126 Sodobna pedagogika/Journal of Contemporary Educational Studies Let./Vol. 70 (136) Št./No. 3/2019 Str./pp. 126–148 ISSN 0038 0474 Iva Buchberger, Vesna Kovač, and Branko Rafajac Predictors of Distributed School Leadership in Croatian Primary Schools 1 Abstract: This paper discusses the features of distributed school leadership (DSL) in Croatian pri- mary schools with regard to its three central theoretical and practical dimensions: the participation of stakeholders in the process of school leadership, desirable social relations between the participants in the process of school leadership, and stakeholders’ influence over the decision-making process. The paper also assesses the development level of three selected groups of DSL predictors: supporting atti- tudes and activities of school principals, supporting characteristics of stakeholders participating in the decision-making process and supporting organisational and material resources. The results indicate that, in Croatia, the number of primary schools with less developed DSL is approximately equal to the number of schools with more developed DSL. The statistically significant positive correlation between the development level of all three predictor groups and the development level of the DSL features was also confirmed. These results suggest that introducing certain policy mechanisms could contribute to the enforcement of the DSL features in Croatian primary schools, which could in turn be of great help to policymakers at all levels of the decision-making process within the education system. Keywords: distributed school leadership, predictors of distributed school leadership, teachers’ partici- pation in school leadership, organisational and material school resources, leadership of school principals UDC: 37.091.113 Scientific paper Iva Buchberger, PhD., postdoctoral researcher, The University of Rijeka, Faculty of Humanities and Social Sciences, Department of Education, Sveučilišna avenija 4, HR-51000 Rijeka, Croatia; e-mail: ibuchberger@ffri.hr Vesna Kovač, PhD., full professor , The University of Rijeka, Faculty of Humanities and Social Sciences, Department of Education, Sveučilišna avenija 4, HR-51000 Rijeka, Croatia; e-mail: vkovac@ffri.hr Branko Rafajac, PhD., emeritus professor, The University of Rijeka, Faculty of Humanities and Social Sciences, Department of Education, Sveučilišna avenija 4, HR-51000 Rijeka, Croatia; e-mail: branko. rafajac@uniri.hr 1 Acknowledgements: W e wish to thank Associate Professors Nena R ončević, PhD , and Darko Lončarić, PhD , for their helpful suggestions and assistance in developing a research design and statisti- cal data processing. Buchberger , K ovač, R afajac 127 Introduction The researchers’ commitment and interest in the analysis of distributed school leadership (DSL) as the preferred leadership model spring from two central factors. First, in today’s social environment, school organisation is becoming increasingly complex, and thus principals’ roles and responsibilities are being significantly ex- panded (Harris 2008; Pont et al. 2008). A need for different stakeholders to share roles and responsibilities has consequently emerged. Secondly, the findings of recent empirical studies suggest a statistically significant correlation between the features of DSL and both school effectiveness and higher student achievement, as well as numer ous other variables such as teachers’ job satisfaction, teachers’ job commitment, etc. (Buchberger et al. 2018; Bruggencate et al. 2012; Choi Wa Ho 2010; Curry 2014; Hallinger and Heck 2010; Hulpia et al. 2011; Minckler 2013; Robinson et al. 2008; Somech 2010). Research on DSL has been encouraged in the educational policy’s environment, and recent policy documents (EC 2012; EC 2015) have also encouraged, explained and recommended this model of school leadership. Previous discussions have emphasised key features and dimensions of DSL (Choi Wa Ho 2010; Curry 2014; Hallinger and Heck 2010; Hulpia et al. 2011; Somech 2010). These discussions have been dominated by the argument that DSL should be seen as a complex phenomenon which includes pluralistic, cooperative and active stakeholder participation in the decision-making process; the importance of a holistic approach to the concept of DSL has also been highlighted (Holloway et al. 2017, pp. 4-5). It should be noted that, along with the DSL phenomenon, recent studies have attempted to outline the circumstances supporting DSL development, which can be described as DSL predictors (e.g. Choi Wa Ho 2010; Harris 2014, 2008; Hulpia et al. 2011; etc.). Analysing the Croatian education context revealed a lack of empirical findings on the presence and features of DSL (Braš Roth et al. 2014; Buchberger et al. 2017; K ovač et al. 2014; OECD 2016). The data on the supporting circumstances, i.e. the key DSL predictors which are the focus of this paper, could serve as a valid basis for the development of strategies to strengthen and improve DSL and contribute to the pluralistic, collaborative and active parti- 128 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac cipation of different stakeholders in decision-making processes in various school contexts. DSL Determinants and Dimensions DSL is understood as the participation of different stakeholders in the school leadership process, which is achieved by sharing and assuming roles in order to execute leadership activities. Based on the authority and responsibility of those involved, influence is spread to different stakeholders inside and outside the school, prompting social interaction, i.e. the formation of social relationships (Gronn 2003; Hallinger and Heck 2010; Harris 2007; Hulpia et al. 2011; Lambert 2005; Spillane 2005). Based on the aforementioned definition and previous scientific discussions, the dimensions of DSL can be categorised according to (a) participation; (b) in- fluence; and (c) social relations (Buchberger 2018, 2016), which will be further discussed in the following text. Participation The dominant participation of different stakeholders in school leadership activities, regardless of their formal positions, is attributed to DSL (Bush 2014; Bush and Middlewood 2013; Crawford 2012; Harris 2007, 2008, 2014). The participation of different stakeholders in leadership activities is not merely a delegation of activities and tasks by the leader (principal) (Harris 2014; McBeth 2008; Spillane 2006); rather, it is characteristic of DSL and implies the stakeholders’ autonomy and a proactive taking of the initiative in leadership activities. Participation in the decision-making process occurs when individuals or groups pursue their interests and autonomously contribute to choosing between different possibilities (Choi Wa Ho 2010; Curry 2014; Hallinger and Heck 2010; Hulpia et al. 2011; Somech 2010). When determining the features of such par- ticipation, the subjects participating in the decision-making process are clearly distinguished and involve different stakeholders—specifically , teachers, principals, other non-teaching staff, students, parents and other external stakeholders. The participation of teachers is particularly important, given their direct contact with students and their knowledge of curricula-related matters. Teachers can make the best decisions regarding the school’s fundamental objective—improving student achievement (Choi Wa Ho 2010; Hallinger and Heck 2010). Involving teachers in the decision-making process related to curricular and teaching issues increases the level of expertise relevant to that process. This provides access to key information which teachers possess by virtue of their immediate engagement in the classroom, which also contributes to the more effective decision-making process (Somech 2010, p. 180). Predictors of Distributed School Leadership in Croatian Primary Schools 129 Influence As a common element in many definitions of leadership, influence 2 is determ- ined by an individual’s or group’s intentional actions intended to achieve the goals, motivation or activities of other individuals or groups (Bush and Glover , 2014, p. 4). DSL can be seen as a process of sharing, redistributing and building mutual influ- ence among different stakeholders, irrespective of their formal roles (Harris 2014; McBeth 2008). Somech (2010) defined teachers’ participatory decision-making (ac- cording to Dachler and Wilpert 1978; McCaffrey et al. 1995) as the strong influence of teachers and other stakeholders on the proposal, consideration and adoption of final decisions related to issues perceived as important. Social relations In recent theoretical discussions, particular focus was devoted to the specificities of the social relations within which the process of leadership occurs (Northouse 2013; Spillan 2006). The definitions of DSL allude to developed social relations among involved stakeholders, which are grounded in collaboration, trust and sup- port (Harris 2014; Spillane 2005, 2006). The dimension of collaboration highlighted here includes the encouragement of dialogue and discussion in the decision-making process (Harris 2014, pp. 39-41). Having identified the development of collaborative relations as a prominent determinant of DSL, it can be stated that such collabor- ations’ existence can be recognised in the presence of developed group cohesion, precise division of roles among the stakeholders and stakeholders’ dedication to the achievement of their school’s common goals. In this respect, what stands out are the clearly defined and accepted mutual obligations and responsibilities of the stakeholders, the distribution of the resources among them and a certain degree of stakeholder readiness for change and adaptation (Adelman and Taylor 2000; Lasley et al. 1992). After the final interpretations of the DSL concept, it should be pinpointed that each leadership model is theoretical and thus difficult to achieve in a pure and ideal form in educational practice. Thus, the educational practice of school leadership could be described as a practice with more or less developed features of DSL. DSL Predictors Previous studies on DSL have singled out important predictors, i.e. vari- ables which could act as supporting circumstances for the development of 2 In this paper we decided to use the concept influence rather than power as an element of DSL because influence is described as non-coercive and democratic and is thus suitable for DSL. However, it is important to mention that researchers such as Lumby (2013, p. 586) have criticised DSL and singled out some authors (Fullan 2006; Harris 2008) for talking about a principal who uses his or her individual power to create an environment in which DSL can grow. 130 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac DSL. 3 Based on the presented theoretical discussions and conducted studies on DSL, DSL predictors can be categorised into three groups: (a) the supporting attitude and activities of principals; (b) the supporting characteristics of stakeholders particip- ating in the decision-making process; and (c) supporting organisational and material resources (Buchberger 2018, 2016). Supporting attitude and activities of principals This group of predictors refers to principals’ contribution to providing support, as well as building trust and collaborative relationships between stakeholders inside and outside the school (Harris 2014, p. 41). The role of a principal as a formal leader is crucial to monitoring and supporting informal leaders’ engagement—i.e. the engage- ment of leaders whose influence does not derive from a particular formal position (for instance, members of a school board, teacher mentors, etc.). Moreover, the findings of several empirical studies have contributed to the notion of principals’ supporting role, as the main focus of these studies was principals’ roles in the school leadership process and in ensuring greater school effectiveness (see, for example, Bruggencate et al. 2012, p. 703). These findings point to principals playing a significant role in activities related to the establishment of positive social relations, collaboration, mutual trust and open, supportive communication. The importance of supporting the professional development of stakeholders while encouraging and implementing innovation at school is also emphasised. Supporting characteristics of stakeholders participating in the decision-making process Previous studies have confirmed that a significant predictor of DSL develop- ment is a positive attitude on the part of stakeholders, as well as the perception of participating in leadership (decision-making) activities as an important duty and an integral part of their job. Consequently , theoretical discussions on DSL suggest that one of the constraints for DSL development is stakeholders’ resistance to participa- tion in the school leadership process (Harris 2014), especially when they believe that leadership is not an aspect of their job. Choi W a Ho (2010, p. 617) stated that some teachers tend to limit their role to implementing the prescribed curriculum, thus eliminating any aspect of school leadership as part of their duty. This is reflective of the importance of stakeholders’ attitudes towards school leadership. The studies mentioned above also revealed that the interest of stakeholders is an additional feature which contributes to DSL development. It should be recalled that proactive involvement of other stakeholders, who are very interested in the school leadership process, takes the form of realizing DSL in educational practice (Harris 2008; McBeth 3 A number of empirical studies on DSL and school effectiveness (e.g. Leithwood and Sun 2012) have revealed that certain organisational factors might have significant effects on specific dimensions of school leadership, particularly those related to organisational conditions (i.e., working environment, perceived working conditions, school culture and school climate) and those related to teachers’ and other stakeholders’ organisational behaviour (teachers’ job satisfaction, teachers’ commitment, trust, etc.). Predictors of Distributed School Leadership in Croatian Primary Schools 131 2008; Spillane 2006). In addition to the already-mentioned characteristics, it should also be emphasised that DSL is a challenging process which requires a high level of developed competences, ranging from strategic and critical thinking to the ability to motivate others (EC 2012). Supporting organisational and material resources Given the previous DSL studies, two key resources from the category of organ- isational and material resources can be distinguished: the availability of information and the availability of the time needed for the decision-making process. Based on the form participation takes—which should include informed, argument-grounded decision-making—it could be concluded that a precondition for achieving participation is nothing less than providing stakeholders with the information necessary to make informed decisions (Buchberger and K ovač 2017, p. 32). However , the importance of time to the process can be indirectly deduced from the results of those studies which have presented a lack of time as factor limiting DSL development (Harris 2014, p. 72). DSL, which includes the decision-making process based on these arguments and debates, implies that sufficient time and information for the decision-making is needed in such an environment. Methodology The following objectives were defined for this study: – To examine the distribution of Croatian primary schools based on the presence of participation, influence and social relations, and whether these dimensions belong to the category of schools with more or less developed DSL; – To examine the distribution of Croatian primary schools based on the presence of DSL predictors related to the supporting attitude and activities of principals, the supporting characteristics of stakeholders and supporting organisational and material resources; and – To examine whether there is a correlation between the development level of DSL predictors and the development level of DSL dimensions in Croatian primary schools. 4 The sample population used in this study comprised all central primary schools in Croatia (884), from which a sample of 80 schools was generated using the random 4 This research was conducted as part of Iva Buchbergerʼs doctoral thesis Leadership Facilita- tors and Characteristics in Croatian Primary Schools, which was part of a scientific research project called Investigating School Leadership From a Distributed Perspective in Croatian Schools (IScLEAD) (Croatian Science F oundation, project IP -2014-09-1825) and a project called The Examination of School Leadership Features in Croatian Primary Schools (The University of Rijeka, project 13.04.1.3.13). Branko R afajac, PhD , served as project manager . 132 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac selection method; 59 schools were finally included in the survey . The participants were teachers working in each randomly selected school (the examination was performed on 30 teachers employed by each sample school; in schools employing fewer than 30 teachers, all of the teachers were included). Due to the final number of schools in- cluded in the survey the empirical results cannot be used to suggest firm conclusions which can be generalised to all primary schools in Croatia. However, certain trends and tendencies about relations between key research variables have been revealed and discussed further below. The Questionnaire on Features and Predictors of Distributed School Leadership (Appendix 1), which was created and validated for the purpose of this study by the project research team, was used to gather data. The starting point of the questionnaire’s design process was the operationalisation of theoretical constructs and the templates of questionnaires used in similar studies which examined the correlation between school leadership and effective schools. 5 To verify and improve the validity of the instrument’s content, an expert validation was conducted which included a validity assessment of the content of all proposed items created on the basis of prior theoretical validation. The purpose of the expert validation was to ensure that certain descriptive DSL features, included in the Questionnaire on Features and Predictors of DSL, refer precisely to those features which ought to be described in the specific context of Croatian schools. The number of items per scale was as follows: Participation, 23 items; Influ- ence, nine items; Social relations, nine items; Supporting attitude and activities of principals, nine items; Supporting characteristics of stakeholders, seven items; and Supporting organisational and material resources, six items. All scales are Likert-type scales which rate the items from 1 (strongly disagree) to 5 (strongly agree). In order to further verify the instrument’s metric characteristics, i.e. the internal consistency reliability estimation, the measure of internal consistency (Cronbach’s α) was calculated. With respect to this, the following guidelines for the reliability estimation were used: α < 0,7, indicating that the scale is not reliable; and α ≥ 0,7, indicating that the scale is reliable (DeVellis 2003; Pallant 2011). The Cronbach’s as for the scales are presented in Table 1 below. Scale Cronbach’s a Participation 0,893 Influence 0,892 Social relations 0,935 Supporting attitude and activities of principals 0,951 Supporting characteristics of stakeholders 0,832 Supporting organisational and material resources 0,880 Table 1: Coefficients of the scales’ internal consistency 5 These questionnaires were the following instruments: The Distributive Leadership Inventory (Hulpia et al. 2011; Hulpia et al. 2009); the Teacher Collaboration Scale; the Perceived School-Leader Sup- port Scale; the Actual Participation in Decision Making Scale; the Satisfaction Concerting Participation in Decision Making Scale (Formalization of Participation in Decision Making) (Honingh and Hooge 2014); and the Subscale of Collaborative Leadership (Hallinger and Heck 2010). Predictors of Distributed School Leadership in Croatian Primary Schools 133 Data were processed via Statistical Program for Social Sciences (SPSS version 20.0, IBM). The basic measures of descriptive statistics (the measures of central tendency, the measures of variability , percentages) were calculated in order to describe the DSL features and predictors. Based on the questionnaire’s items referring to DSL features and predictors, composite variables were created according to the previously stated scales: the composite variable Participation: DSL feature; the composite variable Influ- ence: DSL feature; the composite variable Social relations: DSL feature; the composite variable Supporting attitude and activities of principals: DSL predictor; the composite variable Supporting characteristics of stakeholders: DSL predictor; and the composite variable Supporting organisational and material resources: DSL predictor. Further data processing and data analysis were carried out at the level of composite variables. The median split method and cluster analyses were used to examine the devel- opment of DSL features and predictors in Croatian primary schools and the corres- ponding percentages and frequencies were calculated. The data on schools obtained were divided into clusters (K-means) using the Median Split method, based on: (a) the leadership model (less developed DSL and more developed DSL); and (b) the development level of leadership predictors (less developed DSL predictors and more developed DSL predictors). Prior to conducting these analyses, the original data were transformed into standardised values (T values). 6 Further statistical data processing (the χ 2 test) was carried out at the level of the created cluster variables. To examine the contingency correlation between the level of DSL development and the level of the DSL predictors’ development, the χ 2 test was used. The χ 2 test was also used on the composite variables: (a) the development level of DSL and (b) the development level of DSL predictors. In addition, with the aim of determining the contingency correlation between the DSL development level and general school characteristics (e.g. school size, school environment), as well as the development level of the DSL predictors and general school characteristics, the χ 2 test was used. 7 Classification of Croatian Primary Schools Based on the Development Levels of DSL and DSL Predictors The preliminary results of the examined Croatian primary schools’ distribution based on the development levels of DSL and DSL predictors were obtained by calcu- lating the average T values of composite variables prior to the cluster analyses and school classification process. The results are presented in the Table 2 below. 8 6 T values were obtained using the formula: 50 + (10 × Z value). 7 To examine whether there was a contingency correlation between the DSL development level and both the school size and environment, as well as the development level of the DSL predictors and the school size and environment, several independent χ 2 tests were carried out. The analysis confirmed that there was no statistically significant contingency correlation in any of the examined cases, so the obtained findings were not further elaborated or interpreted for the study. 8 It should be noted that the higher value of the composite variable related to the leadership models indicates a greater presence of the DSL model’s features, which implies the dominant participation of various stakeholders in school leadership activities. This indicates that the involved stakeholders show a high level of interest and proactive involvement, along with developed social relationships based on col- laboration, dialogue, mutual support, trust and appreciation. Moreover , the higher values of the composite variable related to the distributed leadership predictors indicate a greater development level of principals’ supporting activities, interests, competences and positive attitudes towards the leadership of the involved stakeholders and the provided organisational and material resources needed for the leadership process. 134 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac SCHOOL ID T value DSL T value DSL predictors SCHOOL ID T value DSL T value DSL predictors S1 60.0145 59.5921 S31 51.7728 51.3427 S2 51.4275 51.9785 S32 45.6685 45.6835 S3 53.4942 41.5322 S33 50.9504 47.7321 S4 50.0945 52.4169 S34 51.0387 51.7515 S5 49.2501 48.3806 S35 45.9790 43.6047 S6 42.7778 43.2057 S36 51.7913 48.5420 S7 51.0556 55.3079 S37 48.8832 49.5148 S8 49.9343 46.5305 S38 57.6146 55.8565 S9 46.4306 46.4742 S39 50.0945 47.4375 S10 43.5154 45.9362 S40 51.7563 47.6469 S11 41.7482 45.5357 S41 52.7126 53.0038 S12 54.6132 53.6012 S42 53.7745 53.9741 S13 49.2188 49.0383 S43 44.2021 46.5867 S14 50.0465 49.3953 S44 53.6717 54.2839 S15 54.8864 58.3800 S45 52.4746 50.8735 S16 54.1389 52.7478 S46 54.4402 53.6360 S17 45.2159 48.6639 S47 46.0865 44.8920 S18 53.1578 55.1954 S48 58.9040 56.1308 S19 45.1692 44.3361 S49 59.7800 54.1585 S20 46.7576 47.5754 S50 46.5220 46.0038 S21 49.3737 51.4256 S51 48.8219 50.7471 S22 54.0888 52.8554 S52 56.2348 56.6583 S23 53.7622 52.5650 S53 41.7755 45.2047 S24 57.6490 55.2235 S54 45.4895 49.3100 S25 50.6526 53.3476 S55 45.8699 46.1269 S26 52.3870 53.9234 S56 46.3642 49.5376 S27 39.9555 42.3727 S57 52.8950 53.5869 S28 49.0812 48.5560 S58 54.4659 55.8199 S29 54.7750 55.2592 S59 49.9921 51.1418 S30 42.7611 41.0202 Table 2: The average T values of composite variables – leadership model and leadership predictors for each sample school Predictors of Distributed School Leadership in Croatian Primary Schools 135 The results indicate that the highest T values of composite variables were 60,0145 for DSL and 59,5921 for the DSL predictors. The lowest T values of composite variables were 39,9555 for DSL and 41,0202 for the DSL predictors. It could be preliminarily concluded that schools with average T values of composite variables < 50 belong to the clusters with less developed DSL features (that is, DSL predictors), whereas schools with average T values of composite variables ≥ 50 belong to clusters with more developed DSL features. 9 Once the preliminary categorisation was complete, a cluster analysis (K-means) was conducted in order to finally classify examined schools according to the develop- ment level of the DSL features and predictors. The iterative method of classification was used, with a maximum of ten iterations allowed. On the basis of composite variables (participation, influence, and social relations), the classification of schools was conducted with regard to the presence of DSL features. Two clusters were iden- tified: more developed DSL and less developed DSL. To estimate the success level of the classification, the analysis of variance for independent groups of results was carried out, which revealed a statistically significant difference between the groups for all three variables (participation F(1,57) = 65,781, P < 0,5; influence F(1,57) = 60,079, P < 0,5; and social relations F(1,57) = 132,237, P < 0,5), indicating the methodological justification of the implemented classification. Leadership features Frequencies % More developed DSL 35 59,3 Less developed DSL 24 40,7 Table 3: The classification of the examined schools with regard to their DSL development level The results presented in Table 3 above indicate that a higher, although not significantly higher , number of Croatian primary schools have more developed DSL. On t he basis of composite variables (supporting attitude and activities of prin- cipals, supporting characteristics of stakeholders participating in the decision-making process, and supporting organisational and material resources), the classification of schools was conducted with respect to the development level of leadership predictors. Two clusters were identified: more developed DSL predictors and less developed DSL predictors. To examine the success of the classification, an analysis of variance for independent groups of results was conducted. The analysis showed that there was a statistically significant difference between groups for all three variables—the variable supporting attitude and activities of principals F(1,57) = 64,624; P < 0,5; the variable supporting characteristics of stakeholders participating in the decision-making process F(1,57) = 73,113; P < 0,5; and the variable supporting 9 In Appendix 2, the average T values of composite variables for each sample school are presen- ted. A cluster analysis showed that 33 schools had a more developed participation, while 26 schools had less developed participation; 29 schools had more developed influence, and 30 schools less developed influence; 36 schools had more developed social relations, while 23 schools had less developed social relations. The average T values of each scale’s items for the selected schools are presented. 136 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac organisational and material resources F(1,57) = 80,804; P < 0,5) – which implies the methodological justification of the implemented classification. Leadership predictors Frequencies % Less developed DSL predictors 30 50,8 More developed DSL predictors 29 49,2 Table 4: The classification of examined schools with regard to the DSL predictors’ development level The conducted classification of participated schools according to the develop- ment level of DSL predictors (Table 4) indicated that the percentage of schools with more developed DSL predictors is relatively equal to that with less developed DSL predictors. 10 Correlation between Development Levels of DSL Features and DSL Predictors The χ 2 was used to examine whether there is a contingency correlation between the development level of the DSL features and the development level of the chosen DSL predictors. The results indicated a statistically significant contingency correlation between the DSL features’ development level and the DSL predictors’ development level (χ 2 (N = 59) = 32,196, P < 0,01; φ = 0,773). This in turn indicated schools with more developed DSL are more likely to be identified as schools with more developed DSL predictors. Comparing the empirical and theoretical frequencies showed that the obtained results were higher than the theoretical expectations in both cases (see Table 5). DSL Features Less developed More developed DSL predictors More developed Empirical frequencies 1 29 Theoretical frequencies 12,2 17,8 % leadership model 4,2 82,9 Adjusted standardised residuals -5,9 5,9 Less developed Empirical frequencies 23 6 Theoretical frequencies 11,8 17,2 % leadership model 95,8 17,1 Adjusted standardised residuals 5,9 -5,9 Table 5: Contingency table of DSL features and predictors 10 The cluster analysis showed that 29 schools had a more developed supporting attitude and activities of principals, while for 30 schools this predictor was less developed; 34 schools had a more developed sup- porting characteristics of stakeholders, while for 25 schools this predictor was less developed; 28 schools had a more developed supporting organizational and material resources, while for 31 schools this predictor was less developed (see Appendix 2). Predictors of Distributed School Leadership in Croatian Primary Schools 137 Discussion The results of the analyses conducted showed that the schools examined are in the slight majority of schools with more developed DSL. It should be noted that Croatian legislation supports the active involvement of not only principals but also the various stakeholders in schools’ decision-making processes. However, it is also important to remember that merely by ensuring support, this legislation ensured only the participatory dimension of the DSL (e.g. the participation of teachers, non-teaching staff, parents and founder—the remaining dimensions assessed in this research cannot be fully ensured. With respect to influence, most schools had a less developed influence dimension, meaning that even those stakeholders who participate in decision-making processes do not consider their process-related influence to be remarkable. It could also mean that they do not participate in the decision-making process related to the issues they perceive as important (Somech 2010, p. 177). This study’s findings suggest that teachers and other stakeholders rarely become proactively involved in leadership activities (Buchberger 2018), which could in turn reduce their proactivity in the decision-making process. Bearing in mind that the above categorisation of schools is the result of the perspective of one group of stakeholders—teachers—it is important to investigate the perspectives of other relevant stakeholders involved to various degrees in the decision-making process. Current studies suggest that teachers are usually more critical when as- sessing their participation in decision-making than other stakeholders, such as school principals. Principals also tend to assume teachers participate more than teachers do themselves (K ovač et al. 2014, p. 177). It can thus be assumed that external stakeholders such as parents or representatives of local educational authorities in school decision-making bodies might assess their participation in decision-making as being less than was indicated by teachers. The classification of sample schools according to the development level of DSL predictors implies an approximately equal distribution of schools in both clusters; the school classifications based on the development level of each predictor , however , produced slightly different ratios. Hence, the predictor supporting organisational and material resources was less developed in the majority of schools. This findin g is consistent with previous studies which show that Croatian teachers estimate the availability of information teachers need for the decision-making process, as a support ing organisat ional reso urc e, t o be very low (K ovač et al. 2015, p. 53). Furthermore, discussions regarding the availability of time teachers need for the decision-making process, as a second organisational resource, often indicate that the time teachers invest in non-teaching activities is not recognised (for example, their activities as mentors) (Matejčić Čotar 2017). This surely has a negative impact on teachers’ motivation to spend time on different activities other than teaching (project activities, leadership activities, etc.). This finding was expected, as the issues of insufficient financial resources allocated to the education sector (Braš Roth et al. 2017) and a lack of incen tives for teachers’ good work (K ovač et al. 2015) are continuously discussed in public and are probably among each school population in Croatia as well. Furthermore, the current discussions related to unequally dis- 138 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac tributed material investments in schools at the level of local self-government units (K ovač et al. 2017) could have impacted the obtained sample schools’ categorisation. The results of this study suggest that the development level of organisational and material resources which can be controlled and modified at different levels of the education system should be monitored separately in order to detect the correlation of particular groups of resources and the development level of DSL features. The findings related to the predictor supporting attitude and activities of principals are also of great interest as they coincide with the results of previous studies, implying that principals are often unwilling to delegate authority to other stakeholders, especially teachers (Peko et al. 2009, p. 76), and that the methods principals use to motivate their employees are inadequate (Blažević 2014, p. 15). In the Croatian education policy environment, the preconditions for principals’ autonomous functioning are not sufficiently fulfilled, a finding which is also supported by the results of PISA and TALIS research which described the relatively low level of school autonomy in some key decision-making aspects (OECD 2015; Braš R oth et al. 2017). In addition, if not provided opportunities to participate in appropriate training programs, it can be assumed that not all principals have equally developed the competences necessary for DSL implementation. The DSL predictors related to the supporting attitudes and activities of prin- cipals and supporting characteristics of stakeholders in the decision-making process function primarily at the level of a school as an organisation. However , recent studies have shown that it is also possible to affect these predictors and determine their contribution level to DSL development at the level of the policy environment and education system. Some international assessments of the implementation of the system-level policies and programs designed to provide teachers with incentives to motivate them to become part of the school’s management team (Holloway et al. 2017) point to the limitations of such initiatives, unless they are focused on all key DSL dimensions. A program recently implemented in the United States, 11 which stands as an example of a top-down initiative aimed at involving teachers in school leadership, ultimately resulted in strong tensions and dissatisfaction of involved teachers. Holloway et al. (2017) found that the key sources of teachers’ dissatisfac- tion are related to teachers feeling that their participation in leadership activities is limited, especially with respect to the activities they value the most, such as devel- oping positive social relations and empowering associates, and that their work in this regard is instead usually restricted to completing additional administrative tasks. Another interesting point should be mentioned. Based on the results related to the correlation between DSL development level in schools and general school charac- teristics (school size and school environment), as well as the correlation between DSL predictors in schools and general school characteristics, the examined correlations may not be statistically significant. It was expected that smaller schools or schools situated in rural areas would better facilitate the participation and collaboration of teachers and other stakeholders in the decision-making process (e.g. Hulpia et al. 2011; Bruggencate et al. 2012; Hallinger and Heck 2010). However, the findings of 11 The Teacher Advancement Program (TAP), which was implemented in the United States by the National Institute for Excellence in Teaching (NIET). Predictors of Distributed School Leadership in Croatian Primary Schools 139 this research do not support this expectation: They indicate instead that, for instance, a higher level of DSL development would be found in such schools. It appears that differences between schools are more likely to occur due to other environmental factors which were not examined in this research (for example, factors related to the type of school support provided by the local community or local self-government units). It may also be that, due to the ever-expanding globalisation and continuous information and technological progress, school differences based on general char- acteristics such as size or environment are being eradicated (Hodžić 2010). This research confirmed the assumption and revealed the tendency that there is a significant correlation between the development level of DSL features and the supporting attitudes and activities of principals, the supporting characteristics of stakeholders in the decision-making process and supporting organisational and material resources. Consequently , this finding can be used by all decision-makers at different levels of the education system in the process of analysing the mechanisms which empower DSL features. Concluding Remarks This paper was a review of empirical data confirming the assumption that DSL would be more developed in schools with more developed supporting attitudes and activities on the part of principals, supporting characteristics of stakeholders in the decision-making process and supporting organisational and material resources. According to these findings, and in the event that some would want to implement policy recommendations related to DSL effectiveness, decision-makers at different levels of the education system can invest in numerous mechanisms to strengthen all three groups of DSL predictors. Both national and local authorities should consider enhancing material investment in schools by creating more appropriate criteria and methods for resource allocation, creating additional training programs for principals and teachers which would enhance their DSL competencies and de- fining incentives for teachers which would recognise and reward their involvement in school leadership (for example, rewarding teachers’ leadership activities in the context of their career advancement). It should be emphasised that this study is one of the few conducted in Croatia (as well as in other national systems) focused on the school as a sample unit and which implements a research design aimed at describing school characteristics. Using the cluster analysis technique enabled the classification of sample schools based on the observed composite variables of DSL and DSL predictors, making this one of the first studies to offer concrete empirical data on the observed schools’ leadership models and the development levels of certain predictors for each sample school. This approach allowed the continuation of communication between re- searchers and schools, thus justifying a high level of practical applicability of this research and ensuring it can contribute to the creation of relevant guidelines for innovating professional practice. 140 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac The results of this research prompt new research questions, of which the fol- lowing should be mentioned: which demographic characteristics of schools (apart from size and environment) can affect the greater development of DSL features and predictors, and, above all, how and under what circumstances should DSL be developed, if necessary? Ongoing discussions are revealing possible directions of new studies, such as monitoring the effects of initiatives intended for DSL development at the state, local community or school level, with the purpose of determining its actual contribution to improving school effectiveness and student achievement. References Adelman, H. S. and Taylor, L. (2000). Connecting Schools, Families, and Communities. Professional School Counseling, 3, issue 5, pp. 298–307. Blažević, I. (2014). Rukovodeća uloga ravnatelja u školi. Školski vjesnik, 63, issue 1–2, pp. 7–21. Braš R oth, M., Markočić Dekanić, A. and Markuš Sandrić, M. (2017). PISA 2015 Prirodoslovne kompetencije za život. Zagreb: Nacionalni centar za vanjsko vrednovanje obrazovanja. Braš R oth, M., Markovčić Dekanić, A. and Ružić D . (2014). TALIS 2013: Glas učitelja i ravnatelja za bolje ishode poučavanja. Zagreb: Nacionalni centar za vanjsko vrednovanje obrazovanja – PISA centar . Bruggencate, G., Luyten, H., Scheerens, J. and Sleegers, P . (2012). Modeling the Influence of School Leaders on Student Achievement: How Can School Leaders make a Difference? Education Administration Quarterly, 48, issue 4, pp. 699–732. Buchberger, I. (2018). Pretpostavke i obilježja vođenja u hrvatskim osnovnim školama (Doc- toral thesis). Rijeka: Filozofski fakultet. Buchberger , I. (2016). Školsko vođenje kao čimbenik učinkovite škole. Napredak, 157, issue 1–2, pp. 165–186. Buchberger , I. and K ovač, V . (2017). Kritičko donošenje odluka u kontekstu školskog vođenja. Život i škola, 63, issue 1, pp. 29–40. Buchberger , I., K ovač, V . and Ažić Bastalić, A. (2018). Přínos distributivního vedení škol k pracovní spokojenosti učitel ů. Studia paedagogica, 23, issue 1, pp. 55–80. Bush, T . (2014). Applying Distributed Leadership across contexts. Educational Management Administration & Leadership, 42, issue 5, pp. 601–602. Bush, T . and Glover , D. (2014). School Leadership Models: What do we know? School Leader- ship & Management: Formerly School Organisation, 34, issue 5, pp. 553–571 Bush, T . and Middlewood, D. (2013). Leading and Managing People in Education. Los An- geles, New Delhi, Singapure, Washington DC: SAGE. Choi Wa Ho, D. (2010). Teacher Participation in Curriculum and Pedagogical Decisions: Insights into Curriculum Leadership. Educational Management Administration & Leadership, 38, issue 5, pp. 613–624. Crawford, M. (2012). Solo and Distributed Leadership: Definitions and Dilemmas. Educa- tional Management Administration & Leadership, 40, issue 5, pp. 610–620. Curry, K. A. (2014). Team Leadership: It’s Not for the Faint of Heart. Journal of Cases in Educational Leadership, 17, issue 2, pp. 20–40. Predictors of Distributed School Leadership in Croatian Primary Schools 141 European Commission. (2012). Supporting the Teaching Professions for Better Learn- ing Outcomes. Retrieved from: http://eurlex.europa.eu/LexUriServ/LexUriServ. do?uri=SWD:2012:0374:FIN:EN:PDF (accessed on 28. 4. 2013). European Commission. (2015). Strengthening teaching in Europe. Retrieved from http:// ec.europa.eu/dgs/education_culture/repository/education/library/policy/teaching- profession-practices_en.pdf (accessed on 6. 10. 2016). Gronn, P . (2003). The New Work of Educational Leaders: Changing Leadership Practice in an Era of School Reform. London: SAGE Publications, Inc. Hallinger , P . and Heck, R. H. (2010). Leadership for Learning: Does Collaborative Leadership Make a Difference in School Improvement? Educational Management, Administration and Leadership, 38, issue 6, pp. 654–678. Harris, A. (2014). Distributed Leadership Matters: Perspectives, Practicalities, and Potencial. Thousand Oaks: Corwin, A SAGE Company . Harris, A. (2008). What is Distributed Leadership? Retrieved from http://www.almaharris. co.uk/distributed_leadership.htm (accessed on 29. 1. 2014). Harris, A. (2007). Distributed leadership: conceptual confusion and empirical reticence. International Journal of Leadership in Education, 10, isuue 3, pp. 1–11. Hodžić, A. (2010). Globalizacija i obrazovanje, Pogled kroz prozor. Retrieved from: https:// pogledkrozprozor .wordpress.com/2010/08/29/globalizacija-i-obrazovanje/ (accessed on 15. 7. 2017). Holloway , J., Nielsen, A. and Saltmarsh, S. (2017). Prescribed distributed leadership in the era of accountability: the experiences of mentor teachers. Educational Management Administration & Leadership, pp. 1–18. Honingh, M. and Hooge, E. (2014). The effect of school-leader support and participation in decision making on teacher collaboration. Educational Management Administration & Leadership, 42, issue 1, pp. 75–98. Hulpia, H., Devos, G. and Van Keer, H. (2011). The Relation between School Leadership from a Distributed P erspective and T eachers’ Organizational Commitment: Examin- ing the Source of the Leadership Function. Education Administration Quarterly, 47, issue 5, pp. 728–771. Hulpia, H., Devos, G. and Rosseel, Y. (2009). Development and Validation of Scores on the Distributed Leadership Inventory. Retrieved from http://www.academia.edu/241761/ Development_and_validation_of_scores_on_the_distributed_leadership_inventory (accessed on 15. 10. 2014). K ovač, V ., Buchberger , I. and R afajac, B. (2015). O obrazovnoj politici iz različitih perspektiva. Rijeka: Filozofski fakultet u Rijeci. K ovač, V ., R afajac, B., Buchberger , I. and Močibob M. (2014). Obilježja obrazovne politike iz perspektive hrvatskih učitelja i nastavnika. Napredak: časopis za pedagogijsku teoriju i praksu, 155, issue 2/3, pp. 161–184. K ovač, V ., Rukavina K ovačević, K. and R afajac, B. (2017). Učinkovito upravljanje školama na lokalnoj razini. Croatian Journal of Education: Hrvatski časopis za odgoj i obra- zovanje, 19, issue 1, pp.175–207. K ovač V ., Staničić, S . and Buchberger , I. (2014). Obilježja i izazovi distributivnog vođenja škola. Školski vjesnik, issue 63, pp. 131–148. Lambert, L. (2005). Leadership for lasting reform. Educational Leadership, 62, issue 5, pp. 62–65. 142 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac Leithwood, K. A. and Sun, J. (2012). The Nature and Effects of Transformational School Leadership: A Meta-Analytic Review of Unpublished Research. Educational Admin- istration Quarterly, 48, issue 3, pp. 387–423. Lumby, J. (2013). Distributed Leadership: The Uses and Abuses of Power. Educational Management Administration & Leadership, 41, issue 5, pp. 581–597. Lasley, T . J., Matczynski, T . J. and Williams, J. A. (1992). Collaborative and Noncollabora- tive Partnership Structures in Teacher Education. Journal of Teacher Education, 43, pp. 257–261. Matejčić Čotar , S . (2017). Priča o jednom kraju koji je zapravo postao početak. P ogled kroz prozor (digital journal). R etrieved from https://pogledkrozprozor .wordpress.com/2017/01/31/ prica-o-jednom-kraju-koji-je-zapravo-postao-pocetak/ (accessed on 11. 7. 2017). McBeth, M. E. (2008). The Distributive Leadership Toolbox: Essential Practices for Successful Schools. Thousand Oaks; New Delhi; London; Singapore: SAGE. Minckler, C. H. (2013). School leadership that builds teacher social capital. Educational Management Administration & Leadership, 42, issue 5, pp. 657–679. Northouse P . G. (2013). Leadership: Theory and Practice. London: SAGE Publications, Inc. Organization for Economic Cooperation and Development [OECD]. (2016). PISA 2015 Results (Volume II): Policies and Practices for Successful Schools Retrieved from: https://www.oecd-ilibrary.org/docserver/9789264267510-en.pdf?expires=157026415 7&id=id&accname=guest&checksum=C3D646F67BEC848EC192FE67E94703CD (accessed on 24. 6. 2018). Organization for Economic Cooperation and Development [OECD]. (2015). Education Policy Outlook: Making Reforms Happen. Retrieved from http://www .oecd.org/edu/EPO%20 2015_Highlights.pdf (accessed on 6. 10. 2016). Pallant, J. (2011). SPSS priručnik za preživljavanje. Beograd: Mikro knjiga. P eko, A., Mlinarević V . and Gajger V . (2009). Učinkovitost vođenja u osnovnim školama. Odgojne znanosti, 11, issue 2, pp. 67–84. Pont, B., Nusche, D. and Moorman, H. (2008). Improving School Leadership. Volume 1: Policy and Practice. P aris: OECD . Robinson, V . M. J., Lloyd, C. A. and Rowe, K. J. (2008). The Impact of Leadership on Student Outcomes: An Analysis of the Differential Effects of Leadershi p Types. Educational Administration Quarterly, 44, issue 5, pp. 635–674. Somech, A. (2010). Participative Decision Making in Schools: A Mediating-Moderating Analytical Framework for Understanding School and T eacher Outcomes. Education Administration Quarterly, 46, issue 2, pp.174–209. Spillane, J. P . (2006). Distributed Leadership. San Francisco: Jossey-Bass. Spillane, J. P . (2005). Distributed Leadership, The Educational Forum, 69, issue 2, pp. 143–150. Predictors of Distributed School Leadership in Croatian Primary Schools 143 Annex 1. Questionnaire on Features and Predictors of Distributed School Leadership LEADERSHIP FEATURES ITEMS 1 – Strongly disagree 2 3 4 5 – Strongly agree 1 In our school, the decision-making is based on arguments and relevant information. 1 2 3 4 5 2 In our school, the principal usually makes decisions related to the school functioning independently. 1 2 3 4 5 3 In our school, only the principal defines the vision and goals of the school. 1 2 3 4 5 4 In our school, the principal consults and relies only on his/ her closest associates in the decision-making process. 1 2 3 4 5 5 In our school, only the principal and non-teaching staff participate in the decision-making process at the school level. 1 2 3 4 5 6 In our school, parents participate in the decision-making process related to the school functioning through formal bodies only (school board, parent council). 1 2 3 4 5 7 In our school, all or almost all employees are involved in the decision-making process. 1 2 3 4 5 8 In our school, important decisions are made by organizing extensive consultations with a large number of different stakeholders. 1 2 3 4 5 9 In our school, teachers participate in the decision-making related to the school issues. 1 2 3 4 5 10 In our school, representatives of the school founders actively participate in the decision-making related to the school issues. 1 2 3 4 5 11 In our school, local community representatives participate in the decision-making process related to the school functioning, since the school board is not their only means of cooperation with the rest involved in school. 1 2 3 4 5 12 In our school, the principal entrusts school employees with activities essential to the school functioning (e.g., activities of a quality assurance team). 1 2 3 4 5 13 In our school, the responsibility for the decisions made is borne by the principal and other school employees. 1 2 3 4 5 14 In our school, teachers are keen to take over the tasks associated with the decision-making at the school level. 1 2 3 4 5 15 In our school, apart from the principal, only the non- teaching staff has an important role in the decision- making. 1 2 3 4 5 16 In our school, all employees have an important role in the decision-making process. 1 2 3 4 5 17 In our school, teachers’ opinions and suggestions are taken into consideration even when decisions are made on the issues beyond their teaching domain. 1 2 3 4 5 144 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac 18 In our school, the participation of teachers in the decision- making process is considered important. 1 2 3 4 5 19 In our school, parents have a prominent role in making decisions related to the important school issues. 1 2 3 4 5 20 In our school, the representatives of the school founders influence the decisions on the important school issues. 1 2 3 4 5 21 In our school, students have a prominent role in the decision-making process. 1 2 3 4 5 22 In our school, suggestions and opinions are openly expressed at the meetings of teacher and/or class council. 1 2 3 4 5 23 Among our school’s employees, there is an exchange of information important to the decision-making process. 1 2 3 4 5 24 Stakeholders involved in our school understand their role in the decision-making process clearly. 1 2 3 4 5 25 In our school, the relationships between the employees are based on mutual support. 1 2 3 4 5 26 There is a mutual trust among our school’s employees. 1 2 3 4 5 27 In our school, the employees respect each other professionally and humanly when making decisions. 1 2 3 4 5 28 In our school, the principal makes the decisions and supervises their implementation alone. 1 2 3 4 5 29 In our school, the principal, with his/her several closest associates, defines the vision and goals of the school. 1 2 3 4 5 30 In our school, all important decisions are made by the school board and the principal, without the involvement of other employees. 1 2 3 4 5 31 In our school, other employees, parents, and students participate in the decision-making process. 1 2 3 4 5 32 In our school, parents who are not members of the school board and the parent council also participate in the decision-making process. 1 2 3 4 5 33 In our school, special bodies were established in addition to the regular ones, which are aimed at encouraging the school development (e.g., a team for the improvement of the school functioning) and involve the school employees. 1 2 3 4 5 34 In our school, teachers participate in the process of proposing and making decisions proactively. 1 2 3 4 5 35 In our school, teachers participate in making all the relevant decisions related to the school functioning. 1 2 3 4 5 36 In our school, teachers possess the appropriate level of autonomy when it comes to the decision-making process. 1 2 3 4 5 37 In our school, parental involvement in the decision- making process is considered important. 1 2 3 4 5 38 In our school, students’ involvement in the decision- making process is considered important. 1 2 3 4 5 39 At the meetings, the teacher council discusses the problem- solving and the improvement of the school functioning. 1 2 3 4 5 40 When deciding on the issues related to the school functioning, teachers openly express their opinions and suggestions. 1 2 3 4 5 41 Our school’s employees are willing to implement good decisions. 1 2 3 4 5 Predictors of Distributed School Leadership in Croatian Primary Schools 145 LEADERSHIP PREDICTORS ITEMS 1 – Strongly disagree 2 3 4 5 – Strongly agree 1 In our school, relevant information for the decision- making process is available to school employees. 1 2 3 4 5 2 In our school, teachers have enough time to participate in the decision-making process. 1 2 3 4 5 3 Due to the current distribution of working hours, our school’s non-teaching staff has enough time to participate in the decision-making process. 1 2 3 4 5 4 In our school, there are enough material resources for carrying out the activities that contribute to school development. 1 2 3 4 5 5 In our school, the general perception is that decision- making at the school level represents additional work load for teachers. 1 2 3 4 5 6 Most of our school employees are interested in participating in the decision-making process at the school level. 1 2 3 4 5 7 Most of our school employees are capable of making decisions at the school level. 1 2 3 4 5 8 In our school, the principal encourages teachers and non- teaching staff to engage in professional development. 1 2 3 4 5 9 In our school, the principal commends/rewards the school employees for a job well done. 1 2 3 4 5 10 In our school, the principal contributes to the development of mutual trust among school employees. 1 2 3 4 5 11 In our school, the principal contributes to the development of collaboration among school employees. 1 2 3 4 5 12 In our school, the principal establishes good collaboration with local self-government representatives (county and city education departments). 1 2 3 4 5 13 In our school, the principal establishes relations with local communities based on trust and collaboration (e.g., museums, libraries, etc.). 1 2 3 4 5 14 Due to the current distribution of working hours, our school’s teachers have enough time to participate in the decision-making process. 1 2 3 4 5 15 In our school, the non-teaching staff has enough time to participate in the decision-making process. 1 2 3 4 5 16 In our school, the general perception is that decision- making at the school level presents an obligation for all members of school employees. 1 2 3 4 5 17 In our school, teachers are interested in participating in the decision-making process related to the school leadership activities. 1 2 3 4 5 18 In our school, the non-teaching staff is interested in participating in the decision-making process related to the school leadership activities. 1 2 3 4 5 19 Our students’ parents are interested in participating in the decision-making process at the school level. 1 2 3 4 5 146 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac 20 In our school, the principal provides school employees with work-related support. 1 2 3 4 5 21 In our school, the principal encourages positive social relationships among school employees. 1 2 3 4 5 22 In our school, the principal establishes the relationship with parents based on trust and cooperation. 1 2 3 4 5 Annex 2. The presentation of the average T values of composite variables for each sample school SCHOOL ID T value participation T value influence T value social relations T value attitude and activities of principals T value characteristics of stakeholders T value organizational and material resources S1 58.89 58.09 58.52 57.50 58.05 60.73 S2 50.54 48.64 55.96 52.92 51.88 49.55 S3 54.89 50.41 50.33 36.70 53.37 41.00 S4 49.13 49.71 54.12 53.48 49.54 52.45 S5 50.75 48.55 45.27 49.94 46.93 48.50 S6 44.15 45.13 43.07 42.65 45.16 45.45 S7 49.08 51.88 59.17 56.80 51.14 54.67 S8 50.41 48.95 50.70 48.69 46.17 47.74 S9 46.97 46.66 47.11 48.91 46.47 44.32 S10 43.26 45.69 45.44 47.32 45.98 45.57 S11 40.61 43.53 47.56 47.60 46.10 43.74 S12 54.90 53.32 53.83 53.45 52.02 53.66 S13 49.61 47.27 50.78 49.84 47.58 49.69 S14 48.65 51.61 51.33 47.65 48.54 53.32 S15 55.12 56.26 56.22 56.57 56.93 58.50 S16 50.08 56.00 58.80 45.85 52.99 61.02 S17 46.53 50.15 45.05 48.37 46.86 51.59 S18 50.89 53.75 55.81 54.66 54.31 54.44 S19 48.43 45.89 44.75 43.73 51.36 46.01 S20 49.88 45.97 43.78 45.98 50.98 47.71 S21 49.73 49.64 52.41 49.62 54.48 49.63 S22 53.94 53.77 55.63 50.15 56.71 51.78 S23 55.53 50.38 51.89 53.10 51.55 51.49 S24 57.13 56.16 56.83 52.77 55.25 56.58 Predictors of Distributed School Leadership in Croatian Primary Schools 147 S25 50.49 49.79 50.26 50.55 54.64 51.52 S26 51.89 53.76 51.08 52.88 53.22 54.70 S27 38.66 39.51 44.33 42.15 45.80 42.78 S28 49.89 51.06 47.59 53.55 43.85 46.09 S29 54.65 53.76 56.46 55.47 51.95 55.78 S30 44.13 43.87 43.95 42.00 41.40 43.31 S31 52.59 49.00 52.74 54.31 48.03 49.35 S32 46.10 46.90 44.35 46.25 46.74 45.70 S33 52.93 47.49 43.85 47.57 48.75 47.98 S34 51.28 49.35 50.97 50.96 51.51 52.10 S35 47.27 46.69 43.13 45.53 41.82 45.41 S36 50.46 53.37 52.17 47.05 50.72 49.33 S37 47.79 47.60 50.37 49.22 50.62 47.74 S38 57.30 57.34 56.42 55.72 52.64 55.85 S39 50.37 49.55 50.18 46.77 48.06 49.03 S40 51.07 49.16 51.40 50.30 45.79 47.16 S41 51.75 50.68 53.94 53.26 53.14 50.97 S42 54.73 52.47 50.66 51.85 53.56 55.85 S43 46.02 46.26 43.74 45.11 51.74 45.21 S44 52.67 52.27 52.55 55.39 52.07 52.62 S45 52.20 51.68 52.73 52.01 50.02 49.47 S46 54.16 54.93 53.53 55.39 50.98 51.67 S47 46.98 51.27 42.58 45.37 46.85 45.13 S48 53.52 60.48 55.25 53.01 56.71 57.55 S49 57.27 56.74 57.30 55.96 55.22 52.81 S50 45.78 48.06 46.15 46.60 46.85 44.93 S51 50.52 48.50 47.44 48.28 52.72 49.96 S52 55.61 55.74 57.09 55.87 56.88 54.89 S53 41.76 47.02 40.68 43.26 47.01 48.52 S54 44.19 48.09 46.72 51.72 47.17 49.11 S55 44.58 47.55 47.56 44.74 48.02 48.20 S56 46.09 49.08 47.56 49.84 48.70 50.05 S57 53.22 52.37 53.10 53.80 51.46 53.68 S58 54.97 54.93 50.28 54.85 54.93 55.47 S59 49.85 51.69 49.22 48.97 50.76 55.09 148 Sodobna pedagogika/Journal of Contemporary Educational Studies Buchberger, Kovač, Rafajac Iva BUCHBERGER (Univerza na Reki, Hrvaška) V esna K O V AČ (Univerza na R eki, Hrvaška) Branko RAFAJAC (Univerza na Reki, Hrvaška) NAPOVEDNIKI DISTRIBUIRANEGA VODENJA V OSNOVNIH ŠOLAH NA HRVAŠKEM Povzetek: V prispevku obravnavam o značilnosti distribuiranega vodenja na hrvaških osnovnih šolah, zlasti njegove tri osrednje teoretične in praktične razsežnosti: (i) participacijo vseh deležnikov v procesu vodenja šolske institucije, (ii) zažele ne socialne vezi med udeleženci v tem procesu ter (iii) vpliv , ki ga i m a j o r a z l i čn i de l e ž n i ki n a s p r e j e m a n j e od l oči te v . O b te m v p r i s pe v k u oce n i m o tu di dos e ž e n o r a v e n tr e h izbranih skupin napovednikov distribuiranega vodenja, in sicer (i) podporna ravnanja in dejavnosti rav- nateljev , (ii) podporne značilnosti deležnikov , vključenih v procese sprejemanja odločitev , in (iii) podporni organizacijski in materialni pogoji. Glede na rezultate raziskave je mogoče skleniti, da je na Hrvaškem število osnovnih šol z manj razvitim distribuiranim vodenjem približno enako številu šol, kjer je moč zaslediti bolj razvito distribuirano vodenje. P otrdili smo tudi statistično značilno pozitivno korelacijo med doseženo ravnijo vseh treh skupin napovednikov ter doseženo ravnijo značilnosti distribuiranega vodenja. R ezultati tako nakazujejo, da bi uvedba določenih sistemskih mehanizmov lahko prispevala k uveljav ljanju značilnosti distribui ranega vodenja v osnovnih šolah na Hrvaškem, kar bi lahko bilo v pomoč odločevalcem na vseh ravneh sprejemanja odločitev v hrvaškem vzgojno-izobraževalnem sistemu. Ključne besede: distribuirano vodenje, napovednik i distribuiranega vodenja, participacija učiteljev pri vodenju šole, organizacijski in materialni pogoji šole, ravnateljevo vodenje E-naslov: ibuchberger@ffri.hr