Making policy in the classroom

Abstract

The concept of street-level bureaucracy (Lipsky, 1980, 2010) examines the form and extent discretion takes in teachers’ and other public policy enactors’ work and how they negotiate their way through sometimes contradictory policy imperatives. It provides a framework for straddling top-down and bottom-up perspectives on policy making. In this article, I argue that comparative education research should take advantage of the analytical framework this perspective offers. It requires, first, mapping out policies resulting in the characteristics of teachers’ discretion in a particular national or local context and, second, to observe how teachers make use of this discretionary space in their daily work. Lipsky has shown strategies employed by street-level bureaucrats to alleviate workload pressures and how they make policy in this way. Applying street-level bureaucracy in comparative education research illuminates why straightforward policy transfer is problematic and how it can be employed to explore practices around inclusion and exclusion.

Introduction

Education is increasingly treated as a global commodity which can be imported and exported in order to achieve desired outcomes of attainment. The decision about the direction of policy transfer is usually driven by the results of large-scale international surveys. There are several problems with comparison at mainly nation-state level and making top-down policy decisions based on it. As an alternative approach to this form of comparative research I will introduce Michael Lipsky’s (1980, 2010) theory of street-level bureaucracy as a useful tool in comparative education. The strength of a focus on street-level bureaucrats and their discretionary power is that it provides a framework for straddling top-down and bottom-up perspectives on policy making.

In the high-speed world of politics, policy formation and implementation, large-scale comparisons and rankings of educational outcomes, like the OECD flagship Programme of International Student Assessment (PISA) or TIMSS (Trends in International Mathematics and Science Study), can lead to policy epidemics (Levin, 1998; Phillips and Ochs, 2004). The temptation to ‘borrow’ policies that apparently work, indicated by the position of a country in the consulted league table, is great. Global ‘policy speak’ as an outcome of and motivation for policy importing and exporting (Ball, 2013) can be read as evidence of convergence in education. However, a closer look reveals that at times it is just the name of a policy but not the content that has been borrowed and that the policy in transit is mediated by the national and local context including other policies which may not be directly linked to education. The form policy import takes is influenced by the general approach to public policy, the understanding of how policy is made and the nature of the links between aims, policy formulation, implementation and desired outcomes (Hupe et al., 2014).

Top-down approaches in policy analysis are interested in whether stated policy aims lead via policy formulation and implementation to the desired outcomes and if this is not the case, to explore what interferes in this compliance model of policy (see Hill, 2009b, for an overview). Here discretion can be interpreted as dangerous opportunities for practitioners to follow their own, personal goals. The contrasting bottom-up alternative is less constrained by the assumptions about the process itself and focuses on possible actions and choices at the site of the problem (Elmore, 1980). Lipsky’s (1980, 2010) work includes both the top-down and bottom-up perspectives and, with that, highlights the nature of policy as it is made at the ‘coal face’. The focus on discretion demands attention to the rules or policies within which the discretion of workers in public services is invited or taken into account. Apart from the inevitability of discretion in human encounters, the employment of discretion on the part of the street-level bureaucrat may influence the meaningfulness of the service in the eyes of the client (Tummers and Bekkers, 2014) and in this way shapes socio-cultural perceptions of public policy.

Street-level bureaucracy

Michael Lipsky’s (1980) perspective of street-level bureaucracy highlights the importance of discretion across occupations which encounter citizens in their role as public service workers. Street-level bureaucrats work in a context of ambiguous or vague goals with an abundance of rules which are at times contradictory and allow more than one way of action or inaction (Lipsky, 1980, 2010; Maynard-Moody and Musheno, 2003). Discretion acts as a lubricant in the complex public policy machine (Evans, 2011). Street-level bureaucrats do have considerable discretion because their work entails making decisions about their clients where the individual requirements only emerge during the contact with the client and may therefore be negotiated and because the consequent decisions do affect these people’s life chances (Lipsky, 1980, 2010).

Demand for public services tends to increase and, even when service levels are extended, eventually demand exceeds supply in quantitative or qualitative terms, or both. The implication is that street-level bureaucrats perform their work with limited or even inadequate resources (Lipsky, 1980). Specific services do target specific clients, but it is during the encounter that street-level bureaucrats make decisions on whether they are dealing with a deserving client. Maynard-Moody and Musheno (2003) emphasise that this is the initial instance of a discretionary power that street-level bureaucrats possess. As the relationship between street-level bureaucrat and client unfolds, the question of worthiness will be revisited and used to establish the street-level bureaucrat’s power.

Street-level bureaucrats are presented with the dilemma arising from the paradox of treating all citizens alike, but being responsive to individual cases. Professionals like teachers, social workers, police officers ‘interact with citizens in the course of the job and have discretion in exercising authority; in addition, they cannot do the job according to ideal conceptions of the practice because of the limitations of the work structure’ (Lipsky, 2010: xvii). Under certain circumstances, these professionals working with people develop specific strategies to cope with pressure and narrow the gap between their ideal of their role and what they feel able to provide. These strategies include limiting demand, e.g. by routines, better utilisation of resources and availability, modifying the concepts of the job and modifying the concept of the client. Initially, Lipsky (1980) had used the term street-level bureaucrats for those whose professional traditions and locally established work practices had emerged from strategies addressing the dilemma arising from indeterminacy, little control, limited resources and the potential for an expanding workload. The concept of street-level bureaucrat was received and at times misrepresented as meaning all teachers, police officers, social workers and so on in public services as pointing to their misery and work overload (Lipsky, 2010). Nevertheless, many of these workers are to some extent happy with their work and like what they do.

Street-level bureaucracy has been adopted by a number of researchers to explore the strategies undertaken by specific professional groups in specific circumstances, covering a wide range of areas of public policy (see overviews in Brodkin, 2012; Hupe, 2013; Rønning, 2011; Rowe, 2012) and allows an exploration of ‘change in organisational practice, discretion of frontline workers and complex decision making in the context of formal policy ambiguity and uncertainty’ (Brodkin, 2003: 151). A focus on one kind of street-level bureaucrat in a range of settings can throw light on distinct forms of discretion rooted in micro-cultures of frontline practice (Ellis, 2011). In this case, these street-level bureaucrats are positioned in a specific set of policy intentions and policy texts at the top level. A focus on discretion in comparative work opens up the potential to relate the behaviour, experiences and habitus of a particular professional group with the policy framework within which they operate. The bottom-up element in this approach allows including a broader range of policies than a top-down perspective and is better suited to include socio-cultural conditions and is therefore more likely to avoid ethnocentric pitfalls ripe in comparative research.

Empirical studies on street-level bureaucrats produced attempts to conceptualise the slippery term ‘discretion’. Discretion is neither good nor bad, but can have beneficial or negative effects for clients, street-level bureaucrats themselves and their organisations. It cannot be assumed that discretion is the requirement to negotiate a decision within or against established rules and conditions, but rather may be the actions taken due to the lack of knowledge of regulating policies (Dorch, 2009) or even of a policy framework that does not exist (Berkman and Plutzer, 2010). Smith (1981 in Hupe, 2013) differentiates between de jure discretion, or the authority to act and de facto discretion, or the power to act, though not necessarily according to the established rules. Another way to conceptualise discretion in relation to rules is to typify discretion as within the rule, discretion amongst the rules, discretion beyond the rules and discretion against rules (Barberis and Buchowicz, 2015). This distinction offers a useful framework for research within one specific policy framework or for case study research drawing on cases from different nation states. Alternative ways of conceptualising discretion more readily offer tools for comparative analysis. Much research is concerned about the squeeze on street-level bureaucrat’s discretion. The theorisation of distinct though interrelated forms of rule discretion, value discretion and task discretion showed that new public management reduced school teachers’ and social workers’ rule-making capacity (Taylor and Kelly, 2006). Tummers and Bekkers (2014) differentiate street-level bureaucrats’ discretionary decision according to the relationship with the client. They may be moving towards the client, moving away from the client or turning against the client. Discretion can be examined as an independent variable, as a condition to examine outcomes for clients or as dependent variable, as taken by professionals in a specific set of conditions (Hupe, 2013).

The examination of strategies developed by teachers as street-level bureaucrats in order to cope with their particular workload allows us to explore and compare the links between policy on various levels and outcomes. These strategies may be observed as they are established, as a response to increased pressures or they may be seen as consolidated in particular practices as a professional habitus at a specific point in time and in a specific context. When this analysis is extended to include cross-national data, it provides a framework for comparative work. In the following section, I will set out the context within which teachers can be conceptualised as street-level bureaucrats followed by a discussion of the strategies used by street-level bureaucrats with particular references to teachers. To demonstrate the potential of street-level bureaucracy as a perspective in comparative education research, I discuss some insights gained from Lipsky’s writing when I explored data as part of the CATE project (Comparing and Analysing Teaching in Northern Europe), observing a range of native language and mathematics lessons in English and German secondary schools and interviewing teachers (Kelly et al., 2013a, 2013b, 2014). The section and subsections of the article are based on the data I generated for this project. The data are two full, audio recorded lesson observations, each followed by an interview with the teacher. My sample comprised two teachers in England and four teachers in Germany. My critical exploration was strengthened by the collaborate analysis of all data sets of this project and extensive working discussions with the academics in the project team.

The schooling of children

The education of children and young people is a central feature of any modern state, usually expressed by compulsory and free basic education, and is called upon to fulfil a number of purposes, many of which come under the umbrella of welfare state provision. Schools offer routes to qualifications and, connected to this, allocate and/or select individuals for future roles. They are the place of socialisation additional to the family, aiming to support social integration and support the acquisition of cultural knowledge and practices. At the structural level, aims range from ensuring economic growth and competitiveness to offering a lever to promote a just society. The social purpose education encompasses is in empowering young people and allowing them to be active citizens now and in the future. From a political and cultural perspective, education is presented as the key to regaining a role in the globalised world for nation states and individuals and it has recently taken on board the ‘Promotion of citizenship and the common values of freedom, tolerance and non-discrimination’ (European Commission, EACEA and Eurydice, 2016). In addition, schools have custodial tasks (Flechsig and Haller, 1975), are administrative institutions (see Fend, 1981, 2008; Herzog, 2011) and can provide other services, like (free) school meals and access to medical services to almost all children of a generation (Hill, 2009a). The prominence each of the purposes takes and how they translate into practice varies between countries, locations and individual schools.

Governments can take on the role of direct providers of schools, funders and regulators of education. Education policy aims to determine how education is funded and the structure of the system including its control; who is allowed to teach and the qualifications of teachers; teachers’ working conditions, including payment of teachers; subsidies for students and/or their families; and, of course, it makes demands on the content of teaching and assessment of pupils. Governments are involved in providing and/or regulating services provided by schools not directly related to education, like after school care, sport and drama clubs, and pastoral care. Teachers and schools may be charged with responsibility for child protection by the government.

From public policy into the classroom

Policy objectives as such are not very useful in understanding what happens in the classroom (Dorf et al., 2013; Hill, 2009b). Education is severely influenced at the point of delivery, because it cannot be separated from its cultural context (Alexander, 2008) and because of the abundance of purposes of education. By the same token, it is difficult to read current policy aims into observed practice. These difficulties of identifying a tight and easily observable fit between policy and practice invite one to explore the policy practice relationship from a different angle. The classroom is the heart of education where activities take place aiming to ‘transform’ children via learning as a necessary process of developing knowledgeable, responsible and autonomous adults.

Teachers in the classroom

Bernstein’s (1996) pedagogic device offers a useful and elaborate theory and analytical tool to link politics and policy with teachers’ behaviour and pupils’ attainment and the various selection processes and transformation of knowledge. Teachers are at the receiving end of the struggles within and between the official recontextualising field or the education policy arena and the pedagogical recontextualising field, the experts of teaching and learning, which translate knowledge into school knowledge and practices of teaching. Depending on their ability to gain access to the various debates around these education issues, and their willingness to engage with their details, they can consciously participate, for example by responding to calls for consultations, political activity or their own academic research, and are by default further interpreters of the struggle. The teacher is the vital link between these recontextualising fields and the reproductive field, the school, or in other words, teachers are the key connection between policy and practice through further transformation of knowledge in classrooms by teachers and students together. This role of teachers has been described in several ways. Croll (1996) calls them the ultimate implementers of education policy. ‘Policy enactment’ captures better the complex process of negotiation and contestation, and the interplay of interpretation and translation. Some of this research concentrates on mid-level managers of schools (Spillane et al., 2002), other mid-level policy actors (Singh et al., 2013) or examines how policy actors account for enactment and meaning making (Ball et al., 2012; Braun et al., 2010). Yet it is in the classroom where policy is made (Lipsky, 1980, 2010) and actualised (Maynard-Moody and Musheno, 2003).

Teachers as street-level bureaucrats

Teachers, like other street-level bureaucrats, work with the public and in this role, represent the frontlines of government policy. Objectives for schooling are related to the variety of purposes of education, a number of other anticipated outcomes (see above) and are further influenced by existing stakeholders. For each of the goals there are several rules that apply, some binding, some negotiable and some discretionary. Rules exist in the form of acts, directives, expectations of professional bodies, institutional rules or agreements between individual teachers, parents and pupils, to name but a few. Some of these are visible and public. Others are embodied in teachers’ practices, related to institutional structures and develop in the context of the discourse on effective teaching.

Teachers have in common with other street-level bureaucrats that their clients are non-voluntary – at least insofar as we define the pupils as clients. A twist in the application of Lipsky’s theory is that teachers work most of the time with a group of pupil clients, so their professional relationship with pupil clients is witnessed by other clients. The potential shift of power towards clients, the greater visibility of discretionary decisions by teachers may be offset by children’s weaker position in society and being perceived as victim, threat or investment (Hendrick, 1994). Maynard-Moody and Musheno (2003) highlight that the decision about the person and whether he or she qualifies as a client precedes policy enactment. Children may have the right to be educated and may be required to attend school, though they may be excluded from educational settings for short periods or indefinitely excluded from a particular school or a particular type of school. The basis for exclusion is the unworthiness of the pupil, due to certain behaviours in relation to themselves, other people and or academic achievement. Decisions that are mainly of value discretion can be disguised as following rules within elements of rule discretion. When there is uncertainty whether the parents are the clients or customers, especially in a quasi-market or private system of education, the analysis gains complexity.

Street-level bureaucrats perform their work with limited or even inadequate resources (Lipsky, 1980, 2010; Maynard-Moody and Musheno, 2003). Education, like other public services, draws on ideas of social justice and equity, protection of citizens and supports the functioning of society. This is the backdrop against which demand will always exceed provided services. For example, the demand for longer compulsory schooling, extended school days, a broader range of subjects covered or smaller classes could easily outstrip available resources. These demands can be triggered by international comparisons. For example, in the aftermath of the PISA shock in Germany, smaller classes, the lowering of the compulsory school age, extension of teaching time, more German lessons, etc. were called for (Klieme et al., 2010; Verbeet and Weinzierl, 2011). On one level, in contrast to other street-level bureaucracy, teaching is fairly predictable at least concerning the number of children requiring schooling. However, preferences for specific school types and parents’ ability to choose schools, the introduction of quasi-markets, perceptions and reactions to special educational needs, a shift of emphasis within the many goals of education or the recent demand of serving the interests of economic competitiveness (Ball, 2013), lead to reforms of the established system and a new allocation of resources. Observing teachers at work shows that there are always exceeding demands on teachers’ attention, expertise and time, despite strategies to efficiently use resources, like various forms of whole class teaching, peer teaching, and flipped classroom.

Street-level bureaucrats enter their profession committed to ideal conceptions of their job and what can be achieved. These ideals may have been part of the motivation of why the profession was chosen as a career and been inculcated during training. Consequently, they contribute to the basis for value discretion, but this may change over time. At the ‘coalface’ of public policy, they encounter the impossibility of fulfilling these aspirations. The reality of their jobs is that they work on segments, have limited control of the outcome of their work, do not control the ‘raw material’ or even the pace of work and are supposed to fulfil indeterminate objectives (Lipsky, 1980, 2010). Teachers teach specific subjects in specific schools; they can support pupils but not control their attainment; they must work with children enrolled as pupils; are bound up by the structures of schooling including curricula, term times, hours of teacher-pupil contact; and work towards changing expectations about outcomes to achieve. Nevertheless, the acceptance of discretion of street-level bureaucrats is important in upholding the belief that these workers hold the key to the well-being of their clients. Only a professional judgement on the condition of the client and his or her situation allows an adequate decision on action and/or further professional engagement of a specific kind. In education, there is a strong emphasis on the relationship between teacher and pupil to promote learning. It requires the knowledge and skills of the professional and refers often to tacit knowledge gained through professional experience. The requirements of being licensed as a teacher usually include placements and assessed work experience.

The dilemma arising from the paradox of treating all citizens alike and, at the same time, being responsive to individual cases (Lipsky, 1980, 2010) as well as the stickiness of policy pathways lead to contradictory expectations and structures. Teachers who were not able to develop a strong professional identity (often labelled as semi-professionals), for example due to the absence of the taught and researched specialism of pedagogy (as in the German Erziehungswissenschaften, the science of education), are in danger of being a fragmented profession and open to ‘modernisation’ (Beck, 2008). For example, in England, the emphasis on individualism in education was well reflected in the ‘golden age of teacher control’ until the mid-1970s (Whitty, 2001) and encompassed occupational professionalism. The ‘discourse of derision’ (Ball, 1990) smoothed the way to a tight prescription of what and how to teach, to controls and a discourse of accountability and, in this way, restricted teachers’ professionalism (Ward and Eden, 2009; Whitty, 2001), turning occupational professionalism into organisational professionalism (Evetts, 2009). Nevertheless, the unlimited curriculum before the 1988 Education Reform Act is as much a myth as the inability of teachers to assert their authority after the introduction of the National Curriculum in England (McCulloch et al., 2000). Despite various means to inculcate a particular version of teaching practice, government documents setting out contemporary understanding of effective education emphasises that children need individual responses appropriate to their learning (see, for example, Standards for teachers in England (Training and Development Agency (TDA), 2007) and their revision (Department for Education (DfE), 2011).

For Taylor (2007), recent reforms of the education system in England have eroded the discretionary power of teachers to such an extent that the characterisation of teachers as street-level bureaucrats is not applicable anymore. It should be noted that the respondents in this research differentiate between the loss of autonomy in making rules within the school and the ability to make decisions in the classroom. In other words, there was a stronger squeeze on rule discretion but less curtailment of task discretion. However, the characteristics of public services and the relationality of education result in limits to the standardisation of the teacher-pupil encounter, ‘Thus, like putty, discretion can be squeezed by oversight and rules, but never eliminated; it will shift and re-emerge in some other form in some other place’ (Maynard-Moody and Musheno, 2003: 10).

Making policy in the classroom

Policies impacting on teachers’ practice may come at a dizzying pace (Ball, 2013) and may introduce tasks not connected to teaching, for example, marketing of the school (Lundström and Parding, 2011) or connected to the extension of time children spend in all-day schools as in Germany (Autorengruppe Bildungsberichterstattung, 2014). The teachers we observed and interviewed as part of the CATE study showed a range of awareness of current policies regulating their practice in the classroom. Some teachers expressed that their interest in the education policy was rooted in the ability to help promote good practice, for example by taking responsibility for teacher trainees. Others explained that only by joining in and taking responsibility for implementing new government initiatives would they have a chance of understanding the translation of policy into practice in a way that was suitable for their school or their personal approach to teaching and learning.

Teachers make policy in the classroom. They work in the context of cultural and social traditions and, in a specific policy framework that regulates which children attend which school and by means of inclusion/exclusion, teachers make policy at classroom level. Teachers are faced with demands expressed by children and parents, as well as expectations from colleagues and their superiors. How they deal with these will be used as an example of how useful the concept of street-level bureaucrat is for comparative enquiry.

Limit demand by maximising available resources and client compliance

Lipsky (1980) identified several ways in which street-level bureaucrats may limit the demands of clients by maximising the utilisation of available resources, developing routines and simplifications and by obtaining clients’ compliance. These strategies are developed within and are influenced by existing structural characteristics. Class size and composition, contact time and staffing levels are regulated by state or local policy and are the outcome of historical processes. These structures impact on the extent of value discretion available to teacher training situated in the specific philosophical approach found in different countries (McLean, 1990) and can provide the means to teach large groups of children efficiently or the expectations to pay attention to the needs of individual children (Alexander, 2000; Luxton and Last, 1998; Meyer, 2004) and lead to certain preferences.

Education policy in Germany organises school attendance in one or more grades (Klasse) of a one year cohort in which children remain throughout their time in the school and in which they are taught in all subjects. If the number of children in a particular grade falls below a specified level, it may be amalgamated with another grade. Teachers prefer to work with small classes and there are several strategies that can be employed to influence numbers. In street-level bureaucracy terms, class size in general (children enrolled) and on a daily basis (children present) is the outcome of decisions on the worthiness of clients or of which child deserves to be taught in a particular grade by a specific teacher. A system that requires pupils to repeat a year if attainment is too low and in which children can move up or down to other kinds of secondary school invites some manipulation, for example by offering special support to individual students to avoid exit due to low attainment and risk resulting in amalgamation.

Similar strategies by teachers of supporting pupils can be found in systems operating in a quasi-market, as in England, where the survival and resourcing of a school depends on maintaining a specific enrolment level. However, here it is unlikely to reduce workload, even in the long run.

There is a fear out there from a teaching point of view and certainly from a school leadership point of view, fear of failure and fear of failure means you do almost anything you can to get the results within the rules. Which means we do take bits of course work through with them. We do break our backs to a certain extent. And this is quite a tension within the leadership group as well – we have regular discussions which are around the idea, do we let them fail. We should let them fail probably more than they do. And we go probably too far in terms of helping them out.

(Teacher, England)

The discretion granted to teachers through assessment leads to specific forms of practice. If tests and other pieces of work are set and marked solely by the teacher, their ability to influence records of attainment is greater than in a system where tests are set nationally and assessment of these pieces of work are undertaken externally. In a system where low attainment can lead to exclusion from the grade and/or school, frequent tests and other forms of assessment become tools to achieve compliance by pupils. Additionally, a focus on components of summative assessment can reveal further means of compulsion and motivation. Are there certain behaviours that lead to less favourable marks which are not related to the knowledge or skills held by pupils? For example, are students marked down who do not bring a specific notebook, cannot show their homework or who show any kind of disruptive behaviour during lessons?

When exclusion from a specific grade does not hold a strong grip on pupils, either because the boundaries around the group of children are weaker due to streaming and teaching in different groups across the year cohort, because the repetition of a year is unusual or comprehensive schools pre-empt movements from one school type to another, other means of achieving pupils’ compliance are dominant. Specific rules justify punishment and other forms of exclusion. I observed special seating arrangements or special tasks for children who had digressed. In one school, a system of regular public monitoring of the behaviour of a pupil was established. Often these possibilities are available to teachers in the form of school rules, and our observations showed that teachers in the classroom chose and bent these rules according to their view of tasks and understanding of pupils’ needs. Rules around school uniforms, chewing gum and mobile phones presented themselves readily for managing pupil compliance.

Redefining the job

Another strategy to narrow the gap between the ideal of teaching and the daily experience is to redefine the job. Observed practice in classrooms reveals how the demand by clients, from content-related questions to personal matters is controlled, and it also shows teachers’ construction of their job. Team teaching may be one way to alleviate work pressure as does making use of a teaching assistant, or excluding children in one form or the other, but it also shows what teachers believe to be their responsibilities. The abundance of goals in combination with teachers’ ‘unsettled occupational identity’ in the context of the national discourse of education impacts on the concept of the work teachers apply. The strength of a professional identity and the nature of granted discretion as perceived by teachers influences whether goal selection is seen as a positive part of practice or a more subversive role of teaching.

There are two areas of day-to-day work in classrooms which are particularly useful in highlighting the concept of the job of teaching and, connected to it, the concept of the pupil deserving to be taught, which is important for street-level bureaucrats as these concepts are embedded and developed in national policies. One area is the use of homework, because it allows the exploration of the boundaries of teachers’ responsibility and the extent to which school systems relate and are linked to the lives of children and their parents. The other area concerns the rules of participation in the form of speaking in the classroom, because it illustrates the relationship between teacher and pupil.

Schools may have homework policies setting out how much time children are expected to spend on homework, when it is set and parents’ involvement. Homework can be set to consolidate what has been learnt during the lesson, prepare the next step or a new topic, or involves some practical work to prepare teaching and learning materials. Research on whether homework contributes to widening or reducing the gap between high and low achievers and which conditions contribute to one or the other is inconclusive (for example Rønning, 2011; Sharp et al., 2001). In Germany, the idea of homework is dominant. Stereotypically, reflecting post-war traditions, homework is what German children do at home in the afternoon, amongst other extra-school activities. When parents are not available or able to support their children, alternatives have to be found. Not surprisingly, the use of private mentors is prolific in Germany (Schneider, 2005).

Teachers’ use of homework showed a wide range of practice. Some teachers set very little homework because they were resigned to the fact that most pupils would not engage with it. In an education system like Germany, which assumes that teaching is most successful if taking place in groups of homogeneous children regarding knowledge and skills, and practices that contribute to differences in learners, it is less useful. During the lessons in Germany that I attended, homework was set as the same task for all children and was often found in the textbook used. It was assumed that in this way weaker learners could catch up and practise the application of content. Advantages of using textbook tasks were: less time was required for preparation, results could be controlled and evaluated as a whole class activity and, in this way, workload was reduced. However, we observed the setting of individually tailored homework, addressing perceived needs of individual pupils. This was more prominent in the English context, linked to the ideal of teaching as tailored engagement with individual pupils. Homework can be used as assessment tasks and can, if feedback is provided, increase the workload of teachers.

An important aspect of the regulatory discourse is how speaking in the classroom is managed. In contrast to other street-level bureaucrats and client relationships, the hierarchical power relationship is underlined by an adult teacher meeting the child pupil, linking what happens in the classroom to the context of social and cultural expectations about the relationship between generations. Pupils are expected to raise their hand when they want to speak. This rule is known by everybody and, at times, used outside the classroom, for example in meetings or, tongue in cheek, during informal conversations to get a word in. Inside the classroom, transgressing this rule is the arena for teachers establishing proper client behaviour. In our observations, the raising of the hand rule was often ignored or forgotten by pupils in their enthusiasm to contribute or desire to shift pedagogical discourse in another direction. Teachers used this rule flexibly for the purpose of controlling the amount of contribution by particular students, to build confidence (by inviting weaker pupils to have a go at an easier question) and to reduce the noise level if desired. However, since the rule does not work in reverse, teachers can ask for contributions even when a pupil did not raise his or her hand. Here the doors are open to encouragement or to instilling fear. Overall, teachers had established what counts as an acceptable noise level to them and what kinds of conversations between children are permitted. In some of the German classrooms children worked together, without being prompted by the teacher, helped each other out with pens or erasers and some dipped in and out of the tasks at hand. We observed fewer instructions on how to work than in England.

All teachers have to handle the tension of, on the one hand, serving all pupils and on the other, supporting individual children. This is solved by finding a balance between the contradictory goals. Solutions depend on the concept of pupil/client developed by teachers. Observations showed that private matters of children were always present in the classroom. I observed conflicts between children, heard children talking about family matters that had an impact on their work and children asking questions that were of a private nature. Many teachers emphasised that pupils must have the right to bring aspects of their lives into the classroom. I also observed how teachers showed interest in certain family situations and prided themselves on making the lesson content relevant by linking it to the everyday experiences of children. Others tried to make a difference between what happens in the classroom during the lesson and what takes place outside the classroom.

I think the personal level is very important. Well, they don’t know me as a person, but as a teacher. They do know that as a teacher I have to be sometimes more strict and that I have to say uncomfortable things, which they do not like. However, they know, too, that during break time they can talk to me about anything in inverted commas. I find it very important that they know this. One is business, the other one is that I respect you as human being. Like the pupil in the first row. She is really very weak. We often argue in maths because she doesn’t do her homework and does not participate. But we still get on well.

(Teacher, Germany)

This teacher draws a clear distinction between the concept of the pupil/client and the concept of the child. Interestingly, some teachers talked about colleagues who emphasise the teacher-pupil relationship and exclude the adult-child dimension. Consequently, they observed less work satisfaction in their colleagues and felt that in this way, the ideal of the teacher is somewhat betrayed. Making the distinction between pupil and child softens the harshness of judgement regarding the worthiness of a client and decisions within the national policy framework. A teacher may not be able to support all pupils to attain grades to a high level, but he or she can offer other kinds of support and recognition to all children in their professional care. Here there are important links to the concept of childhood and how adults and social institutions share responsibilities.

Conclusion

The exploration of education policy by applying Lipsky’s (1980, 2010) framework of street-level bureaucracy to comparative research and analysis has shown the potential of this approach. It straddles the top-down and bottom-up perspectives on policy making and avoids some of the pitfalls of each. Exploring the practice of teachers in the classroom against the backdrop of policies regulating demand, setting out what the job of teachers is and defining clients (pupils and their parents) in relation to teachers is a useful framework for comparative, cross-national research. Some of these aspects are not new and are addressed when comparing countries. However, a specific focus on the means of the ‘squeeze’ on discretion, its effects on teachers’ ability to make decisions in practice and the patterns emerging in local and national contexts offers a new angle of enquiry. It may reveal specific distinctions between ‘authorised’ and ‘unauthorised’ discretion or between compliance and subversion (Brodkin, 2011) exerted by teachers, whether expressed in the policy documents, institutional ethos or as perceived by teachers, and with it remarks on the strength of professional identities.

Street-level bureaucracy allows the exploration of the job as understood and practised by teachers and other educators. Questions worth asking focus on their ideals and what they think is expected of them by the government, their organisation, their managers and colleagues. This may be linked to their experience of training and preparation for their professional role.

A further step in using the street-level bureaucrat as an analytical tool is the relationship of teachers and rules. It includes questions about knowledge of the regulatory framework and about the extent of agency to influence it at the point of formulation. Then their practice can be examined using the typology of discretion within the rule, amongst rules, beyond rules or against rules (Barberis and Buchowicz, 2015). Teachers’ task discretion should then be explored as nestled in the various aspects of rule discretion.

Those working in education, and teachers as street-level bureaucrats in particular, struggle with the definition of the client, which is closely linked to value discretion. Obviously, children as pupils are clients. Yet, many choices are made by parents who then take the characteristics of clients. When the professional relationship of teachers is with this double client, questions of worthiness have a twist and may contribute to the understanding of inequality. What is the definition of a worthy client parent and what is the definition of the worthy client pupil? How is this established in practice? What mechanisms are in place and used to teach good client behaviour and do teachers change their construction of the worthy (double) client over time? Do teachers move towards, away from or against (double) clients (Tummers et al., 2015) when the workload or other pressures increase?

Another advantage of the street-level bureaucrat perspective is that it lends itself to taking a fresh look at how education objectives and practices are linked to objectives and practices which seem to be situated outside the realm of education. The conceptualisation of children and childhood must be considered in order to understand goals and the balance of education systems, the structure of the education system and the professional ideals of teachers. Therefore, a focus on teachers’ discretion extends inquiry beyond expected, preferred or even acknowledged outcomes and focuses on policies made in the classroom.

The exploration of applying the street-level lens to education also shows the benefits of ethnographic research. Some aspects of the decisions on the worthiness of pupils and the exclusion or inclusion of children from education at the micro level can only be accessed by classroom observations in combination with extensive knowledge of the socio-political and cultural contexts. It may be one approach that fills the gap of comparative (Foster et al., 2012) research based on classroom observations. Researchers must face common problems in comparative work, such as ethnocentricity, language abilities and cultural variations as expressed in language, and so are difficult to translate (Phillips and Schweisfurth, 2015). The unit of analysis will be teachers and their practice in classrooms and schools. However, to identify all the policies on various levels and to capture the socio-cultural context will be more challenging and, in this sense, problematise the unit of analysis. And, as Hill (2009b) and Evans (2011) point out, it also requires a better understanding and focus of the intersection of professional and management issues. As an analytical tool in comparative research on education policy, the street-level bureaucrat’s perspective has the potential to illuminate aspects difficult to grasp with other means. The unique links it makes between the macro and the micro offer bridges to research with a focus on each of these arenas. One of the attractions of findings from comparative street-level bureaucracy research is it can reveal that crude policy transfer is impossible and why this is so.