish principals’ identities. The analytical work undertaken is detailed below.

Methods and data

Regarding the survey, email addresses to principals were obtained from the Swedish National Agency for Education. According to the official statistics (see Swedish National Agency for Education 2019), there are currently 3600 principals in Sweden in full-time employment. With regard to how the principal’s assignment is formed in the Swedish context, a number of considerations were made. In the current study, temporary appointed or hired-in principals were excluded because they have lesser possibilities to answer detailed questions on identity, especially on the situated or socially located dimension. Deputy principals were also excluded because, in contrast to many head principals, these principals generally have less time in the profession and thus experience of working as principals. Therefore, they also have limited possibilities to reflect upon what it means to be a principal and how identity is formed over a longer period of time (cf. Ritacco and Boliviar 2019). In addition, they also do not have an overall responsibility from a school leadership perspective, for instance, regarding legal matters and economic issues such as budget and salary setting – aspects that we relate to the professional identity dimension. Principals in both public and independent schools, however, were included. One main argument for integrating and merging the different forms of school ownership was that no significant differences were detected between the forms. Principals with responsibility for both preschools and schools were also included. One argument supporting this sample is the above-mentioned reform declaring that preschool leaders nowadays are entitled to be principals from the 1st of July 2019, and they also have to attend the National School Leadership Training Programme. Moreover, it is common, especially in rural areas, that principals are responsible for both preschool and compulsory school.

Taking this into consideration, the survey targeted approximately 3000 principals. After sending the initial email, it became obvious that the email addresses were not entirely updated; consequently, some work was done to find the ‘missing’ principals. New emails were sent to these principals. Thereafter, two reminders were sent out to all the principals.

Despite these challenges, the data collected meet the purpose and objectives of this article. The final sample of principals (n = 1286) represented 249 out of the 290 local municipalities. There was a nice spread between the four types of municipalities defined by the Swedish Board of Agriculture (large city, city, rural areas, and sparsely populated rural areas). Another strength of the empirical data is that there was a good spread, in terms of age, number of years in the profession, size of the school unit, gender, and a satisfying spread in terms of school ownership.

Regarding the non-response analysis, some possible factors need to be highlighted. One difficulty already highlighted was the struggles in obtaining updated e-mail addresses from the National Agency for Education. Another factor is that Swedish principals tend to change jobs more often than their European colleagues (see OECD 2016), which means that positions are vacant and that quite a lot of principals are replaced by a temporary principal. An additional factor was also that some e-mails got stuck in the municipal spam filters; therefore, all reminders were sent directly to the schools. Thirteen participants were dropped because of missing data or they were unable to meet the criterion of being head of a preschool, primary, or secondary school.

The questionnaire

This section provides details about the analysis instrument for this article. To start with, the questionnaire consisted of 42 questions. However, only questions regarding the principals’ backgrounds and emotional identity were used in this analysis. The first part of the questionnaire included questions about the following variables.

• School authorities (independent or municipal)

• Type of school (preschool, primary, or secondary)

• Type of municipality

• Gender

• Age

• Length of time as principal

Some questions in the questionnaire were constructed in accordance with the different dimensions of identity, elaborated on by Crow, Day, and Møller (2017). The professional dimension, for instance, included questions on the influence of steering documents and the Swedish Agency for Education. The situated or socially located dimension included questions on the influence of local factors such as school culture and expectations from other principals and teachers. Finally, the personal dimension included questions on, for example, personal engagement in the school, and in the local society.

The respondents were asked to state the extent to which they agreed with each statement, by indicating on a scale ranging from 1 to 6. The validation of the three dimensions (e.g. a factor analysis) is presented in Nordholm et al. (forthcoming), and these composite measures were used in this article. Listed below are the questions and how they were formulated in the questionnaire. Composite measures for the three dimensions were constructed by combining questions in each dimension to an index.

To what extent do these aspects influence you as a school principal:

The professional dimension:

Steering documents and policies

The Agency for Education

The Swedish School Inspectorates

Current research

The personal dimension:

My own personal values

My personal engagement and interest in the school

My interest in the local society

The situated or socially located dimension:

The school’s culture

Expectations from teachers

Expectations from other principals

The data in the questionnaire were analysed using the statistical package SPSS. In a first step, the descriptive statistics for all items relating to the different dimensions of identity were computed. Means for the 10 items varied between M = 3.10 and M = 5.45, showing that principals rated influence of steering documents higher than expectations from other principals. Skewness for these items varied between .91 and −1.31, showing both positive and negative skewness. Indexes for the three dimensions were computed by summing the means for each item and dividing the sum by the number of items in each dimension. Means for the indexes varied between M = 3.91 for the local dimension and M = 5.01 for the professional dimension. These computed indexes were used as dependent variables in the variance analysis reported in the results section. The Cronbach Alpha for the three dimensions varied between .78 and .82, showing a rather high internal consistency in the constructed indexes.

Results

This section presents the results of the SPSS-analysis of the possible differences between the groups of principals. The analysis was computed by using a variance analysis, with the score for the computed index for each dimension as a dependent variable and gender, length of time as principal, and type of school as independent variables. Particular attention is drawn to the results where significant variances could be discerned regarding emotional identity between the different groups of principals. These variances are presented one at a time below.

In Table 1, descriptive data are shown for the participating principals in the study. There were 954 females and 315 males in this study. Twelve principals did not answer the question about gender.

Table 1. Number of males and females.

Sex	Frequency	Percent
Female	954	74.5
Male	315	24.6
Other	12	1.0
Total	1281	100.0

The participants in this study were between 26 to over 66 years old. As illustrated in Table 2, the majority of the participants in the study were between 46 and 55 years of age. Only twelve participants were older than 66 years.

Table 2. Number of subjects in each age group.

Age group	Frequency	Percent	Cumulative Percent
26–35	39	3.0	3.0
36–45	412	32.2	35.2
46–55	558	43.6	78.8
56–65	260	20.3	99.1
66– … ..	12	.9	100.0
Total	1281	100.0

The majority of the participants had worked as a principal from one to three years. About 13 percent of the participants had worked as principals for up to one year. More than half of the sample had worked as a principal for less than three years.

In Table 3, descriptive statistics for the three dimensions are shown, separated by gender. As shown in Table 3, there were differences between males and females according to all three dimensions of identity. Females rated the professional dimension as higher compared to males (M = 5.11) and (M = 4.70), which was also a significant difference, F (1, 1266) = 92.86 p < . 000. There was also a difference between males and females regarding the situated or socially located dimension, (M = 4.05) and (M = 3.72). This difference was also significant, F (1, 1265) = 36.92 p < .000. Finally, there was a difference between males and female regarding the personal dimension M = 4.17 and M = 4.03, and this difference was also significant, F (1, 1265) = 8.26, p < .004.

Table 3. Descriptive statistics for males and females separated by the identity dimensions.

Descriptive		N	Mean	Std. deviation
Professional	Female	954	5.11	.65
	Male	314	4.70	.70
	Total	1268	5.01	.68
Situated or socially located	Female	953	4.05	.82
	Male	314	3.72	.87
	Total	1267	3.97	.85
Personal	Female	953	4.17	.78
	Male	314	4.03	.76
	Total	1267	4.14	.78

In all measures, female principals rated the dimensions as higher compared to the males. Going into detail, starting with the professional dimension, it is worth noting that both males and females estimated this dimension as higher compared to the other two identity dimensions. Still, there are significant variances between the two groups. Regarding the situated or socially located dimension, the analyses showed similar variances between female and male principals. With regard to the personal dimension, the differences between female and male principals remain. Noteworthy, however, that the mean for women increases strongly in this dimension, while the mean for men is increasing significantly compared to the situated or socially located dimension.

The analysis also targeted emotional identity, from the perspective of principals’ age. The distribution of age is shown in Table 2, and the mean ratings for the different dimensions are shown in Table 4.

Table 4. Descriptive statistics for different age groups, separated by the identity dimensions.

Identity dimensions	Age group	N	Mean	Std. deviation
Professional	26–35	38	4.70	.703
	36–45	409	4.98	.69
	46–55	555	5.06	.64
	56–65	258	5.01	.71
	66– … ..	12	4.76	.99
	Total	1272	5.01	.68
Situated or socially located	26–35	38	4.07	.68
	36–45	409	3.93	.88
	46–55	554	3.97	.80
	56–65	258	4.04	.87
	66– … ..	12	3.75	.94
	Total	1271	3.97	.84
Personal	26–35	38	4.06	.60
	36–45	409	3.99	.84
	46–55	554	4.19	.75
	56–65	258	4.25	.73
	66– … ..	12	4.60	.62
	Total	1271	4.14	.78

The analysis detected significant variances between different age groups within the personal dimension, F (4, 1267) = 6.73, p < .001. As evident, principals between the ages of 26–35 rated this dimension as rather high compared to the subsequent age group 36–45. As the table shows, it also becomes evident that the mean rises steady as the principals’ age increases. Comparable variances and relationships could not be detected within the professional and situated or socially located identity dimensions.

Another focus of the analysis was to explore identity based on the length of time in the profession (also see Table 5).

Table 5. The frequency of principals, separated by length of time as principal.

Time as principal	Frequency	Percent	Cumulative Percent
Less than 1 year	165	12.9	12.9
1–3 years	556	43.4	56.3
4–8 years	335	26.2	82.4
More than 8 years	225	17.6	100.0
Total	1281	100.0

As shown in Table 6, there are similar mean ratings between different groups of principals according to the length of time in the profession. More precisely, the analysis showed that newly appointed principals rated the personal dimension as higher compared to principals with more than 8 years as principal, F (3, 1267) = 5.36, p < .001. No other significant differences were found.

Table 6. Descriptive statistics for length of time as principal, separated by the identity dimensions.

Identity dimension		N	Mean	Std. deviation
Professional	Less than 1 year	44	5.06	.67
	1–3 year	400	5.02	.64
	4–8 year	371	4.94	.71
	More than 8 years	457	5.03	.69
Situated or socially located	Less than 1 year	44	4.13	.82
	1–3 year	399	3.98	.79
	4–8 year	371	4.00	.82
	More than 8 year	457	3.93	.90
Personal	Less than 1 year	44	4.19	.79
	1–3 year	399	4.03	.69
	4–8 year	371	4.11	.85
	More than 8 years	457	4.25	.78

The last area of focus for the analyses was to examine variances between the different types of principals (see Table 7).

Table 7. Descriptive statistics for different types of principals, separated by the identity dimensions.

Identity dimension		N	Mean	Std. deviation
Professional	Principal Compulsory school	971	4.98	.68
	Principal Preschool	301	5.11	.68
	Total	1272	5.01	.68
Situated or socially located	Principal Compulsory school	971	3.91	.83
	Principal Preschool	300	4.17	.85
	Total	1271	3.97	.84
Personal	Principal Compulsory school	971	4.14	.74
	Principal Preschool	300	4.12	.88
	Total	1271	4.14	.78

As Table 7 demonstrates, differences were detected between the principals in preschool and compulsory school for the professional dimension, F (1, 1266) = 7.91, p < .005 and situated or socially located dimension of identity, F (1, 1265) = 22.19, p < .001. The analysis showed that principals with a responsibility for preschools rated the situated or socially located dimension as higher – i.e. they experienced that this dimension affects them more in their leadership compared to school principals. They also rated the professional dimension as significantly higher than principals in compulsory schools. No significant difference could be observed between the groups for the personal dimension, p > .80.

Some remarks

It may also be worthwhile to say something brief about the analyses where significant variances were not found. For example, no significant variances could be found between the four municipality types defined by the Swedish Board of Agriculture. The ANOVA analyses also did not identify any significant variances regarding identity between independent and public principals. Finally, no variations could be identified based on school size or between principals who worked in a few or in many schools.

Discussion

The analyses showed that all three identity dimensions constitute important parts of the emotional identity of Swedish principals. It is worth highlighting in this matter that female principals rated all three dimensions as significantly higher than their male colleagues. Another general result holds that the professional dimension constitutes a greater part of the principal’s identity compared to the situated or socially located and the personal dimensions. It is also worth noting that the situated or socially located dimension had the lowest impact in terms of identity, both among female and male principals, even if this relationship is most evident among the male principals.

Looking at the details and variances between the different groups of principals, the analysis detected variances between the different age groups within the personal dimension. Principals between the ages of 26–35 rated this dimension as rather high compared to the subsequent group 36–45. The analyses also showed how the personal dimension becomes stronger as the principals’ age increases. One potential explanation for these results is that Swedish principals are generally younger and more unexperienced compared to colleagues in other countries, both as teachers and as principals, (cf. OECD 2016). Thus, with increased age and time in the profession Swedish principals become more confident in their role and the personal dimension therefore becomes more prominent in their leadership. Another potential explanation is the prevailing expectation of strong state governance and that newly appointed principals conform to (institutionalised) notions on ‘the strong state’ (cf. Nordholm and Andersson 2019). Arguably, with increased experience, these principals become more confident in their role, for example, in terms of responsibility and mandate; thus, they can also affirm the personal dimension more in their leadership. However, even if one could relate these results to some kind of expected development in the profession, there might still be a need to highlight the personal identity dimension more clearly, for instance, in principal training, so it is clearly found throughout the professional life of principals (cf. Angelle 2017; Bezzina and Bufalino 2019; Bird et al. 2013).

Regarding the situated or socially located dimension, significant variances were detected between the different types of principals, namely between principals with a responsibility for preschool and compulsory school. More specifically, preschool principals rated the situated or socially located dimension as higher compared to school principals. Similar variances could not be detected between the groups within the professional and personal dimensions. These results may be linked to the preschool’s proximity to the local community and that it is often the first (and sometimes the only) authority that parents and children meet in the local community, not the least in rural regions. Thus, there is also a very close relationship between the surrounding community, to colleagues, and the principal’s leadership. Another possible explanation could be found in the prevailing forms of school governance in Sweden and to the systemic-structural aspects discussed above. More precisely, preschool principals are generally less exposed to external inspection by the Swedish Schools Inspectorate compared to compulsory school principals (cf. Novak 2018). This probably also increases the opportunities to consider and integrate the local context in their leadership.

The fact that the Swedish principals carefully consider what the government and its authorities do and recommend is arguably favourable, but at the same time, there could be a risk if principals continuously expect that answers to complex leadership questions are to be sought ‘upwards’ in the system. Seen from such a perspective, there are also important differences to note compared to a strong professional identity underpinning school improvement and educational change (cf. Ritacco and Bolviar 2018, 2019). Combined with the result that the situated or socially located dimension is estimated as the lowest, most clearly among male principals, there is reason to ponder how Swedish principals understand and adapt their leadership to the local context. Understanding the culture of the school, how the surrounding community affects the local leadership, how colleagues’ expectations enable and limit leadership, and so forth are, as noted, important elements for school leadership and successful school improvement (Hallinger 2011; Hallinger and Heck 1996; Seashore 2015). Building such an understanding, therefore, is important, not the least in the National Principal Training Programme, which is mandatory for all principals in schools and preschools in Sweden.

Another result worth considering is that female principals rated the three dimensions as significantly higher than their male colleagues. That is not to say that male principals overlook important parts of their leadership because they also rated all dimensions as somewhat high, but the differences between men and women are still vibrant. Here, there is however an obvious need for further studies. In the light of Corin and Björk’s study (2016, 2017), these results could however be an expression that there are different expectations on male and female principals in Sweden and therefore, there is a need to pay attention to these issues further, both in future research but also in other contexts where issues related to working conditions and equality issues are addressed.

From a broader perspective, Crawford (2007) emphasised that emotions are intrinsic to the practice of principals, and therefore all organising actions of principals are inseparable from, and influenced by, emotions. That declared, the results of the article show that the three dimensions in the study by Crow, Day, and Møller (2017) contribute to a broader understanding of principals’ identity and how identity influences principal’s work. They also encouraged further research within the field, and we believe that the current article adds an important piece of research to this end. Using Sweden as a case, the analysis instrument thrives in identifying the overall results and significant variances between different groups of principals.

Conclusion

This article explores Swedish principals’ emotional identity by applying an analysis instrument built on three dimensions elaborated on in the study by Crow, Day, and Møller (2017), namely, the professional, the situated or socially located, and the personal dimension. Aside from the intention to broaden the understanding of Swedish principal’s identity, particular attention is paid to variances, in terms of gender, age, length of time in the profession, and to the type of principal. Empirically, the article builds on data obtained from a survey of Swedish principals, completed in January 2019.

The findings show that all three dimensions constitute important parts of principals’ emotional identity. In addition, female principals rated all three dimensions as significantly higher than their male colleagues. Another general result is that the professional dimension constitutes a greater part of the principal’s identity compared to the situated or socially located and the personal dimensions. It is also worth noting that the situated or socially located dimension had the lowest impact, in terms of identity, particularly among male principals. Regarding other variances between the different groups of principals, the analysis disclosed that the personal dimension expands with increasing age. Considering the situated or socially located dimension, preschool principals rated this dimension as higher compared to compulsory school principals.

Apart from these empirical findings, this article also offers a theoretical contribution because the analysis instrument developed and applied succeeds in capturing the three dimensions elaborated on in the study by Crow, Day, and Møller (2017). Therefore, this article bridges data to theory in a way that, to our knowledge, has not been done previously even if important work has been conducted with related perspectives and intentions (cf. Ritacco and Bolviar 2018; 2019).

Regarding limitations, it is important to emphasise that the article builds on data obtained in the Swedish school context. The construction of principals’ identity is situated in time, space, and place, and this should constitute a point of departure for studying and understanding the concept. As detailed, there are some characteristics regarding school system governance in Sweden with relevance for principals’ identity to consider, e.g. a high degree of de-centralisation, NPM, and market adaptation, even if similar movements are seen internationally (cf. Allen 2010; Rönnberg 2015). Consequently, the article encourages further studies in different contexts and in doing so, there could be a need to modify the analysis instrument and to interpret the results based on the specific school context.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The term independent schools in Sweden comprises a range of school organisers, i.e., from large school companies to small cooperatives. The schools are financed through a state voucher system and do not charge any fees.