Abstract
Introduction
Although the context of schooling in various countries can differ greatly, research on school management and leadership can be supportive for improving education in developed as well as developing countries. Information on the effectiveness of school inputs is important for evidence-based sensible managerial decisions and practices at different levels in the country’s education system. At the same time, we have to deal with the issue of universal versus context-specific effectiveness of management practices. There are several indications that some principles of management are applicable universally, while others are much more sensitive to local and cultural variation. While attention for cultural dimensions of school and management improvement is important, researchers worldwide encounter numerous similarities to school (and management) effectiveness in different cultures (Brophy, 2000; Kantamara, Hallinger, & Jatiket, 2006; R. H. Hofman, Hofman, & Gray, 2008).
Throughout the world, policy makers are seeking to restructure and renew educational systems that have been struggling to keep pace with rapidly changing environmental demands. Social and economic development in the developing nations of Southeast Asia over the past decade is comparable to three generations of change in the industrialized societies of Europe and the United States. While Southeast Asia policy makers have conceived ambitious new educational policies consistent with their evolving social, political, and economic aims, the same governments are finding it difficult to implement these policies (Hallinger & Kantamara, 2000; Kantamara et al., 2006; Reynolds, Creemers, Stringfield, Teddlie, & Schaffer, 2002). What is needed is smart management, which is a type of management that sets goals and objectives that are specific, measurable, acceptable, realistic, and time-bound (Doran, 1981). The crucial point here is that educational and managerial choices can make the difference between good and less effective education. When school management focuses on ineffective practices or the wrong use of basically good policy (e.g., decentralization of education without sufficient guarantees for quality for all), even rich countries will possibly score low in terms of education effectiveness. It is also a fact that poor countries making the right choices will be able to score high by comparison. An example for this situation can be found in the TIMSS study (1995) where Thailand and Slovenia score at the level of the Netherlands and even higher than the United States on math (eighth grade); 10 years later Thailand is doing less well than the United States (respective ranking of 35 and 44), with Slovenia on the 19th rank and the Netherlands ranked at 5th position (OECD/PISA, 2009). R. H. Hofman et al. (2008) and W. H. A. Hofman, Hofman, and Gray (2010) conducted trend analyses showing that European countries are quite stable in terms of their math performance and that the average math performance of low scoring countries is not easily improved (Hofman, Hofman, Gray, & Daly, 2004).
Since the publication of the Coleman report the term school effect has been associated with the size of differences between schools in students’ achievement outcomes, after adjusting in some way for intake. There is now a widespread consensus that controlling for students’ prior achievement is indispensable if schools are compared on the basis of attainment in examinations or tests. Thus, effectiveness is defined as that part of the variance in current measured achievement not predicted by a student’s prior achievement and other controls and identified at the school level (Sammons & Luyten, 2009). Strong, especially educationally focused, management is considered to be one of the key characteristics of an effective school (Teddlie & Reynolds, 2000). This article focuses on student learning from a management perspective and examines how the stimulation of an effective educational management climate affects the learning of students. The first underlying assumption in this article is that school leadership exerts strong influences on the students’ social and learning environments.
A second underlying assumption in this article is that different schools need different management practices. There are strong differences between schools in terms of their student populations, environment, and place in the community. Some secondary schools suffer from specific problems such as truancy, dropout, lack of consensus among teachers and management, and conflicts within the school staff (Hattie, 2009; R. H. Hofman, Hofman, & Guldemond, 2003). Such problems and conflicts could be linked to the contextual characteristics of the school, such as the type of school population, or to influences of community stakeholders surrounding the school.
Research is needed to demonstrate whether one effective management style is the best style for all schools or whether the best management style depends upon certain characteristics of secondary schools. It is likely that management styles are contextually sensitive to differences in student populations and this will also influence the contacts with parents and with the community around the school.
There are three kinds of secondary education in the Netherlands: prevocational secondary education, which takes 4 years; senior general secondary education, which takes 5 years; and pre-university education, which takes 6 years (Eurydice, 2010).
The objective of the study presented here is to clarify how general and vocational education schools can be distinguished according to different leadership or management styles. An empirically based picture of the types of management that the two types of secondary schools (school principals and heads of faculty) employ will be presented. The basic question is:
Do schools for general and vocational secondary education in the Netherlands differ in terms of effective or smart management styles?
Theoretical Framework: Configuration Theory of Management
Recent research (e.g., Fink & Brayman, 2006; Fullan, 2007; Geijsel, Sleegers, Leithwood, & Jantzi, 2003; Hargreaves & Fink, 2006) has shown the importance of a multidimensional approach toward school leadership and school management, and as a result, this study proposes a configuration approach that takes into account such a multifaceted perspective toward the management of secondary schools.
Drawing on findings from organizational and contingency theory (Mintzberg, 1979), we postulate that studying specific types or configurations of secondary schools will be more fruitful than focusing on the influence of single variables on the effectiveness of schools.
Moreover, we should take into account the fact that management practices in secondary education are multidimensional and that school principals develop a variety of management styles. Some school principals concentrate on the managerial and administrative aspects of their task, whereas others stress the importance of educationally and instructionally focused leadership. Furthermore, there are schools in which a great deal of the educational decision making is being transferred to the faculty leaders (so-called leadership substitutes; Firestone, 1996; Geijsel et al., 2003; Hargreaves & Fink, 2006; Sergiovanni, 1992). Consequently, research into the effectiveness of secondary schools should closely examine the relationship between the management practices of the school principal and that of the faculty heads. Only then can the various management practices found in secondary schools be justified. Research into the social, instructional, and monitoring practices of the faculty alongside the work of the school management could result in powerful lessons concerning the role of management and help us to improve both students’ learning processes and the effectiveness of schools (Lomos, Hofman, & Bosker, 2009).
Central to our configuration theory is the assumption that the effectiveness of schools depends upon how their more formal management characteristics fit in with the specific, more cultural management elements that seem to produce an integrated and effective school. In line with this view, we wish to point out the relevance of different integration mechanisms that schools may use to develop a consistent and effective school organization.
Integration Mechanisms in Secondary Schools
Integration is the general term under which unifying arrangements in organizations are subsumed. A broad set of possible integration practices is available to school principals and faculty heads in secondary schools. Mintzberg’s (1979, 1983) original organizational theory distinguishes between six coordination mechanisms that are highly suitable as a paradigm of effective school management. Although not all of them apply to the managerial practices of secondary schools, part of these integration mechanisms could in general be a helpful tool to describe and typify secondary schools. By linking these with the outcomes of effective school literature, we can distinguish a set of four integration mechanisms that could be important management tools for principals and faculty heads of secondary schools.
The first integration mechanism stimulates a “mutually adjusted influence structure” in secondary schools. Effective school principals consider it vital to ensure that teachers’ views are taken into account, and this positively relates to an effective and self-improving school (Fink & Brayman, 2006; Hargreaves & Fink, 2006; Harris, Chapman, Muijs, Russ, & Stoll, 2006; Leithwood, Tomlinson, & Genge, 1996; Mortimore, Sammons, Stoll, Lewis, & Ecob, 1988). The study of Bryk and Frank (1991) also reveals that research on school organization underscores communication within the faculty and highlights the significance of shared decision making in educational matters. However, not only the teachers and the faculty are important in promoting a policy of “mutual adjustment” in the school. The influence on the school’s policy and the educational processes of various other school members, such as the members of the school board, the parent-teachers association, other parents, and students, also seems to contribute to more effective, mutually adjusted schools. R. H. Hofman et al. (2008), as well as Mortimore et al. (1988), found that regular parental involvement in the school life and in school board decision making is more influential than that of formal parental organizations.
The second integration mechanism is called “educational supervision,” and in our case it concerns the educational leadership features of the school head and the faculty leaders. Effective leadership focuses on instructional leadership (both school head and faculty head) and on the stimulatory and motivational supervision of the teachers’ instructional process (Creemers & Kyriakides, 2006; Hallinger & Heck, 1996; Hattie, 2009). Effective management styles create and promote an achievement-oriented school policy, which is based on the regular monitoring of student progress and the optimal functioning of the teachers in the school. According to Hattie (2009), the type of leadership is an important moderator, and based on a synthesis (meta−meta-analysis) of more than 800 meta-analyses about influences on learning, including leadership (styles), he concludes that a transformational leadership style is the most powerful. School leaders who focus on students’ achievement and instructional strategies are the most effective. Hattie presents a mean effect size of Cohen’s d = .57 leadership influences on student achievement.
The third mechanism involves “standardization through work processes and output,” and it concerns the degree to which principals and faculty heads have made arrangements about teaching strategies, learning objectives, learning content, use of homework, and tests (Hattie, 2009; R. H. Hofman et al., 2003). It focuses on the standardization of teachers’ behavior toward students: the extent to which teachers behave according to a set of prescribed school rules concerning truancy and students arriving late, rules on the teachers’ testing policy, as well as rules set for classroom behavior. Promoting management practices based on this integration mechanism makes it clear for both the teachers and students as to what the school staff stand for as well as which set of rules and norms students should comply with. This should result in a safe and educationally supportive learning environment that will promote better learning attitudes and higher achievement motivation (Bryk, Lee, & Holland, 1993; Hattie, 2009; Teddlie & Reynolds, 2000).
The fourth mechanism that secondary schools’ principals and faculty leaders can use is referred to as “standardization through skills and norms,” and this focuses on (site-based) staff development policy and practices used in the school and the department (Boyle, Lamprianou, & Boyle; 2005; Hattie, 2009; Mortimore et al., 1988). Effective school managers seem to distinguish themselves from the less effective ones by demonstrating a strong interest in the professionalization of their team members as a tool to improve the school as a social and educational entity (Boyle et al., 2005; Levine & Lezotte, 1990; Lomos et al., 2009). A smart management style focuses on the binding and motivating factor of continuous professional development in a school because without such a binding factor schools will be “loosely coupled” organizations (Rosenholtz, 1985; Weick, 1976). Consequently, this mechanism concentrates on the monitoring of teachers’ skills and the improvement of (“starting”) teachers’ instructional skills (Hattie, 2009; R. H. Hofman et al., 1999; Teddlie & Reynolds, 2000).
In this study we develop a configuration theory of educational management by assessing the degree and extent to which these integration mechanisms are applied by principals and faculty heads in vocational and general education and use that empirical information to typify the management practices of secondary schools. The following section presents information on the design of the study and the variables it contains. Thereafter, the background and procedures that have led to the construction of different types of secondary school management will be described.
Method
The multilevel model realistically reflects the nested or hierarchical nature of data found in school-effect studies. The analyses were carried out with VARCL (Longford, 1993). We first of all clarified which part of the total variance was situated at the school (including department) level and which part at the student level. By including student input characteristics in the second model and school input characteristics in the third multilevel model (the so-called covariate models), we provided fair effectiveness scores (value added) of schools. Next, four theoretical models were used to estimate the degree to which integration mechanisms at management level contribute to variation in the effectiveness of secondary schools in general and vocational education.
However, it should be noted that we assumed that apart from the independent main effects that seem to point to effective management practices, the variation between the differences in the schools’ effectiveness will be more strongly explained by the interactions of these principal effects. Therefore, we determined the joint effects of composed indicator variables (or configurations of integration mechanisms) on the student outcomes. A multidimensional scaling procedure is employed to create such configurations using our indicators of effective management (i.e., the set of integration mechanisms). In these analyses we made use of variable scores transformed into z scores to take differences in range values into account.
In the next section, we will present the variables and indicators of our study before we discuss how the school configurations are constructed.
Data
This study analyzes data from two types of secondary schools, vocational and general education, and the math performances of their students. Longitudinal data were collected from 121 secondary schools in the framework of a large-scale cohort research. Our sample includes 2,805 students form 65 secondary schools for general education and 1,081 students from 56 schools for vocational education. Students, teachers, parents, faculty leaders, and school principals were asked to complete a questionnaire.
Student Level: Control Variables
This study uses seven student characteristics as covariates. These variables were determined at the beginning of the first year of secondary school.
The first one is students’ school advice (advice), which concerns the score students receive from the primary school head teacher. This is indicative of their level at the beginning of their secondary school career. The second one concerns the type of school the student actually attends (school type) and includes four types ranging from 1 = prevocational school type, to 2 = (lower) and 3 = (higher) general education, to 4 = pre-university education. The third covariate is socioeconomic background(s), which includes the educational attainment level of the mother. Furthermore, the ethnic background of the student (ethnicity) is measured as 1 = native and 2 = ethnic minority, as well as whether or not the student is raised in a one-parent family (1 = parent).
Another student control variable concerns the student’s intelligence (psb3), which is measured by a nonverbal subtest consisting of 40 items designed to measure students’ ability to argue logically. The last covariate is the students’ motivation for achievement (prestmot), and this is measured on a scale of 11 items with a reliability of Cronbach’s alpha = 0.77. Some examples of the applied items are: “I would rather . . . (not = 1 up to very much = 4) like to be the best of the class” and “In studying I have . . . (low = 1 up to very high = 4) certain requirements of myself.” School advice and psb3 control variables are a reflection of the students’ intellectual abilities, while prestmot indicates students’ efforts and motivation for achievement. School type has been introduced as a covariate to control for the initial variation in math achievement between general and vocational education.
School Level: Control Variables and Integration Mechanisms
This study distinguishes three covariates at the school level. In the first place is the percentage of ethnic minority students (minority %) in the school. Furthermore, we take into account the denomination of the school (public, Catholic, Protestant, neutral, or other; sdenom). The category “other” indicates the schools with missing scores on this variable. The last one concerns the community type of the school, which ranges from rural to urban (urban).
This study works from the premise that school management in secondary education is multidimensional and that school principals develop a variety of management styles. It also presumes that there are schools in which part of the managerial tasks, more often the educational decision making, are being transferred to the faculty leaders. Based on Mintzberg’s coordination mechanisms and from the viewpoint of the two (Mintzberg, 1979) managerial levels in secondary schools, we developed a set of indicators for the integration mechanisms. The content of these have been described in the third section of this article paper, and the psychometric characteristics of the scales that have been constructed to operationalize them are presented in the appendix.
Dependent Variable: Math Performances
Math performance was determined by means of a mathematical test at the end of students’ third school year. Secondary school students were asked to complete a national standardized test, which was developed by CITO (the Dutch national testing service). These tests represent the learning content that must be taught and examined in primary and secondary schools. The content is specified according to a specific set of national objectives and standards. The test we used had a version for students in general education and one for students in vocational education. The two tests contain a total of 31 and 32 items, respectively, with a maximum score of 101 and 77 points, respectively. The relationship between the two math scores is very high (.92). Correlation of the two math scores of students in the third year with the score on math in the first year of secondary school shows correlations of, respectively, 0.70 and 0.73 (Hustinx, Kuyper, & van der Werf, 2005). The reliability of both tests is adequate (Cronbach’s alpha is 0.81 and 0.87, respectively).
Configurations of Integration Mechanisms
A hierarchical type of cluster analysis is employed to find empirically based configurations based on our (indicators of) integration mechanisms.
The results of the cluster analysis are presented graphically in Figure 1. This shows three empirically based types of school, distinguished by how they make use of the set of integration mechanisms.
Cluster analysis results
The graphical presentation of the scores for each of the three configurations shows the kind of management types that are seen in secondary schools.
The first management configuration (n = 67 schools) includes more than half of the secondary schools in our sample. Inspection of the graphs in Figure 1 shows that this type of management scores modest to low on our four sets of integration mechanisms. In short we can typify this type as modest school-based management and nonexistent faculty management.
The second management configuration (n = 36) includes a third of the secondary schools in our sample. This type of school management exhibits quite a lot of variation in our indicators of integration mechanisms. The graphical presentation shows that this type of school management looks the opposite of the first configuration. In short, this type can best be described as faculty-based management focusing on instruction and skills but lacking consensus.
The third management configuration (n = 18) is clearly the opposite of both the first and the second type. Schools with this type of management combine strong school-based management with strong faculty-based management. Only 15% of all secondary schools in the sample qualify as this type. In short, it seems appropriate to typify these schools as strong and integrated school and faculty management.
The question to be answered is whether these types of school management contribute to the math performances of the students attending these schools. Yet, even more important is the question as to whether our three management configurations relate in the same way to the math performances of students in general education as they do to those in vocational education. To answer these questions, two sets of separate multilevel analyses were conducted: one for vocational and one for general education.
Results of the Multilevel Analyses
Variance at School and Student Levels
To start with, we again established the size of the potential contribution of the school on the one hand and the student on the other to the math performances. The interschool variance in mathematical performances of general education students was greater than that for vocational education students, 33% versus 23%, respectively. The outcomes of these multilevel analyses are given in Table 1.
Multilevel Analyses: General and Vocational Education Effects With Math Performances
Note. NSD = no significant difference.
Impact of Covariates at Student and School Level
With respect to the influence of the covariates at pupil level we can conclude that general education and vocational education strongly concur: For both school types there are significant effects of advice, nonverbal intelligence, and social background on the math performances. Up until now the school covariates do not play a role in vocational education. 1 The only school-level covariate that exerts an influence on math performances in general education, even after adjustment for the student level characteristics, is the denomination of schools. Using public schools as our baseline group in the analysis, we observed substantially lower performances within private but religiously neutral (bijzonder-neutraal) schools compared to public ones and private Protestant and Catholic schools for secondary education (see Table 1).
Effectiveness of Integration Mechanisms in General and Vocational Education
Influence model
With respect to the so-called influence model, we noted differences between general education and vocational education (see Table 2). In general education the more effective school management type seems to be the one that allows for a relatively strong influence on the board’s financial, educational, and managerial policy at the different school layers.
Multilevel Analyses: General and Vocational Education Effects With Math Performances
Note. NSD = no significant difference; NSE = no significant effect (standard error).
The vocational student appears to perform worse on the mathematics test when there is a relatively large influence on the school policy and the educational content by the split sites leader. Furthermore, it seems that in vocational education, there is yet another important school covariate: the types of education offered in a school. With the combination of these two outcomes it seems probable that the vocational pupil performs less well at split sites of broad comprehensive schools.
Educational steering
The second content-based model is that of educational steering or supervision by the school management or head of department. It is striking that whereas in general education no supervisory factors play a role, this is very much the case in vocational education. Both the degree of monitoring of teachers’ performance as well as the amount of contact between parents and school were positively correlated with math performances of vocational education pupils (Table 2). The last observation makes it clear that for vocational education pupils in particular, a good contact between parents and school and/or the parental involvement in school is beneficial for the performance of their children. It is also noticeable that when school departments place an accent on integration between subjects, matching didactic work methods, and thematic education, such matters do not benefit the pupil. It is possible that a vocational education pupil will benefit more from a strongly subject-oriented course as opposed to a broader course.
Standardization
Also with respect to our other integration models—which assume that integration in the school is brought about by the standardization of work processes, output, skills, and norms—we observed that various factors affect the performance of vocational education pupils. Positive effects assume the perception of the contribution of the school to pupil performances. There is a positive effect of school managements who place great emphasis on the contribution of the school to the math performances of the vocational education pupil. Furthermore, a positive effect has been established for the presence of a professionalization plan in or for the faculty in vocational education. The negative correlation between the degree to which innovations are realized and the percentage of teachers participating in professionalization in association with this is striking.
In general education we observed, in contrast to vocational education, no influences of the degree of standardization on the math performances. All in all, after concluding the discussion of independent effects, we can establish that there is a clear difference between general and vocational education. Yet the most important question is whether this is also true for the three management configurations.
Smart Management Configurations in General and Vocational Education?
In general education, the average math performance is significantly higher in schools characterized by the third management configuration, which included the secondary schools with strong and integrated school and department management. This means that a management style allowing the different layers a relatively strong influence on school policy is more effective. Schools that decentralize power to the faculty and stimulate parental involvement show particularly good results. An effective management style typically shows strong educational supervision from faculty heads (see Hattie, 2009). In these schools the rules structure is also adequate, that is, it is clear and carried out consistently. Within departments of the effective management configuration type, a relatively high level of commitment is made with respect to curriculum content and procedures concerning homework, exams, extra instruction time, and student feedback. Furthermore, these departments are characterized by activities to enhance professionalism, a lot of teamwork, and finally, strong consensus about the school and faculty policies relating to educational goals, content, didactics, rules, and even financial and personnel management.
Next to this positive effect on student’s math performance, an additional but negative effect was observed for general education with respect to the second management configuration. This is the management style that includes the secondary schools with faculty-based management focusing on instruction and skills but lacking consensus.
The hypothesis that configurations of school and department management characteristics are more likely to have an additional effect on math performances than the independent effects or integration mechanisms is also now confirmed in vocational education. However, unlike in general education, the second management type, in addition to the independent effects, exhibits an additional effect on the mathematical performances. Therefore, the management type of faculty-based management focusing on instruction and skills but lacking consensus exhibits a positive effect on math performance.
All in all we can conclude that there are some differences between vocational and secondary education students with respect to an effective management context. This also fits in with the outcomes of research at the class level, which also revealed that the nature of an effective teaching environment was very different for vocational and general education students. Clearly the application of class and school management is strongly determined by the type of student (Hofman et al., 1999).
The theoretical influence of school-level characteristics and the actual influence of the formulated models on math performances are shown in Table 3.
Explained School-Level Variance (%) in Math Performances
Two issues are clearly visible. The first is that the interschool variance in math performances is greater in general education than in vocational education, 33% versus 23%, respectively. In principle, there are therefore more possibilities for the school in general education to exert an influence than is the case for vocational education. We subsequently observed that the models distinguished, put more precisely the indicators distinguished within these, have a diverse impact on math performances in general and vocational education. For example, the influence model accounts for a quarter of the interschool variance in vocational education, whereas for general education this model can only account for 3% of the interschool variance. A comparable situation is found for the standardization and supervision model. The (additional) influence of the configurations distinguished or the integration mechanisms is more or less the same in general and vocational education.
Put briefly, the explanatory power of the theoretical models is significantly greater in vocational education (70%) than in general education (14%).
Summary and Discussion
In this study—to clarify differences in the performance and functioning of students in secondary education—we have concentrated on the role of configurations of integrated management in realizing an effective school. Following on from Mintzberg (1979, 1989) we adopted various integration mechanisms for the promotion of effective integrated school management: (a) integration and a mutual adjusted influence structure, (b) educational supervision, (c) standardization of work processes and output, and (d) standardization by means of skills and norms.
Differential Effects in General and Vocational Education
Both the degree of monitoring of teacher performance and the extent to which there is contact between parents and school are positively correlated with the math performances of vocational students. This was not the case for general education students. We also see differences between general education students and vocational education students in another aspect. Vocational students obtain higher marks in mathematics in schools where the school management places great emphasis on the contribution of their school to the performance of the students. High expectations from school, good contacts between parents and school, and parental involvement with school are particularly beneficial for the performance of vocational students. Those students in general education seem to benefit less from such an effective management context. Furthermore, it is known that less performing students benefit more than high performers from a clear and well-structured learning environment (Kirschner, Sweller, & Clark, 2006). In terms of instruction, the importance of a direct teaching approach for less performing students and of a flexible teaching approach for higher performing students is mentioned. This line of thought could also be true for the role of the teacher in general and vocational schools. In the vocational school with a homogenous group of less performing students, the faculty must then fulfill a heavily regulatory role strongly directed toward instruction in the class and the skills of the teachers to be able to teach this type of pupil (Muijs, Campbell, Kyriakydes, & Robinson, 2005).
Smart School Management?
The question was whether the three configurations of school and departmental management characteristics could bring about additional effects on the cognitive functioning of students. The outcomes of the analyses for math performances confirm this thought. In fact, we can state that in general education there is a reinforcement premise. The math performances of general education students undergo an extra positive effect from a strong and integrated school and faculty management in this type of school. Furthermore, we established an additional negative effect of the second management cluster in general education: the faculty-based management focusing on skills and instruction but lacking consensus. That this type of management configuration does not function well in general schools can possibly be explained by the fact that too much faculty-based management in this type of school with three types of general education (lower/higher/pre-university) too often leads to loosely coupled schools (Weick, 1976). In particular, the lack of consensus in the management configuration could be strongly disadvantageous for the math performances of the students.
We conclude that what may be termed an effective management context is partly dependent on the type of student (general vs. vocational). This therefore confirms the validity of the contingency theoretical approach to the social context of learning. Effective management is not “one best style” for all schools. What is the “best style” depends on certain characteristics of secondary schools (e.g., Mintzberg, 1979). Management styles are contextually sensitive to influences from the community around the school and differences in the student populations, such as students attending vocational schools.
Implications for Internationalization of Education
We will close now with some reflections on the implications of our findings about school management and leadership for other countries. The important question that should be answered is “How universal or context specific are the results of our study?” We expect there to be as much similarities as well as differences in the management of schools across cultures. We believe that many of the assumptions that underlie effective management in secondary schools are universal indeed. The multifactorial or configuration assumption will be valid for many educational contexts. There is no logical reason to assume fundamental differences between different cultural and educational contexts in the fact that combinations of factors will have an impact on school effectiveness instead of isolated factors. Although education is a complex process everywhere, we agree with Hattie (2009), who points at the importance of the teacher as a key player: “Teachers who account for 30% of the variance. It is what teachers know, do and care about which is very powerful in this learning equation” (p. 3). Hattie presents an overview of the effect sizes of various variables that are related to student achievement (learning gains). In fact, 8 out of 10 most important (effect sizes from Cohen’s d: 0.50−1.13) factors are situated at the level of the teacher (feedback, instructional quality, direct instruction, remediation/feedback, class environment, challenge of goals, peer tutoring, mastery learning).
A second issue concerns that of the context sensitivity of indicators of effective management practices. We think it is possible to define indicators that are empirically tested robust to context variation. Brophy (2000) provides two arguments for this: First, schooling (or the management of schooling) is much more similar than different across countries and cultures, and second, the principles of effectiveness refer to generic aspects of schooling that cut across grade levels, school subjects, and particular curriculum content (Reynolds et al., 2002).
Several studies on education management suggest that certain types of leadership are associated with effective schools (Alton-Lee, 2007; Geijsel et al., 2003; Hargreaves & Fink, 2006; Leithwood, Aitken, & Jantzi, 2001; Timperley, Wilson, Barrar, & Fung, 2007). Hattie (2009), for example, concluded based on an extensive meta−meta-analysis positive effects on student outcomes for more instructional and purposeful leadership, compared with transformational leadership; the latter effects are more indirectly through satisfaction and teacher outcomes.
However, in the setting of many developing countries, principals apparently function as the lower link in an organizational chain that extends from the school through local educational boards, district supervisors, to central staff. Our study shows a positive impact of school management on school effectiveness if the influence of members of the school community on management decisions is relatively high. This finding indicates that the responsiveness of management to the educational knowledge of staff and other parties involved in the school life, such as parents, is crucial. It would be worthwhile to invest time and effort to make principals, department heads, and teachers more aware of the importance of their contribution to decision making, school policy, and the governance of their school. Good relationships between all school parties, like governors, principals, teachers, and parents, are fundamental to effective decision making and governance of schools. The specific knowledge of school management and teachers concerning the local community surrounding their school, their pupils’ home environment, the students’ in- and out-of-school behavior, and peer group should be taken more into account in establishing a fitting school policy and effective school-based governing of schools in general. Although not every country will hold the same opportunities to develop such a local responsiveness of their school management, still, emphasizing the opportunities of school-based management could indirectly lead to a more collectively shared educational and effective school management.
Nonetheless, we have to keep in mind that there are culturally grounded differences in teachers’ responses to certain types of management styles (Hallinger & Kantamara, 2000; Kantamara et al., 2006; R. H. Hofman et al., 2008). Just as we found that effective management styles can differ between academic and vocational education, different types of management strategies will be culturally and context dependent and need adaptation to the type of students and teachers within a secondary school. These adaptations should concern adjustments that derive from the cultural values and norms that shape the behavior of individuals within schools, students as well as teachers. It demands skilful leadership to understand how to take advantage of the cultural strengths within a (local) community (Hofstede, 1991).
Finally, we would like to end this article in line with Hallinger and Kantamara (2000), who prefer to view cultural characteristics as
two-sided coins on which strengths may become weaknesses and weaknesses may become strengths. When one adopts this perspective, a richer field of vision emerges on the domain of leadership. It should stimulate scholars in the industrialised West to look more deeply at their own conceptual models. If so, they will see—perhaps for the first time—the cultural background on which their theories exist. This will open up the possibility of richer and more broadly applicable theoretical development. (p. 16)
Footnotes
Appendix
Management Configurations (Cluster z Scores)
| Cluster 1 | Cluster 2 | Cluster 3 | |
|---|---|---|---|
| n = 35 General; n = 32 Vocational; 55% | n = 21 General; n = 15 Vocational; 30% | n = 9 General; n = 9 Vocational; 15% | |
| Coordination through mutual influence structure | |||
| S influence school principals | .01 | −.24 | 0.55 |
| S influence on boards’ decisions | −.01 | −.28 | 0.73 |
| S contact with parents | −.03 | −.37 | 1.00 |
| F influence faculty | −.25 | .12 | 0.78 |
| Coordination through educational supervision | |||
| S evaluation policy | .13 | −.31 | 0.20 |
| S achievement orientation | .32 | −.55 | −0.02 |
| F emphasis on objectives | −.05 | −.31 | 0.96 |
| F instruction arrangements | −.28 | .32 | 0.37 |
| Standardization through work processes and output | |||
| S rules and orderly climate | .18 | −.59 | 0.66 |
| S policy cancellation lessons | −.17 | .02 | 0.68 |
| F cancellation lessons | −.22 | .56 | −0.45 |
| F arrangements on content, homework | −.23 | −.03 | 1.00 |
| F teachers teamworking | −.27 | .15 | 0.78 |
| Standardization through skills and consensus | |||
| S help starting teachers | −.27 | .14 | 0.80 |
| S monitor skills | −.46 | .60 | 0.47 |
| S consensus | .28 | −.70 | 0.52 |
| F consensus | −.05 | −.22 | 0.73 |
The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
The authors received no financial support for the research and/or authorship of this article.
1.
In the case of vocational education, school type has not been included as a variable in the analyses because vocational education is understood to be a type of education at one level.