Development and validation of a research instrument for measuring the presence of democratic school leadership characteristics

Abstract

The main purpose of this article is to describe the process of development and validation of a research instrument for measuring the presence of democratic school leadership characteristics. The instrument has been developed and validated in the city of Zagreb and Zagreb County in Croatia in four phases: theoretical construction; expert validation (N = 6 experts); pilot research (N = 77 teachers); and instrument structure validation through the main research (N = 651 teachers). The expert validation and pilot research took place in 2018. The main research was conducted in January and February 2019. According to Carmines and Zeller (1979), the results of the factor analysis, which show that the instrument is measuring a single phenomenon, indicate that all four hypotheses have been fulfilled. More specifically, several aspects of the extracted (i.e. unrotated) factor matrix support the following four hypotheses: the (unrotated) factor matrix of all 32 items explains a total of 66.793% of the variance; subsequent components demonstrate the fairly equal proportions of the remaining variance except for a gradual decrease; all items have high factor loadings on the first component (from .603 to .861); and all items have higher loadings on the first than on the second component.

Keywords

Democratic school leadership instrument validity reliability principal

Introduction

There are many definitions of school leadership. Some authors in this field offer their definitions (Gardner, 2007; Kouzes and Posner, 2007; Pont et al., 2008; Stein, 2015), whereas others try to combine definitions from different authors with the intention of trying to specify the main elements of school leadership (Bass and Bass, 2008; Bush, 2003; Bush, 2019; Bush and Middlewood, 2013). Pont et al. (2008) refer to school leadership as a process of intentional influence on others, based on articulated objectives and results that need to be achieved. Very similar to this definition, Gardner (2007) sees school leadership as a process during which an individual and/or a team of people ensure that others, who are their followers, accomplish appointed goals. Kouzes and Posner (2007) describe school leadership as a relationship between those who want to lead and those who have decided to follow them, whereas Stein (2015) says that school leadership is planning, directing, controlling and organizing, as well as influencing people and inspiring them to succeed. Bush and Middlewood (2013) claim that there are three main dimensions of school leadership. The first is the influence of one or many leaders on individuals or a group of people to organize activities and relationships within a group or an organization. The second dimension is based on personal and professional values, and the third is the developed and articulated vision an individual has about an organization. It could be concluded that there are several main elements of school leadership (Pažur and Kovač, 2019): the influence of one or many individuals on their followers; a process of coordinating followers and leading them towards goal accomplishment; and cooperation and participation during the latter process, throughout which the actors involved share responsibility.

Because of the significant connection between school leadership and school effectiveness, different authors have started to discuss and research the relationship between these two constructs (Barnes and Spangenburg, 2018; Bush and Middlewood, 2013; Fullan, 2007; Pont et al., 2008; Stylianou and Savva, 2016). A few of them have made the distinction between different types of school leadership to ascertain which types are more or less applicable to a specific school context (Bass and Bass, 2008; Bush, 2019; Bush and Glover, 2003; Bush and Middlewood, 2013; Collins, 2007; Pont et al., 2008). With regard to these typologies, some authors (Bass and Bass, 2008; Bush and Glover, 2003; Goleman, Boyatzis and McKee, 2002) have recognized democratic school leadership (DSL) as one of many school leadership types. Bass and Bass (2008) claim that DSL, unlike autocratic leadership, is employee-orientated, considerate, supporting, open, informal and warm. Furthermore, this leadership style is orientated towards a consensus (Goleman, Boyatzis and McKee, 2002), with the responsibility for the leading process being shared between more equal participants (Bush and Glover, 2003).

The concept of DSL cannot be understood fully without first explaining the connection between democracy and education. The first theoretical and practical connections between these two concepts started with Dewey (1916), who argued that the school should be a hub of social activity and the key to community engagement. Similarly, Dundar (2013) claims that the concepts of democracy and education are connected because democracy can be ensured through education, and vice versa. Currently, some of the most influential global and European organizations¹ demand that their member countries develop an educational policy that will respond to the needs of democracy (OECD, 2020; UNESCO, 2016). By answering to these demands, different educational policies can adopt one of three approaches of conceptualization (Sant, 2019): education for democracy; education within a democracy; or education through democracy. The concept of DSL is aligned with the approach of education through democracy, which means involving the members of the community (parents, professional educators, students) in the decision-making process. The findings of the research conducted by McCormick (2017) suggest that the school leader is a person who is pulled in many different directions, but also a person who should remain committed to the democratic process of listening and deliberation to make education through democracy possible. According to Dewey’s philosophical tradition of democratic schools, democratic leadership implies that the school principal is responsible for building an educational organization around central democratic values such as supporting equity and social justice, as well establishing them in the wider community. At the same time, he/she is responsible for emphasizing social justice, dignity and the rights and welfare of all individuals in school, including minorities (Kilicoglu, 2018). Unlike some other leadership types (distributed school leadership, transformational school leadership, instructional school leadership),² no scale for measuring the level of development of DSL has yet been developed. Therefore, the main purpose of this article is to describe the process of development and validation of a research instrument for measuring the presence of DSL characteristics.

Elements of democratic school leadership

Based on a detailed literature review, Pažur and Kovač (2019) define DSL as school leadership in which the responsibility for leading and developing an organization is shared between all interested actors and in which the focus of those leaders is on practising democracy with regard to everyday activities in all spheres of school life. They point out three main elements of DSL: the circular nature of leadership; practising democracy; and facilitating the development of active citizens. This definition will be used for a theoretical construction of the instrument explained below.

Authors who talk about DSL (Begley and Zaretsky, 2004; Brooks and Kensler, 2016; Goleman, Boyatzis and McKee, 2002; Hope, 2012; Kensler and Woods, 2012; Mncube et al., 2015; Møller, 2006; Moos and Huber, 2007; Riley, 2003; Williams, 1989; Woods, 2005; Woods, 2007; Woods and Gronn, 2009) agree that the main characteristic of this leadership type is shared responsibility during the leading process. However, because this characteristic is also visible in some other leadership types (Bush, 2019), it is important to examine the DSL model more closely. Essentially, DSL focuses on the cultivation of an environment that supports participation, sharing of ideas, honesty, openness, flexibility and compassion. Therefore, school leaders should employ a participatory approach and ethics of social justice through structures that enable the views of students and teachers to be included in the decision-making process (Jwan and Kisaka, 2017). According to Woods (2005), DSL is a leadership type in which the impulse for organizational development comes from a group that is more than just the aggregate of individuals. The main characteristic of the DSL leadership model is the circulation of initiative. The circular nature of leadership means there is a circular exchange of initiative between different interested actors in which a leader transfers the initiative to colleagues who then react and answer to the existing initiative or start an initiative of their own (Woods, 2005). In addition, democratic leadership is associated significantly with a coherent leadership team and distributed leadership functions, such as the distribution of support and supervision (Kilicoglu, 2018). In other words, the main goal of DSL is the development of a school culture that is based on democratic ideals and practices (Møller, 2006; Riley, 2003; Woods 2005; Woods, 2007). In DSL, the leader should be focused on creating and practising democracy or, in other words, the leader enables all school actors to fulfil their right to freedom of expression, to personal and professional development and to be part of the decision-making process (Møller, 2006; Moos and Huber, 2007; Woods, 2005). This type of school leadership enables students and grown-ups to gain knowledge and develop skills to become active citizens of the communities they are part of (Kensler and Woods, 2012; Moos and Huber, 2007; Riley, 2003).

To implement DSL in the school, the school leader should embody and nurture specific activities, behaviours and values. To promote the circular nature of leadership, the school leader should open a creative space for the development of autonomy and initiative by potential co-leaders (Woods, 2005). Furthermore, by promoting open access to knowledge (Woods, 2005) and by noticing and complimenting the achievements of teachers (Parham, 1944), the school leader contributes to their self-confidence, which is an important precondition for starting and/or taking over leading initiatives. It is very important for a school leader who is developing DSL to display democratic values (Woods, 2005) in the way he/she behaves, interacts with students, teachers and parents and in the way he/she deals with conflict and ethical dilemmas (Woods, 2007). Democracy can be developed only when practised in everyday situations (Moos, 2011; Riley, 2003; Woods, 2004) and when all interested school actors have an opportunity to be a part of the decision-making process (Parham, 1944; Riley, 2003; Woods, 2005). It is important to highlight that this kind of participation needs to be purposeful and must lead towards concrete actions (Woods, 2005). The school leader who creates and practises democracy on an everyday level should create guidelines and structures that enable school actors to participate in the decision-making process (Woods, 2005), assign school tasks to multiple teachers (Parham, 1944) and promote the meaningful work of different school bodies (such as parents’ and/or students’ councils). Practising democracy can also be done by engaging in and nurturing dialogue (Parham, 1944; Woods, 2005) and by using nonviolent communication (Begley and Zaretsky, 2004; Riley, 2003), both of which will lead towards developing a culture in which everybody can share their opinion freely and has an opportunity to develop critical thinking skills (Hyde and Laprad, 2015). The final element of DSL is making a contribution towards creating active and responsible citizens. An important part of this particular element is that a school leader should develop and nurture core democratic values such as mindfulness, consideration, patience and tolerance (Parham, 1944), accept differences and encourage plurality (Møller, 2006), as well as promote equality and inclusivity (Woods, 2005). By applying all the above-mentioned behaviours and developing recognized values, the school leader can direct a school towards the goal of his/her establishment becoming a democratic community (Dorczak, 2014).

According to everything that has already been said, it could be concluded that the role of the school leader in developing DSL should be to act simultaneously and continuously in three intertwined dimensions. More specifically, the school leader should empower other actors so that they can participate purposefully in democracy, create conditions for meaningful democracy and be ‘a model of democracy’ in everyday activities. Based on the literature review, which has led to a theoretical construction of DSL (Table 1), a first version of the instrument for measuring DSL characteristics, which has 33 items, has been developed.

Table 1.

Theoretical construction of the instrument for measuring DSL characteristics.

The circular nature of leadership (power and responsibilities, self-confidence, democratic values)	Practising democracy (school structures, dialogue)	Encouraging the development of active citizens (everyday situations)
The school leader encourages the autonomy and initiative of other school actors (Woods, 2005).	The school leader promotes practical guidelines for the participation of school actors in the decision-making process (Riley, 2003).	The school leader accepts different initiatives proposed by different school actors (Woods, 2005).
The school leader shares authority and responsibility with other key school actors (Gülbahar, 2016).	The school leader promotes teacher involvement in the process of leading working groups and/or projects (Parham, 1944).	The school leader assigns project tasks according to the knowledge and skills of the teachers (Woods, 2005).
The school leader sees the decision-making process as an opportunity for shared learning (Rusch, 1995).	The school leader promotes the effective operation of different institutional structures that enable school actors to participate in the decision-making process (Hope, 2012).	The school leader assigns school tasks to a greater number of school actors in accordance with their expertise and not in relation to the existing hierarchical structures (Møller, 2006).
The school leader creates opportunities for all school actors in relation to critical thinking about learning (Møller, 2006).	The school leader promotes students’ involvement in planning class activities and in defining criteria for evaluating their schoolwork (Møller, 2006).	The school leader nurtures relationships based on mutual trust and respect (Møller, 2006).
The school leader notices and compliments the achievements of teachers (Parham, 1944).	The school leader usually leaves the door of his/her office open (Møller, 2006).	The school leader accepts positive school changes proposed by the students’ council (Hope, 2012).
The school leader helps teachers to find their ‘passion’ and what they are good at (Woods, 2005).	The school leader provides the means whereby and promotes situations in which all interested school actors can fulfil their right to freedom of expression (Woods, 2007).	The school leader encourages the parents’ council to be an active actor in the decision-making process (Parham, 1944).
The school leader acts towards others equally and appropriately (Woods, 2005).	The school leader nurtures and promotes the flow of free information and open dialogue (Rusch, 1995).	The school leader encourages weekly/monthly meetings of all school actors at which important school decisions are made (Hope, 2012).
The school leader is truthful, inclusive and transparent (Delgado, 2014).	The school leader nurtures and promotes teamwork (Rusch, 1995).	The school leader promotes decision-making via a consensus or a majority vote of all school members (Hope, 2012).
The school leader advocates dialogue between all school actors involved in a conflict as the means by which it can be dealt with most effectively (Woods, 2007).	The school leader is focused on becoming acquainted with and replying to the interests and needs of students (Normore and Jean-Marie, 2008).	During decision-making, the school leader rates a student’s voice as equal to that of a grown-up (Hope, 2012).

Methodology

After the theoretical construction of the instrument, further development and validation was performed in three phases: expert validation (N = 6 experts); pilot study (N = 77 teachers); and instrument structure validation through the main research (N = 651 teachers). The instrument, which was written in the Croatian language, was given to the teachers who were asked to what degree (on a 5-point Likert-type scale) a certain action existed according to the practice in their particular school.

Before the expert validation, and after the theoretical construction of the instrument, two nationally recognized academics/researchers in the field of school leadership were consulted with regard to both single items and the instrument as a whole.³ The process of expert validation involved six experts⁴ from three elementary schools from the city of Zagreb and Zagreb County in Croatia and took place from July to August 2018. The experts had the following professional profiles: a principal (30 years of professional experience in school), an English teacher and school counsellor (9 years of professional experience in school), two history and geography teachers (16 years of professional experience in school), a school counsellor (7 years of professional experience in school) and a primary school teacher (8 years of professional experience in school). All of the participants had been active in the implementation of civic education,⁵ so they were familiar with the topic of democracy in schools. The experts completed the questionnaire, and this was followed by an interview, in which they provided further explanation of their answers.

The pilot study was conducted in three elementary schools (one urban school from the city of Zagreb, one urban school from Zagreb County and one suburban school from the larger area of the city of Zagreb) in November 2018. In the three schools, all teachers present at the time completed the questionnaire. The final sample consisted of 77 teachers (15.6% male and 84.4% female). With regard to their job role, there were 29.9% teachers working in the first to fourth grade, 50.6% working in the fifth to eighth grade, 11.8% professional experts⁶ and 7.7% working in other positions (mostly in after-school care).

In the main research, 651 teachers from 20 elementary schools⁷ from the city of Zagreb and Zagreb County⁸ completed the questionnaire.⁹ The main research took place from 20 January to 25 February 2019 and involved 12.7% male and 85.6% female teachers. The sample included teachers from an urban (74.3%) and a suburban (25.7%) area. With regard to the number of students per school, teachers from schools with up to 300 students (12.9%), schools with 301 to 600 students (41.5%) and schools with more than 600 students (45.6%) were included. In relation to the type of leadership being studied, it was important to include teachers from schools that had principals with different lengths of professional tenure. Therefore, there were teachers from schools where the current principal had held the position for between 1 to 5 years (47.3%), schools where the current principal had held the position for between 6 to 9 years (35.5%) and schools where the principal had been in position for 10 or more years (17.2%). It was also important to include teachers with different lengths of professional experience¹⁰ (31.2% teachers had between 1 to 9 years of professional experience, 30% had between 10 and 19 years of professional experience, 23.7% teachers had between 20 and 29 years of professional experience and 12.3% had 30 or more years of professional experience). The sample consisted of teachers working with grades 1 to 4 (43.4%), those who worked with grades 5 to 8 (51.4%) and professional experts (6.7%). During the expert validation and pilot research, the first draft of the instrument was further developed.

Results

Expert validation

Experts considered questions from the questionnaire with the aim of assessing whether each item was understandable and if it was applicable to the school context (Figure 1). In addition, they assessed whether each item was relevant in relation to democratic processes in school. They were also asked to comment on whether they thought any of the items were unnecessary or if some important element of leadership had been excluded. This process was followed by interviews to obtain a better understanding of their comments.

Figure 1.

Expert validation: Example of questions.

After the expert validation, 16 items were assessed as completely understandable and relevant for the school context and 16 items were further developed. One of the proposed items was excluded (its content was already part of another item), and four new items were developed (experts suggested that these important school leadership elements had not been included). Examples of some experts’ comments are presented in Table 2.

Table 2.

Examples of experts’ comments.

	Unclear item	Item that is not appropriate for the Croatian context
Initial item	The school leader includes all school members in the decision-making process.	The school leader organizes institutional structures that facilitate the decision-making process (e.g. co-management, leadership team, student projects).
Experts’ comments	It is unclear who school members are. (E4-P) It is important to highlight that the school leader includes in the decision-making process those members who are competent to be part of it. (E5-U)	Co-management is not recognized in Croatian law. (E4-P) This is not applicable to the school context. (E5-U)
Final item	The school leader includes all school members (experts, teachers, students, parents, administrative staff) in any decision-making process that is relevant for them.	The school leader creates teams that take responsibility for some school processes (e.g. team for self-evaluation, team for civic education, project teams).

After the comments of the expert validation had been incorporated into the instrument it contained 35 items for measuring DSL characteristics. It was then used in a pilot study.

The pilot study

A pilot study is regarded as a significant approach in ensuring the reliability and validity of an instrument, as well as its adequacy (Baker, 1994). In other words, some items will be dropped to attest the reliability and validity of the instrument (Johnson and Christensen, 2008). The teachers from the pilot study were asked to fill in the questionnaire according to what degree (on a 5-point Likert-type scale) a certain action existed according to the practice in their particular school.

The results of descriptive statistics (M, SD) indicated that all of the items had a gently positive skewness, according to which none of them should be excluded from the instrument. An exploratory factor analysis was conducted to identify the structure of the instrument and to verify the contribution of each item to the factor. A factor analysis with oblimin rotation was performed on the data collected and five factors were extracted. Three items were cross-loaded and/or not loaded anywhere and were, therefore, omitted (Table 3).

Table 3.

Omitted items.

Items	1	2	3	4	5
A principal is willing to take responsibility for a decision a teacher makes according to his/her own initiative.	,250	,286	,155	,767	,256
A principal enables teachers to take the initiative in the decision-making process (new ideas, innovations and changes).	,195	,239		,234	,783
A principal enables students to take the initiative in the decision-making process (new ideas, innovations and changes).		,106	,575	,118	,619

In addition, 32 out of the 35 initial items contained in the instrument were extracted through factor analysis. This confirmed the presence of three factors with eigenvalues exceeding 1, explaining a total of 73.228% of the variance (Table 4). The factors extracted have a high range of inter-factor correlation (Table 5), which indicates that the instrument could be measuring a one-dimensional phenomenon of DSL, or it could be measuring three connected sub-dimensions of DSL.

Table 4.

Variance explained by the pilot study.

Component	Initial eigenvalues			Extraction sums of squared loadings
Component	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
1	20,738	64,807	64,807	20,738	64,807	64,807
2	1580	4939	69,746	1580	4939	69,746
3	1114	3482	73,228	1114	3482	73,228
4	,943	2948	76,175

Table 5.

Component correlation in the pilot study.

Component	1	2	3
1	1000	-,503	,741
2	-,503	1000	-,364
3	,741	-,364	1000

A pilot field test was also conducted to assess internal consistency reliability (Cronbach’s alpha). Low values of alpha would indicate that the items in question were capturing the construct poorly (Churchill, 1995). Because all 32 items contained in the instrument complied with the statistical requirements, the Cronbach’s alpha was .981. Additionally, each loaded factor was assessed by the Cronbach’s alpha measure (factor 1 = .972; factor 2 = .940; factor 3 = .920). The results implied that the overall reliability of the whole instrument for measuring DSL characteristics was very high and that all 32 items should be retained.

The main research

The teachers from the main study were asked to fill in the questionnaire according to what degree (on a 5-point Likert-type scale) a certain action existed according to the practice in their particular school.

A confirmatory factor analysis (Table 6) was applied to the data gathered to test the construct of the instrument. The factor analysis of all 32 items confirmed the presence of two factors with eigenvalues exceeding 1, explaining a total of 66.793% of the variance (Table 7). All the items have factor loadings from .603 to .861.

Table 6.

Principal component analysis – component matrix.

Component matrix ^a
	Component
	1	2
The principal¹¹ creates an atmosphere of mutual respect and cooperation.	,861	-,240
When making decisions, the principal considers relevant facts in conjunction with all actors included in the process.	,854
The principal ensures the availability and flow of all relevant information within the school.	,853	-,169
The principal nurtures teamwork.	,849	-,187
The principal communicates with decision-makers and he/she advocates for starting positive changes in the school.	,848	-,179
The principal enables everyone, not just particular school members, to fulfil their right to freedom of expression.	,845	-,244
The principal accepts initiatives (ideas, suggestions, innovations) recommended by teachers.	,844	-,137
The principal includes all school members (experts, teachers, students, parents, administrative staff) in decision-making processes that are relevant for them.	,836
The principal presents and clearly explains current school situations.	,835	-,184
The principal creates an atmosphere in which others share their thoughts and opinions without fear of the consequences.	,822	-,232
The principal ensures that all the important school decisions are made by a consensus.	,819
The principal helps teachers to find their ‘professional passion’ and practise what they are good at.	,816	-,147
During the decision-making process, the principal recognizes students’ needs.	,813
The principal acts towards everyone equally.	,810	-,268
All school members feel free to talk to the principal whenever they need his/her help.	,799	-,306
In the case of conflict, the principal enables dialogue with all those involved.	,794	-,156
The principal is willing to take responsibility for a decision made according to a student’s own initiative.	,792
The principal discusses with teachers how useful their new knowledge and skills gained through professional training are.	,786
The principal tends to include all school members (teachers, students, professional experts, and administrative and technical staff) in creating proposals that the school board will then vote for.	,783	,260
The principal accepts suggestions for positive school changes made by the students’ council.	,783	,341
When sharing out school tasks, the principal offers everyone an opportunity to get involved.	,775
The principal highlights teachers’ achievements.	,752
The principal takes into consideration the knowledge and skills of school members, and allocates important school tasks (e.g. leading projects, leading out-of-school activities) according to members’ strengths.	,749
After tasks have been completed successfully, the principal recognizes and rewards those individuals who have been the most responsible for the outcome.	,748
The principal puts discussions about students’ achievements on the agenda of relevant school meetings.	,738
The principal encourages students to give opinions and feedback with regard to the curriculum.	,738	,330
If necessary, the principal holds (unplanned) meetings with all school members (including students), at which school important issues are discussed.	,733
If possible, when decisions are made, the principal values a student’s voice equally to that of a grown-up.	,733	,443
The principal includes students, both directly and indirectly, in the process of evaluating their schoolwork.	,725	,412
The principal respects the opinion of the parents’ council about proposals for the school board.	,716	,391
The principal creates teams that take responsibility for some school processes (e.g. team for self-evaluation, team for civic education, project teams).	,649
The principal accepts initiatives (ideas, suggestions, innovations) recommended by parents.	,603	,431

Extraction method: principal component analysis.

^a Two components extracted.

Table 7.

Variance in principal component analysis explained.

Total variance explained
Component	Initial eigenvalues			Extraction sums of squared loadings			Rotation sums of squared loadings^a
Component	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %	Total
1	19,807	61,898	61,898	19,807	61,898	61,898	19,052
2	1567	4895	66,793	1567	4895	66,793	14,770

Extraction method: principal component analysis.

^a When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

The factors extracted have a high range of inter-factor correlation (Table 8), which indicates that the instrument could be measuring a one-dimensional phenomenon of DSL.

Table 8.

Component correlation matrix.

Component	1	2
1	1000	,726
2	,726	1000

Discussion

An instrument for measuring DSL characteristics

Even though the factor analysis extended two factors, the data collected indicate that the instrument measures one general factor that explains the single phenomenon of DSL. This is confirmed by four verified hypotheses, which according to Carmines and Zeller (1979), indicate that the instrument has a one-dimensional object of measurement.

More specifically, several aspects of the extracted (i.e. unrotated) factor matrix should support these four hypotheses:

The first extracted component should explain a large proportion of the variance in the item (< 40%). In this case, the factor analysis of all 32 items confirms a total of 66.793% of the variance (Table 7).

Subsequent components should explain fairly equal proportions of the remaining variance except for a gradual decrease. This is the case in the present research and it is visible in Table 7.

All or most of the items should have substantial loadings on the first component (< 0.3). In this case, all items have factor loadings on the first component from .603 to .861 (Table 6).

All or most of the items should have higher loadings on the first component than on subsequent components. In the factor matrix presented, all of the items have higher loadings on the first than on the second component (Table 6).

Consequently, it could be concluded that the instrument developed measures one single phenomenon of DSL characteristics.

Reliability

Reliability can be assessed in various ways, and these are based on the extent to which items belonging to the same scale are associated. It follows from the Cronbach’s alpha coefficient that the reliability coefficient increases with the number of items as well as the item–item correlation (Bentzen and Christiansen, 2009). As a general rule, it is recommended that alpha should be no less than 0.80. In the present study, alpha has a high level of 0.98. Ferketich (1991) recommended that corrected item–total correlations should range between .30 and .70 for a good scale. In this case, corrected item–total correlations were from .583 to .845 for each item, with an average of .769. Results that present the value Cronbach’s alpha would have if a particular item was deleted from the scale indicated that the removal of any item would result in a lower Cronbach’s alpha. Finally, it could be concluded that according to the data presented, the instrument has a high level of internal consistency reliability.

Validity

When creating an instrument, it is important to ensure it is valid, or that it measures what it is supposed to measure rather than reflecting some other phenomenon. Even so, validity is a matter of degree, not an all-or-none property (Carmines and Zeller, 1979). There are several different types of validity, and each of them takes a different approach in assessing the extent to which a measure measures what it purports to. Carmines and Zeller (1979) point out the three most basic types of validity: content validity; criterion-related validity; and construct validity.

Content validity depends on the extent to which an empirical measurement reflects a specific domain of content (Carmines and Zeller, 1979). In the case of creating an instrument for measuring DSL characteristics, content validity is secured by its theoretical construction or, more specifically, by highlighting the main theoretical elements that should be included in the instrument items. Furthermore, for content validity purposes, different experts (researchers, principal of the elementary school, other school employees) have commented on the content of the instrument, mainly through an expert instrument validation. As far as criterion-related validity is concerned, the results were related to another external criterion. During the main research, participants answered the following yes/no question: ‘Do you consider your leadership as democratic?’ As many as 86.5% of them answered ‘yes’ (N = 454; M = 4.35; SD = 0.53), only 13.5% answering ‘no’ (N = 67; M = 2.96; SD = 0.72). It is statistically significant that those participants who answered ‘no’ estimated the number of characteristics of DSL to be lower than those who answered ‘yes’ (t(519) = 19, 117, p < 0.001). In construct validity, agreement is sought on the ‘operationalized’ forms of the construct, clarifying what we mean when we use this construct (Cohen et al., 2007). In this case, construct validity was yielded by the pilot study, with which all necessary adjustments of the instrument were made.

Another two important types of validity are internal and external validity (Cohen et al., 2007). According to the same authors, internal validity seeks to demonstrate that the explanation of a particular event, issue or set of data that a piece of research provides can actually be sustained by the data. External validity refers to the degree to which the results can be generalized to the wider population or other cases or situations. To ensure internal validity, DSL operationalization was based on a thorough search and examination of the literature, the research limitations were considered and appropriate statistical methods were used. As a measure of insurance of external validity, the representative sample was a realistic distribution of teachers’ characteristics in two specific regions, meaning that schools of different sizes (small, medium, large) and different locations (suburban, urban) were included in the research.

Future research and policy implications

DSL as a construct suggests shared leadership and shared responsibility for the leadership process in schools (Begley and Zaretsky, 2004; Brooks and Kensler, 2016; Goleman, Boyatzis and McKee, 2002; Hope, 2012; Kensler and Woods, 2012; Mncube et al., 2015; Møller, 2006; Moos and Huber, 2007; Riley, 2003; Williams, 1989; Woods, 2005; Woods, 2007; Woods and Gronn, 2009), and these two elements were highlighted during the theoretical construction of the instrument. As indicated during the process of expert validation, school leadership is closely connected to one person, the principal. According to the relevant legislation, the principal is the only person legally responsible for all decisions made in a school. Different types of leadership (such as having a leadership team or co-principalship) are not applicable to the Croatian context. Therefore, the concept of DSL in the instrument was adjusted to fit the Croatian context, to which it was applied.

Bearing this in mind, the instrument has many ramifications for further research and policy implications at different levels of educational policy and practice. It could be used by the academic community for further research on DSL and its relationship with other relevant school elements. Furthermore, because many research results show that the development of DSL could have multiple positive effects on teachers (Johnson, 2017; Kars and Inandi, 2018; Peker et al., 2018; Shepherd-Jones and Salisbury-Glennon, 2018), students (Mager and Nowak, 2012; Quinn and Owen, 2016) and the school as a whole (Delgado, 2014; Dorczak, 2014; Gülbeher, 2016), the instrument could be used by school leaders for self-evaluation and as a basis for further school leadership development. Because there is a lack of autonomy in Croatian schools with regard to shared leadership, the results of the research in which the instrument was used could be an evidence-based starting point for decision-makers to effect changes in relation to Croatian educational policy and legislation. It is necessary to facilitate the further development of DSL, especially in the sense that in relation to making schools more equitable, school autonomy is the future (Keddie, 2015).

Conclusion

The process of validation of the instrument implies that it measures the general phenomenon of DSL characteristics. The instrument has adequate metric characteristics and shows a high degree of reliability and validity. As a consequence, it can be recommended for use in general school practice, as well as in future research. Even so, it is important to highlight the limitations of the study. The application of the instrument to the Croatian context only, as well as not validating the instrument by employing internationally renowned academics, opens a discussion as to whether it will show adequate metric characteristics when it is used in other countries. In addition, it is necessary to add that some important aspects of DSL are not highlighted enough because they are not legal obligations of principals in Croatia explicitly, for example, elements of school leadership in the field of human rights and fundamental freedoms. In the future development of the instrument, it will be necessary to explore how to incorporate these issues. Finally, the concept of DSL is extensive, and it could be explained holistically only by including all-important school actors when measuring it and by using a mixed methods approach in which certain thoughts, values and attitudes could be better understood by qualitative research methods.

The development and the validation of the instrument for measuring DSL characteristics will contribute to a better understanding of school leadership as well as of the processes of school democratization. The instrument will also provide direction for practitioners in developing a democratic, more participative and inclusive school environment. DSL is a construct that should be further explored, and the instrument that has been developed has been successful in measuring the characteristics of this type of school leadership.

Footnotes

Declaration of conflicting interests

The author has no conflicts of interest to declare.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Monika Pažur

Notes

References

Baker

(1994) Doing Social Research. 2nd ed. New York: McGraw-Hill.

Barnes

Spangenburg

(2018) When leadership fails – a view from the lens of four employees. American Journal of Business Education 11(3): 49–54.

Bass

(2008) The Bass Handbook of Leadership: Theory, Research & Managerial Applications. New York: Free Press.

Begley

Zaretsky

(2004) Democratic school leadership in Canada’s public school systems: Professional value and social ethic. Journal of Educational Administration 42(6): 640–655.

Bentzen

Christiansen

(2009) Current health as a general health indicator II: Evaluation of reliability and validity. Scandinavian Journal of Primary Heath Care 13(1): 21–25.

Brooks

Kensler

LAW

(2016) Distributed leadership and democratic community. In: English

(ed.) The SAGE Handbook of Educational Leadership. 2nd ed. Thousand Oaks: SAGE Publications, pp. 55–67.

Bush

(2003) Theories of Educational Leadership & Management. London: SAGE Publications.

Bush

(2019) Models of educational leadership. In: Bush

Bell

Middlewood

(eds) Principles of Educational Leadership and Management. 3rd ed. London: SAGE Publications, pp. 3–19.

Bush

Glover

(2003) School Leadership: Concepts and Evidence Available at: http://dera.ioe.ac.uk/4904/1/download%3Fid=17370&filename=school-leadership-concepts-evidence-summary.pdf (accessed 10 February 2018).

10.

Bush

Middlewood

(2013) Leading and Managing People in Education. London: SAGE Publications.

11.

Carmines

Zeller

(1979) Reliability and Validity Assessment. Beverly Hills: SAGE Publications.

12.

Churchill

(1995) Marketing Research: Methodological Foundations. New York: The Dryden Press.

13.

Cohen

Manion

Morrison

(2007) Research Methods in Education. Abingdon: Routledge.

14.

Collins

(2007) Level 5 leadership. In: Fullan

(ed.) Educational Leadership. San Francisco: Jossey-Bass, pp. 27–51.

15.

Delgado

(2014) Democratic leadership in middle schools of Chihuahua Mexico: Improving middle schools through democracy. Journal of International Education and Leadership 4(1): 1–11.

16.

Dewey

(1916) Democracy and Education. New York: Barnes and Noble.

17.

Dorczak

(2014) Leadership for democracy: Developing democratic organizations. Contemporary Educational Leadership 1(3): 19–25.

18.

Dundar

(2013) Students’ participation to the decision-making process as a tool for democratic school. Educational Sciences: Theory & Practice 13(2): 867–875.

19.

Ferketich

(1991) Focus on psychometrics: Aspects of item analysis. Research in Nursing & Health 14(2): 165–168.

20.

Fullan

(2007) The New Meaning of Educational Change. New York: Teachers College Press.

21.

Gardner

(2007) The nature of leadership. In: Fullan

(ed.) Educational Leadership. San Francisco: Jossey-Bass, pp. 17–27.

22.

Genc

(2008) An evaluation of teachers’ views of primary school principals’ practice of democratic values. Social Behaviour and Personality 36(4): 483–492.

23.

Goleman

Boyatzis

McKee

(2002) Primal Leadership: Realizing the Power of Emotional Intelligence. Boston: Harvard Business Review Press.

24.

Gülbahar

(2016) Views of the teachers on the behaviours and the personal characteristics of the school principal which are effective in formation of a democratic school environment. Anthropologist 23 (1–2): 21–32.

25.

Hope

(2012) Small and perfectly formed? Is democracy an alternative approach to school leadership? School Leadership & Management 32(3): 291–305.

26.

Hyde

LaPrad

(2015) Mindfulness, democracy and education. Democracy and Education 23(2): Article 2.

27.

Johnson

Christensen

(2008) Educational Research: Quantitative, Qualitative, and Mixed Approaches. 3rd ed. Los Angeles: SAGE Publications.

28.

Johnson

(2017) Teacher perceptions of administrator leadership styles regarding job satisfaction. ProQuest LLC Dissertations Publishing, Lindenwold University. Available at: https://search.proquest.com/docview/1972896695 (accessed February 2020).

29.

Jwan

Kisaka

(2017) Democracy, ethics and social justice: Implications for secondary school leadership in Kenya. South African Journal of Education 37(3): 1–9.

30.

Kars

Inandi

(2018) Relationship between school principals’ leadership behaviours and teachers’ organizational trust. Eurasian Journal of Educational Research 74: 145–164.

31.

Keddie

(2015) School autonomy as ‘the way of the future’: Issues of equity, public purpose and moral leadership. Educational Management Administration and Leadership 44(5): 713–727.

32.

Kensler

Woods

(2012) A nested view of democratic leadership and community. Journal of School Leadership 22(4): 702–706.

33.

Kilicoglu

(2018) Understanding democratic and distributed leadership: How democratic leadership school principals related to distributed leadership in schools? Educational Policy Analysis and Strategic Research 13(3): 6–23.

34.

Kouzes

Posner

(2007) The five practices of exemplary leadership. In: Fullan

(ed.) Educational Leadership. San Francisco: Jossey-Bass, 63–75.

35.

McCormick

(2017) Leadership in a democratic school. ProQuest LLC, EdD Dissertation, University of Pennsylvania. Available at: https://search.proquest.com/docview/1958947327 (accessed February 2020).

36.

Mager

Nowak

(2012) Effects of student participation in decision making at school. A systematic review and synthesis of empirical research. Educational Research Review 7(1): 38–61.

37.

Mncube

Davies

Naidoo

(2015) Democratic school governance, leadership and management: A case study of two schools in South Africa. International Perspectives on Education and Society 26: 189–213.

38.

Møller

(2006) Democratic schooling in Norway: Implications for leadership in practice. Leadership and Policy in Schools 5(1): 53–69.

39.

Moos

(2011) Educating school leaders for democracy. Scholar–Practitioner Quarterly 4(4): 353–357.

40.

Moos

Huber

(2007) School leadership, school effectiveness and school improvement: Democratic and integrative leadership. In: Townsend

(ed.) International Handbook of School Effectiveness and Improvement. Dordrecht: Springer, pp. 579–597.

41.

Normore

Jean-Marie

(2008) Female secondary school leaders: At the helm of social justice, democratic schooling and equity. Leadership & Organization Development Journal 29(2): 182–205.

42.

OECD (2020) Are Students’ Career Expectations Aligned with their Skills? PISA in Focus No. 104. Paris: OECD Publishing.

43.

Parham

(1944) Democratic school administration. The Social Studies 35(6): 252–255.

44.

Pažur

Kovač

(2019) Democratic school leadership: Analysis of current researches and open issues. Metodički Ogledi 26(1): 33–60.

45.

Peker

Inandi

Gilic

(2018) The relationship between leadership styles (autocratic and democratic) of school administrators and the mobbing teachers suffer. European Journal of Contemporary Education 7(1): 150–164.

46.

Pont

Nusche

Moorman

(2008) Improving School Leadership. Volume 1: Policy and Practice. Paris: OECD.

47.

Quinn

Owen

(2016) Digging deeper: Understanding the power of ‘student voice’. Australian Journal of Education 60(1): 60–72.

48.

Riley

(2003) ‘Democratic leadership’ – a contradiction in terms? Leadership and Policy in School 2(2): 125–139.

49.

Rusch

(1995) Leadership in evolving democratic school communities. In: Annual Meeting of the American Educational Research Association, San Francisco, CA, 18–22 April 1995. Available at: https://eric.ed.gov/?id=ED392117 (accessed June 2020).

50.

Sant

(2019) Democratic education: A theoretical review. Review of Educational Research 89(5): 655–696.

51.

Shepherd-Jones

Salisbury-Glennon

(2018) Perceptions matter: The correlation between teacher motivation and principal leadership styles. Journal of Research in Education 28(2): 93–131.

52.

Stein

(2015) Schools needs leaders – not managers: It’s time for a paradigm shift. Journal of Leadership Education 15(2): 21–30.

53.

Stylianou

Savva

(2016) Investigating the knowledge management culture. Universal Journal of Educational Research 4(7): 1515–1521.

54.

UNESCO (2016) Leading Better Learning: School Leadership and Quality in the Education 2030 Agenda: Regional Reviews of Policies and Practices Available at: www.unesco.org/fileadmin/MULTIMEDIA/HQ/ED/pdf/leadership-report.pdf (accessed March 2018).

55.

Woods

(2004) Democratic leadership: Drawing distinctions with distributed leadership. International Journal of Leadership in Education 7(1): 3–26.

56.

Woods

(2005) Democratic Leadership in Education. London: Paul Chapman Publishing.

57.

Woods

(2007) A democracy of all learners: Ethical rationality and the affective roots of democratic leadership. School Leadership and Management 26(4): 321–337.