School Autonomy Policies and the Changing Governance of Schooling

Since the mid-1990, but in particular since the so-called PISA shock, German-speaking countries have seen major changes in the governance of their school systems. The author of this paper aims to explain some conceptual tools for analyzing governance changes and to give examples of research into governance changes using these concepts. He begins by clarifying the special conceptual view on education systems and on educational reforms provided by the governance perspective, then continues by giving an overview of major educational reforms during the last two decades in German-speaking school systems and explains how they are analyzed and interpreted with the help of governance concepts. He concludes by using the Austrian school autonomy policy, which allowed schools to develop specific “curricular profiles,” as an example for discussing governance changes in a multilevel governance system.

term of governance that has become known as the governance perspective. The aim is to provide "a general analytical framework for studying all kinds of coordination problems among actors" (de Boer, Enders, & Schimank, 2007, p. 138): The coordination between actors produces a "structure of regulation," a social order that regulates interaction and actors' contributions in a given field. This "structure of regulation" is intertwined with a "structure of performance" by which the systemspecific performance is generated. Both social order and performance in a given field are conceived as arising from the coordination of the independent actions of social actors (see Benz, 2004, p. 17).
Starting a decade ago, this conceptual framework has also been introduced into educational research in order to study the changes in the regulation of school systems that German-speaking countries (Germany, Austria, German-speaking cantons of Switzerland, Liechtenstein) have seen since the early 1990s (see Altrichter, Brüsemeister, & Wissinger, 2007). Under the name governance perspective, governance research, or governance studies, a body of work has evolved whose authors aim to understand these changes by concentrating on the question of how regulation and performance of school systems is achieved, sustained, and transformed under the perspective of action coordination between various social actors in complex multilevel systems (see Altrichter & Heinrich, 2007;Kussau & Brüsemeister, 2007;Schimank, 2007).
This seemingly abstract definition-coordination of actors-invites one to spell out what exactly is happening when social processes are governed, regulated, or steered and to study empirically who is contributing what to a system that appears to be governed or coordinated in a specific way (Altrichter, 2010a). The traditional view of system governance assumes a dominant actor, the government, who by the help of its administrative staffs and specific instruments such as legal norms, differential financing, and bureaucratic and executive powers can govern the operative actors, can make the actors at the bottom levels of a system act in a specific way in order to produce the system-specific performance.

Multitude of Actors
The term governance firstly indicates that the assumption that school systems-and their reform-are not necessarily shaped by a single dominant actor, such as the government and its administrative staff. More actors are involved in a system's formation, maintenance, and change. Although proponents of the governance perspective strongly argue that many actors have some influence on the steering of a system, this does not mean that they usually have equal chances of participation and support (see Altrichter & Salzgeber, 2000). Nor does it signify that the central state can no longer be an important actor. However, it does allow researchers to attend to situations in which the state itself is regulated (e.g., by supranational bodies, such as the European Community) or enters negotiation relationships with actors who are in principle under their jurisdiction (e.g., firms, foundations).

Coordination of Action
A central and crucial concept is coordination. Researchers tend to consider something to be regulated or governed if the relevant system actors coordinate their action. The governance perspective uses a nonevaluative concept of coordination to analyze the method and functionality of the actors' combined action (see Altrichter, 2010a;Lange & Schimank, 2004 for analytical instruments).
A major analytic strategy of governance studies is to establish those governance mechanisms that are characteristic of and explanatory for a social system at a specific point in time and space. Different conceptual instruments are used for this purpose (e.g., Benz, Lütz, Schimank, & Simonis, 2007;Dupriez & Maroy, 2003;Lange & Schimank, 2004). For example, de Boer et al. (2007, p. 138) have claimed that at least five specific dimensions may be used to trace characteristic changes during the contemporary transformation of the European higher education systems. These dimensions are as follows (and are organized as governance equalizer in Fig. 4.1): • The dimension state regulation denotes the traditional regulation of public systems by the top-down state authority using legal measures, directives, and distribution of earmarked resources aiming to prescribe in detail the behavior of subsystems.
• The dimension academic self-governance refers to the professionals' power in decision-making, for example, "institutionalized in collegial decision-making within universities and the peer review-based self-steering of academic communities" (de Boer et al., 2007, p. 139). • The dimension external guidance by the state or other stakeholders describes regulatory activities that direct systems and institutions through goal setting, advice, and evaluation usually exerted by the government or other stakeholders. • The dimension managerial self-governance refers to the regulatory power of the internal hierarchies in organizations (such as schools, universities, or hospitals) and to their leadership's power in internal goal setting, distribution of funds, and decision-making. • The dimension competition for scarce resources (such as money, personnel, and prestige) refers to system coordination through market or "quasimarkets" processes.
This rationale was used to analyze both specific changes in university systems (see e.g., Schiene & Schimank, 2007) and more general transformations and differences between European university systems. Figure 4.1 summarizes the findings of de Boer et al. (2007, p. 140) with respect to changes in university governance in England, the Netherlands, Austria, and Germany over the last 20 years: All countries have undergone changes on all five dimensions, yet the degree of change varies between countries and dimensions. The most common feature seems to be that academic self-governance is the main loser, while external guidance by performance targets, the powers of managerial self-governance, and competition between the actors of the university systems have increased in all countries studied.
Schimank's governance equalizer obviously has some heuristic value for analyzing the transformation within systems and differences between systems. In this study, I also found this approach useful for analyzing changes in school systems over time (see Altrichter, Brüsemeister, & Heinrich, 2005) and in the working conditions of different actors in the school system (see e.g., Altrichter, 2010b).

Agency and Structure
It is not the erratic or accidental actions that are interesting for governance analysis, but the structured and structuring actions that contribute to the (relative) sustainability of system coordination. The capability to act in social systems is based on structural elements, on a structure of regulation that organizes actors' rights and competences in a way that is specific to the particular system (see Braun, 2001, p. 247;Kussau & Brüsemeister, 2007, p. 21). Thus, conductors of governance analyses are looking for rules and resources (see Giddens, 1992) that are already existent in a system and how they are used (or not used) and transformed by action.
In order to promote and implement change, promoters of a reform must offer-in part-new norms and resources and must stimulate relevant actors to take them up. If the norms of a reform fit to the motives and values of relevant actors, it will be easier to establish relatively stable constellations. The same holds true for reforms that build on resources that are already available and usable by actors. If this is not the case, reformers must invest in developing the values and resources of relevant actors.
Turning to the performance standard policy as an example of innovation (see Altrichter, Rürup, & Schuchart, 2016): A reform is more likely to take roots in schools if teachers and other relevant actors are committed to boosting students' performance and if they consider competence-based teaching an appropriate and feasible strategy for doing so. Appropriate resources are of help, that is, they know how to practice competence-based teaching, relevant teaching material is available, and so forth. When they receive data feedback about the performance of their classes in standard-related assessment, they must know how to read and interpret it, and if the performance data indicate problems they must also know how to do better-for example, have alternative teaching strategies in their repertoire. Above all, they must be motivated enough to change their teaching.
If these elements are not present, if the actors are not ready to perform the innovation that should be standard in innovatory times, then additional measures are taken to (gradually) close the gap between existing rules and resources and those rules and resources the innovation necessitates. Staff meetings, for example, are held to explain the reform to teachers and motivate them to implement it. Laws and guidelines that pronounce good practice are changed. Performance contracts are introduced to bind the administration and the schools more closely together, or to bind school leaders and individual teachers more closely to the new tasks. Professional development is offered to build up competencies; teaching material is developed and distributed to schools to provide examples of innovative teaching.

Multilevel Systems
Another characteristic of the governance perspective is that complex social systems such as the school system are considered to be multilevel phenomena. This notion points to the fact that not all actors interact with all other actors in the same way. Instead, there are typical constellations of actors, typical levels with special principles of action, that may differ widely from the logic of action on another level.
The concept of multilevel systems draws our attention to questions of crossborder coordination between system levels that appear to be among the most crucial problems of system development. The plans and blueprints for a governance reform (produced and propagated by politicians, the administrative top levels, and some social scientists) are not the whole reform. They are above all structural offers-in part, new rules and resources (see Giddens, 1992)-that are inserted into the transactions of a school system. They must be taken up by actors on various levels of the system, and they must also be translated and redesigned for the specific context, in order to have a chance of acquiring social relevance. The potential effect of these structural offers (and whether or not the propagated effects materialize) can only be determined through their use and through the way various actors (such as teachers, students, school leaders, inspectorate, parents, and textbook producers) adapt and transform their actions and arrive at a new pattern of coordination.
Taking up these structural offers is more than merely following prescribed action programs or implementing given structures (see Ball, Maguire, & Braun, 2012); it necessarily entails constructive and productive features. Actors must make these structural offers more concrete-they must develop them in view of the specific logic of action and of the work conditions of their particular level and translate them into feasible versions of action. Fend (2006) has developed the concept of recontextualization to account for these processes. He considers it important to adequately describe how actors "act together" within the education system (Fend, 2006, p. 174).
Acting on one level of a multi-level system implies that the superordinate level is present for the lower level as a context which, however, will be reinterpreted in view of the context conditions and action resources of this level and will be transformed for practical action. The superordinate level in this way is preserved, but, at the same time, transformed. (Fend, 2006, p. 181 1 ) It is obvious that the multilevel structure is also an issue for my example, the implementation of performance standard policy. Students may find it difficult to make sense of performance standards for their learning. Equally, teachers in Germanspeaking school systems may have difficulties in interpreting performance feedback as didactic cues and reacting with alternative teaching strategies (see Altrichter et al., 2016;Maier & Kuper, 2012).
It has been argued that governance researchers must not limit themselves to the systemic and organizational questions on macro-and mesolevels before classroom learning happens. The central concept of action coordination is also relevant for the microlevel. Classroom teaching and learning may also be understood as a coordinative effort that contributes to the specific performance of a multilevel system: A number of learners and teachers must coordinate their individual actions in such a way that individual and social functions are fulfilled.

Modernization Policies
Governance studies are interested in analyzing In this section, I will turn to recent modernization policies in German-speaking school systems and propose an interpretation of important changes in their coordination modes on a macrolevel. In the third section, I will attempt to analyze how some of these changes are enacted on the mesolevel of the individual school and of the microlevel of staff and classroom interaction.
For the time being, I will concentrate on the German-speaking school systems. Of course, most concepts used in this section are also to be found in other countries of Europe and the Northern hemisphere, as well as quite a number in the global South (see UNESCO, 2016). Many of these policies are obviously "traveling" across national borders. However, Ozga and Jones (2006) have warned that they may acquire a new meaning when they are embedded into different cultural contexts.

Phase 0: Dual Regulation
After post-World War II reconstruction and the investments in education of the 1960s and 1970s, the 1980s were a time of stagnation in many European states, a "strangely motionless time with respect to education policy," as Fend (2006, p. 225) observed for the German-speaking school systems. Some demands were made for new education styles and more autonomy for individual schools and teachers, but the majority of schools and schools systems worked according to the traditional governance mode of dual regulation (Brüsemeister, 2004;Maroy, 2009, p. 72): On the one hand, regulation is based on a state-led administrative hierarchy with general bureaucratic rules; on the other hand, the teaching profession has considerable individual and group-related autonomy when it comes to implementing these rules. Figure 4.2 uses the governance equalizer proposed by Schimank (2007), which was already introduced in section "Analyzing changes in governance" to characterize the governance regime of this phase.  have claimed that there were three waves of reforms that changed the modes of governing German-speaking school systems. These waves of innovation were not exclusive in the sense that the later reforms pushed aside the previous ones; rather, they are rather to be conceived as layers: The new reform wave was placed over the older one, thereby not negating its arguments but merely pushing it somewhat into the background and making new principles more dominant, while still upholding older arguments where it seemed appropriate.

Phase 1: School Autonomy
The first of these major reform ideas was school autonomy. School autonomy policies aim to expand the room for maneuver on the level of individual schools, but also their responsibility for results and development. They do this through decentralization (i.e., redistributing decision-making rights from superordinate administrative levels down the hierarchy to individual schools) or deregulation (i.e., doing away with regulations or making them less detailed). Rights to autonomous decision-making may be granted (or not granted) in various fields, such as the budget, personnel, organization, and educational decisions. The general aims of these policies are to strengthen the "quality and effectivity of education in schools" and the "responsiveness to local needs" (OECD, 2008, p. 524).
Autonomy policies can be found in virtually all developed countries (Blanchenay, Burns, & Köster, 2014;Eurydice, 2007Eurydice, , 2008; however, their actual content and their impact on system governance vary widely. In 1993, Austria passed school autonomy legislation to open up a room for maneuver for in-school decision-making, particularly with respect to curricular matters. This policy allowed schools to develop specific in-school curricula as the basis for so-called Schulprofile (school profiles).
In terms of Schimank's governance equalizer, the situation may be described as follows (see Fig. 4.3): The state decreases the coordination mechanism input regulation through its autonomy legislation. This should enhance competition between schools (although no explicit policies were recorded with respect to this topic). These moves also put some pressure on teachers (who have to invest more time and energy in coordination in order to fulfil the promises of the school profile). It also puts pressure on headpersons who have to orchestrate some coordinated development, although they were not provided with new instruments for leading and managing the school.

Phase 2: School-Based Management
The question of system governance was explicitly raised in the second half of the 1990s. Concepts like school programs, self-evaluation, quality management, new ways of school inspection, coordination of classroom work through sample exercises (i.e., Aufgabenbeispiele), and "parallel tests" (Vergleichsarbeiten; i.e., tests using identical items to compare the performance of different classes) became more prominent. Still, the idea of teachers self-evaluating their schools remained central, with the qualification and loyalty of teachers still seen as prerequisites for the productive development of schools. However, administrations increasingly called for complementary measures: School self-evaluation was to be challenged and checked by external demands (e.g., performance contracts with schools, school programs) and external evaluations (see Holtappels, 2004). These measures were, on the one hand, to provide instruments for in-school management and leadership, for the internal government of schools (see BMUK, 1998). On the other hand, school administration itself began to look for levers to "orchestrate variety" (see EDK, 2000), which had been produced by the policy of school autonomy. Reforms were not communicated as a departure from the previous strategy of school autonomy, nor as a step back to the old centralist models of regulation, but as some complement that should-for the sake of the coherence of the system-provide top and intermediary levels of the school system with new options for control and intervention.
In terms of my analysis of changing governance modes (see Fig. 4.4), I see attempts to increase in-school management and leadership as the main issue in this phase. They are accompanied by early experiments with externally formulated goals. Strengthening the powers of the management should also put some pressure on individual teachers' autonomy.

Phase 3: Evidence-Based Governance
A new round of changes was triggered by the results of the international large-scale assessment studies TIMSS and PISA, which were not flattering for the education systems of the German-speaking countries (see Baumert, 1998;OECD, 2001). This PISA shock paved the way for more and more powerful systemic instruments of external governance of schools. Performance standards (Bildungsstandards) and centrally administered standard-related performance assessment were to form the basis for a more sophisticated output-oriented system governance. Additionally, experts interpreted PISA and TIMSS results as indicating a growing need for classroom development; teaching strategies for stimulating more thorough understanding and for dealing with heterogeneity in a more sophisticated manner were required. These reforms used normative arguments, operational arrangements, and instruments from two major sources.
The first is New Public Management, which may be characterized by the following concepts (Rhodes, 1991): It argued in favor of shifting the focus of governance and control from input to output and aimed to measure the performance of public institutions by controlling their results (Maag Merki, 2016). Secondly, the creators of NPM wanted to make public institutions more responsive to their stakeholders' concerns and to increase their customer orientation. In education, this is reflected by more choices, improved information, and increased reporting to parents, but less often by participation of students and parents in in-school decision-making . Thirdly, comparative testing of schools and publication of the results were intended to reinforce efficiency pressure. Increasing options for school choice and competition for student numbers were to work into this direction, too (Altrichter, Bacher, Beham, Nagy, & Wetzelhütter, 2011).
The second source is the development of sophisticated instruments for comparative performance testing. On the one hand, studies such as PISA and TIMSS provided an advanced technology (and more people able to handle it) for comparative testing; on the other hand, these studies and their public recognition produced international pressure to improve the performance of national education systems in an era of globalization.
The main thrust of the reform lies in formulating system-wide substantial goals and in controlling performance according to these goals (see Fig. 4.5). This, however, should also exert some pressure on the teaching staff, who was confronted for the first time with externally formulated and measured performance goals for students that could, however, also be used to measure the performance of the teachers themselves. The policy also affected the headpersons' discretion in postulating goals for development, which were now externally regulated. On the other hand, external performance standards might also give some lever to the school management to stimulate development in fields that may previously have encountered more teacher opposition. Performance-related information may eventually strengthen parents in their will to discover the best schools for their children and, thus, fuel competition between schools.
The dominant ideas of this third wave of reform are epitomized by a so-called evidence-based model of educational governance, which may be characterized by the following features (Altrichter & Maag Merki, 2016): (i). The evidence-based governance model sets expectations and goals for the performance of the education system and communicates them more clearly than before, for example, through formulating measurable performance standards or developing quality frameworks for school inspections. (ii). Accountability for goal fulfilment: Evaluation and accountability are considered to be key issues in ensuring quality provision for all. Evaluation instruments are to produce evidence as to whether or not expectations have been met by the practical operation of the system units. (iii). Evidence is fed back to all system levels in order to stimulate and orientate system improvement. Actors on all levels of the system-politicians, administrators, school leaders, teachers, students, and so forth-are supposed to use evaluation information to make more rational choices in developing their contribution to the education system and improving their performance. (iv). Involve stakeholders and the wider public: In many cases, evaluation results are not only communicated to the professionals in schools, but also to the schools' stakeholders and even to the wider public through the media. This reflects the idea that schools will be more responsive to developmental needs if they are directly accountable to their constituencies. (v). Link different levels of the system: The idea that cycles of goal formulation, evaluation, and feedback will dynamize improvement is implemented on all (or most) levels of the system: Regions, and in some cases central ministries, are subject to similar instruments of performance management; for example, the results of standard testing are communicated in personalized reports to different system levels. Instruments such as contract management between schools and regional officers, regional officers, and central authorities are used to link information flow and loyalty between system levels. (vi). Align and improve support systems: Finally, existing support systems must be aligned with the governance models and new support instruments must be developed (e.g., developing teaching material for competence-based teaching and diagnostic tests by which teachers can prepare their classes for comparative testing).
Recently, some countries seem to be streamlining their evidence-based accountability systems. Most notably, some countries have introduced "proportional inspections" in order to focus inspection resources to those schools most in need of evaluation and development with well-functioning schools receiving less, or less frequent, external attention (see e.g., Ehren, n.d.;Scheerens, Ehren, Sleegers, & de Leeuw, 2012). These developments may be triggered by financial restrictions, by disappointing research findings, or by the will to find a more prominent space for professionalism. Nevertheless, I do not believe that this can be seen as a thoroughly new phase of school modernization, as the idea of evidence-based control remains central to educational policy. The idea may seem to encounter more criticism than before, but there is no new, comprehensive paradigm that can replace that of evidence-based control. In parallel to Thomas Kuhn's (1996) argument, I assume that a governance regime paradigm also does not vanish to leave a vacancy, but must be pushed aside and replaced by an alternative. In my opinion, such an alternative is not in sight, despite all the disillusionment in educational debates. With only minor adaptations, the evidence-based control model remains a major point of orientation for contemporary education policy.

Research on Governance Reforms: Coordination in and Between More Autonomous Schools
The last section of the paper turns to the meso-and microlevel of schools: How are educational reforms taken up and enacted on the level of schools and classrooms? My empirical example is the Austrian school autonomy policy that allowed schools to develop specific curricular profiles (Altrichter, Heinrich, & Soukup-Altrichter, 2014). Data comes from two research projects based on 11 school case studies. The first study was on secondary schools developing a specific curricular profile in ICT. Cases were sampled in the second study to complement the original ICT cases with different types of school profiles (e.g., language, social learning, arts, and inclusion profiles). Data was collected through qualitative interviews and documents sampling. Individual case studies were written and compared in a cross-case analysis (Altrichter, Heinrich, & Soukup-Altrichter, 2011). Emerging results were confronted with findings by studies using other methodologies. Given the data basis the following statements should not be generalized too quickly; they claim to be valid for the case studies of our research and certainly may be taken as hypotheses for further research. Many schools used the Austrian autonomy legislation to develop specific school profiles. These are usually packages of specific curricular elements (characterized by a thematic and/or methodical specialty) with some additional features (such as extracurricular learning opportunities, special aspects of school culture, specific services). Individual schools use this profile to make themselves visible for special target groups of students and parents, and to attract a sufficient number of students in times of decreasing enrollment.
The empirical data revealed that various reasons for developing a specific profile exist. For example, a group of teachers may hold specific educational ideas and jump at the chance to roll out an innovative plan. Too few students may be enrolling in the school. Assignments by the regional administration may induce schools to develop a profile. Or a problem analysis by the school itself may have unearthed challenges that can be met by the new profile. It is not rare for different groups of teachers in the school to contest the profile and the process of its establishment.
Whatever the original reason for the profile may have been, one condition must be met for its continuation and survival: It must be successful, that is, it must attract a sufficiently high number of students to enable the school's administrators to fill the classes and to choose among applicants.
The transitions or gaps in a noncomprehensive school system are "situations of choice" (in Austria, students are channeled to different educational options after primary school at the age of 10 and after lower secondary school at the age of 14): Depending on their market position, students/parents can choose schools or schools can choose students. A successful curricular profile is a major means of boosting the attractiveness of the school and of improving its opportunities for choice.
Attractive schools can use these choice opportunities to select "good students," who are-according to the choice rationales emerging in this study's interviewswell-performing students and students who come from social groups interested in and supportive of education. Success in this area further increases the school's attractiveness: Most teachers like to teach "good students;" most parents prefer to send their children to schools whose clientele is unlikely to include "difficult students." In a systemic perspective, it may be said that coordination by competition is becoming more important in the school system. This is an (unintended or 'transintentional') result of this specific enactment of the autonomy policies, although no relevant political force in Austria had explicitly argued for more competition in the school system.
These new profiles not only increase the diversity of choices but they are in a hierarchy of attractiveness. As the Austrian lower secondary system is bipartite (i.e., distinguishing between high-status academic Gymnasiums (AHS) and low-status Neue Mittelschulen), the choice did not previously take place on an equal level, but favored the Gymnasium. However, autonomous school profiles produced a new and additional hierarchy within the two systems, and as a consequence new selection processes as well as new opportunities of success and failure, because schools are allowed to choose between students if there are too many applications (Altrichter et al., 2014).
But what happens to the "many unchosen" who are excluded from the attractive profiles? Researchers comparing "profile classes" with "leftover classes" that cater to those who are not chosen (or are not choosing) regularly show that these latter classes usually contain more boys, more immigrant children, and more low performing students. In short: Students with problematic school careers and who are most in need of support are most likely to be in "leftover classes" or "leftover schools" (see Specht, 2011).
But it is more than that. A study by Ferdinand Eder (2011) indicates that more lessons are cancelled in "nonprofile leftover classes" and that qualified replacement of these lessons takes place far less often. Thus, there is some indication these "leftover classes" or "leftover schools" that emerge in the course of developing attractive curricular profiles are, on average, characterized by less careful teaching and assistance, affecting exactly those students who might be in particular need of support.
What do these autonomy changes mean for the traditional governance regime of the Austrian school system, which has been characterized as a hierarchicalprofessional dual regulation (Brüsemeister, 2004; see the black bars in Fig. 4.6)? A school reform inspired by New Public Management is indicated by grey bars in the same figure (see de Boer et al., 2007, p. 149): less state regulation and teacher autonomy, and more institutional management, competition, and performance control.
The Austrian policymakers pushing for school autonomy explicitly attempted to decrease "state input regulation" and to strengthen "institutional leadership and management" while remaining comparatively silent with respect to other dimensions. However, the concrete processes of developing school profiles have affected more coordination principles (see the hatched bars in Fig. 4.6). This is particularly true for "competition," which (although present in earlier coordination relationships; Zymek, 2010) has been upgraded to a more important coordination principle between schools. "Institutional leadership and management" have changed far less. I judge that the case studies show that demands on in-school management have increased without being accompanied by structural changes with respect to school leadership. In-school management still needs skillful leaders and sensitive attention to the traditional coordinative principles in the teaching community. The new instruments of evidence-based governance (which are indicated by the coordination principle "external control through performance goals" in Fig. 4.6) are not yet relevant in the schools included in this study, although the ministry invested strongly in propagating school programs and quality management. This situation may change after the performance standards and standardrelated tests, which were only in their initial stages at the time of the data collection, are broadly implemented.

Conclusion
During the last two decades, many European school systems have undergone major changes in the way they are governed. In its first section, this chapter proposes some conceptual tools for making sense of governance changes. These concepts are used in the second section to analyze changes in the governance of German-speaking school systems. It is suggested that three waves of modernization policies may be distinguished: A first wave introduced some degree of "school autonomy;" a second one emphasized leadership and in-school management; while in a third phase, instruments for an "evidence-based governance" of schooling were gradually built up. While section "Modernization policies" argued on the macro level of policies, the third section turns to the meso and micro level of schooling to study how reform policies are taken up, appropriated, and transformed in schools and classrooms. Data derives from a qualitative empirical study on the implementation of school autonomy policies in Austria (Altrichter et al., 2014).
The changes I have described in this paper have not fully overturned the traditional governance regime of hierarchical-professional dual regulation. Rather, a hybrid coordination constellation has been made even more hybrid (Dupriez & Maroy, 2003) by new accents. If one dares to convert these qualitative trends into the logic of Fig. 4.6, the hatched bars result. They indicate that the changes point to the direction postulated by New Public Management. However, old coordination principles have not been fully replaced, but persist, in a somewhat "weakened" state, side by side with relatively "new" (or newly emphasized) coordination principles that have been promoted by the waves of modernization. The result is a "new unclarity" that is typical for situations of transformation, and which is experienced by actors as a growing number of reference points to be attended to in their daily actions and decisions.

Note
The paper follows the presentation the author gave in the Heidelberg symposium "Geographies of Schooling" on September 14, 2016. It draws on the author's previous work, in particular Altrichter (2010a), Altrichter and Heinrich (2007), and Altrichter et al. (2014).
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.