Accountability and the Pressures To Exclude : A Cautionary Tale from England

Recent studies have produced conflicting findings about whether test-based rewards and sanctions create incentives that improve student performance, or hurdles that increase dropout and pushout rates from schools. This article reports the findings from a study that examined the impact of England’s accountability reforms and investigated whether the confluent pressures associated with increased testing, school ranking systems, and other sanctions contributed to higher levels of student exclusion (expulsion and suspension). The study found that England’s high-stakes approach to accountability, combined with the dynamics of school choice and other curriculum and testing pressures led to a narrowing of the curriculum, the marginalization of low-performing students, and a climate perceived by teachers to be less tolerant of students with academic and behavioral difficulties. A comparison of higherand lowerexcluding schools, however, found that these effects were more pronounced in the higher-excluding schools, which lacked strong systems and internal structures for supporting staff communication, teacher collaboration, and students’ individual needs. The study offers an international perspective on recent trends toward greater accountability in education, pointing to a complex inter-relationship between the pressures of national policies and the unintended consequences on schools’ organizational and Education Policy Analysis Archives Vol. 13 No. 26 2 teachers’ instructional capacities. The study’s findings raise particular implications for the United States and show that in the design of accountability systems, attention must be paid to how the pressures from accountability will affect the capacity of schools and teachers to respond to student who are low-performing and struggling academically. I. Accountability – Assumptions and Consequences The central assumption of many contemporary accountability schemes is that by holding schools, districts, teachers, and students responsible for results on a range of achievement and performance measures, teaching will improve and expectations for students will rise. Across the US, states have adopted new standards and testing requirements, increasing accountability and aligning their assessment systems to meet new federal mandates, with the hope and expectation of producing higher gains in student achievement. As a result of the Bush Administration’s reauthorization of the Elementary and Secondary Schools Act in 2001, also known as the No Child Left Behind Act (NCLB), states and districts must meet the federally established goal of “adequate yearly progress,” or face severe federal sanctions, penalties, and corrective actions, such as the replacement of staff, implementation of new curriculum, increased parental choice, and finally, complete takeover by the state. Accountability is being touted as the panacea for improving schools and raising achievement; yet little is known about how such pressures will affect the practices of schools and teachers, and what the long-term consequences will be for those students who are low-achieving. Under NCLB and state efforts to hold schools more accountable, schools have greater incentive to exclude weaker students from testing and staying in school. While NCLB attempts to address this concern through its subgroup requirements, the implications for increased grade retention and tracking, practices known to aggravate the dropout and discouragement of students remain largely unclear. Whether high school dropout rates will increase for students most at risk of failure; or whether more schools will exploit “zerotolerance” policies to exclude difficult students whose low achievement might mar a school’s overall performance needs further investigation, particularly in states that have responded to NCLB by using test results to reward and sanction schools and to award a high school diploma. Recent studies on accountability, however, have produced mixed results about the impact of test-based rewards and sanctions (Braun, 2004) and conflicting findings about whether the pressures on schools to raise achievement increases dropout and lowers graduation rates in schools. One line of research has found that states with high-stakes accountability systems have produced higher gains in student achievement without evidence of increased dropouts (Carnoy & Loeb, 2002). Some analysts have suggested that this is because of the focusing effects on the curriculum, which encourages students and teachers to pay attention to the exam and its content (e.g. Bishop, 1998). However, a second group of studies have questioned these gains, finding that some accountability systems have created incentives to push low-scorers into special education (Allington & McGill-Franzen, 1992; Figlio & Rustique-Forrester: Accountability and the Pressures To Exclude 3 Getzler, 2002); hold low-achievers back in grades (Allington & McGill-Franzen, 1992); and push out low-performing students so that schools’ average scores will look better (Haney, 2000; Advocates for Children, 2002). A recent study by Greene & Winters (2004) found that graduation rates have not declined in states that require high school exit exams. However, other studies have found increasing rates of retention and higher dropout in the states that have instituted tougher graduation requirements, (Lilliard & DeCicca, 2001), a widening graduation rate gap between white and minority (black and Hispanic) students (Orfield et al., 2004), and disproportionately negative effects for minority, low-income, and students with special educational needs (DarlingHammond, 2001). In states with test-based accountability systems, there is growing body of evidence that suggests that dropouts and exclusions are on rise. For example: · In the case of Texas, while the state’s rapid student achievement gains caused the state to be hailed as “The Texas Miracle,” analyses by the Center for Research and Evaluation on Testing found that retention rates in ninth grade and dropout rates for high school students have increased substantially since the 1980s (Haney, 2000). A state audit in Houston, which examined the records of middle and high schools, charged the district with falsifying student records after finding that more than half of the 5,500 students who left in the 2000-01 school year should have been declared dropouts, but were not (Schemo, 2003). The Houston case raises interesting implications because the district’s reported gains in student achievement were attributed to Texas’s system of accountability, upon which NCLB’s provisions on accountability have been modeled (Phillips, 2003). · In Massachusetts, which began requiring a high school exit exam for graduation in 2002, graduation rates have decreased from 76 percent in 2002 to 72 percent in 2003. Analysis by Wheelock (2003) found that student exclusions1 in the state increased 22% between the 1998-99 school year and the 2000-01 school year. According to the state’s official records, the 1,621 students who were excluded in the 2000-01 school year is the highest number of exclusions that has been reported during the past decade (ibid.). · In New York City, community organizations and local advocacy groups have expressed that many of the city’s public schools appear to be trying to improve their high school test scores by pushing out the weaker students who are unlikely to pass the state’s high school graduation tests, which comprise five comprehensive Regents exams. A recent article by The New York Times reported that last year more than 37,000 school-age students were in G.E.D. programs run by New York City’s school system, up from 25,500 since 2001 (May 15, 2001, p. A1). A report by Advocates for 1 In Massachusetts, student exclusion is defined as the removal of a student from participation in regular school activities for disciplinary purposes permanently or indefinitely or for more than ten days (Massachusetts Department of Education, 2003). Education Policy Analysis Archives Vol. 13 No. 26 4 Children (2002) reported that more than 55,000 high school students were “discharged” during the 2000-2001 year, a number that dwarfs the 34,000 seniors who actually graduated from high school. While some reports have found that testing can prompt teachers to focus on previously neglected subjects and classroom topics (Center on Education Policy, 2004), earlier studies on test-based reforms have pointed to the narrowing effects on the curriculum (Madaus, 1991; Madaus et al., 1992), the deskilling and demoralizing impact on teachers (McNeill, 1988) and the diminished incentive to spend time on classroom activities and tasks that promote critical, higher-level thinking skills, but which are not measured on state tests (see discussion in Shepard, 2000). Other research suggests that “teaching to the test” leads to a superficial coverage of topics, less depth in instruction, and less time for using diagnostic assessments to identify specific areas where students needed attention and intervention (Whitford & Jones, 2000). This line of research, and more recent studies point to the potential and unintended consequences of high-stakes policies (e.g., Amrein & Berliner, 2002) and the hidden and considerable costs of testing (Center on Education Policy, 2004). At the heart of these controversial research and policy debates are questions about the unintended pedagogical and social consequences of different systems of accountability. One set of questions pertains to how schools, districts, and teachers may be responding in both positive and negative ways to the increased pressures to raise student achievement. While many of the advocates of standards-based reforms had hoped that the push for accountability would spur greater investments in teaching, curriculum improvements, and targeted resources to build capacity (O’Day, Goertz, & Floden 1995) and “others imagined that teaching to a good test would be

A recent study by Greene & Winters (2004) found that graduation rates have not declined in states that require high school exit exams.However, other studies have found increasing rates of retention and higher dropout in the states that have instituted tougher graduation requirements, (Lilliard & DeCicca, 2001), a widening graduation rate gap between white and minority (black and Hispanic) students (Orfield et al., 2004), and disproportionately negative effects for minority, low-income, and students with special educational needs (Darling-Hammond, 2001).In states with test-based accountability systems, there is growing body of evidence that suggests that dropouts and exclusions are on rise.For example: • In the case of Texas, while the state's rapid student achievement gains caused the state to be hailed as "The Texas Miracle," analyses by the Center for Research and Evaluation on Testing found that retention rates in ninth grade and dropout rates for high school students have increased substantially since the 1980s (Haney, 2000).A state audit in Houston, which examined the records of middle and high schools, charged the district with falsifying student records after finding that more than half of the 5,500 students who left in the 2000-01 school year should have been declared dropouts, but were not (Schemo, 2003).The Houston case raises interesting implications because the district's reported gains in student achievement were attributed to Texas's system of accountability, upon which NCLB's provisions on accountability have been modeled (Phillips, 2003).
• In Massachusetts, which began requiring a high school exit exam for graduation in 2002, graduation rates have decreased from 76 percent in 2002 to 72 percent in 2003.
Analysis by Wheelock (2003) found that student exclusions1 in the state increased 22% between the 1998-99 school year and the 2000-01 school year.According to the state's official records, the 1,621 students who were excluded in the 2000-01 school year is the highest number of exclusions that has been reported during the past decade (ibid.).
• In New York City, community organizations and local advocacy groups have expressed that many of the city's public schools appear to be trying to improve their high school test scores by pushing out the weaker students who are unlikely to pass the state's high school graduation tests, which comprise five comprehensive Regents exams.A recent article by The New York Times reported that last year more than 37,000 school-age students were in G.E.D. programs run by New York City's school system, up from 25,500 since 2001 (May 15, 2001, p. A1).A report by Advocates for Children (2002) reported that more than 55,000 high school students were "discharged" during the 2000-2001 year, a number that dwarfs the 34,000 seniors who actually graduated from high school.
While some reports have found that testing can prompt teachers to focus on previously neglected subjects and classroom topics (Center on Education Policy, 2004), earlier studies on test-based reforms have pointed to the narrowing effects on the curriculum (Madaus, 1991;Madaus et al., 1992), the deskilling and demoralizing impact on teachers (McNeill, 1988) and the diminished incentive to spend time on classroom activities and tasks that promote critical, higher-level thinking skills, but which are not measured on state tests (see discussion in Shepard, 2000).Other research suggests that "teaching to the test" leads to a superficial coverage of topics, less depth in instruction, and less time for using diagnostic assessments to identify specific areas where students needed attention and intervention (Whitford & Jones, 2000).This line of research, and more recent studies point to the potential and unintended consequences of high-stakes policies (e.g., Amrein & Berliner, 2002) and the hidden and considerable costs of testing (Center on Education Policy, 2004).
At the heart of these controversial research and policy debates are questions about the unintended pedagogical and social consequences of different systems of accountability.One set of questions pertains to how schools, districts, and teachers may be responding in both positive and negative ways to the increased pressures to raise student achievement.While many of the advocates of standards-based reforms had hoped that the push for accountability would spur greater investments in teaching, curriculum improvements, and targeted resources to build capacity (O' Day, Goertz, & Floden 1995) and "others imagined that teaching to a good test would be an improvement over low-level basic skills curricula" (Shepard, 2000), researchers have found that local responses to state and national policies can vary by districts and schools because their effects become mediated by a range of policy factors as well as a range of contextual factors (Spillane, 1998(Spillane, , 1999;;McLaughlin & Talbert, 1993).
Historically, curriculum and instructional decisions of schools and teachers have been largely unaffected by federal educational policies, in part because districts and states could decide how to define, measure and monitor student progress.However, the impact of NCLB, with its requirements for student testing and its demands on schools and districts to maintain "adequately yearly progress," represents an unprecedented shift in the role of the federal government, a change that Elmore (2003) describes as "the single largest nationalization of education policy in the history of the United States" (p. 6).
Little is known, given the relatively recent reforms in the United States, about how teachers are responding to the demands of increased accountability in their own classrooms, and the ways in which their beliefs, expectations, and perceptions of their students may be changing amidst the pressures of a profoundly changing policy climate.Surveys of teachers have found that while teachers support accountability (Pedulla et al, 2003;Begley & Jones, 2004), the implementation of more frequent testing and the use of tests to make high stakes decisions for graduation and placement are leading to curriculum decisions and instructional practices that many teachers believe to conflict with their own professional judgment.

Overview of Report
This article is a report of the findings of a research study conducted between 1998 and 2001, which examined the consequences of England's accountability framework by investigating the nature and impact of the pressures placed on schools and teachers.The paper reports the findings of a study, which analyzed the subsequent and steep rise in student exclusions (defined as suspensions and expulsions), which increased 400% between 1990 and 1997, and explored how the internal organization and culture of schools influenced the capacity of teachers to respond to the demands of accountability and to resist incentives to exclude students with behavioral and academic difficulties.
The case of England, which put into place a high-stakes2 accountability system through a combination of national testing, curriculum, inspection, and school choice policies offers a cautionary tale about the potential consequences of combining accountability with a prescriptive curriculum, increased testing and national performance targets.While there are profound differences between England and the US in the governance of public education, there are some useful points in considering the unintended consequences and perverse incentives that may accompany broad and inflexible educational policies.For example, there are some similarities between some components of the reforms implemented in England after 1988, and features of the accountability systems adopted recently in Texas, Florida, Massachusetts, and New York.These include the setting of student performance targets to monitor and measure yearly progress, more frequent testing of students, the ranking and comparison of schools on the basis of student test results, and the awarding of qualifications based on exit exams in order to receive a high school qualifications.While there are differences between the US and UK with respect to the role of local school boards, the structures and funding of schools, and the policymaking powers of national government, Olsen (2004) recently observed that the "striking resemblance" between the elements of No Child Left Behind, and England's national education strategy for raising standards and student achievement could mean that "the lessons learned [from England] may prove a harbinger for the United States" (p.1).

II. Background To The Study England's 1988 Reforms
Beginning in the late 1980s and through the mid-1990s, England implemented, under the Conservative administrations of then-Prime Ministers, Margaret Thatcher and John Major, a set of national educational reforms aimed primarily at raising achievement and increasing public accountability.This package of sweeping and radical educational reforms, known as the Education Reform Act of 1988, resulted in a nationalized curriculum, national assessments, increased school and parental choice, national performance targets, national inspection for schools, and the ranking and comparison (by national newspapers) of schools based on students' test performance.The theory of action behind these reforms was that greater control by the central government, combined with the use of market forces, would improve schools and enable parents to have greater choice among schools on the basis of their quality, a choice partly informed by test results.
Essentially, England's national reform strategy embodied two conflicting policy strategies of decentralization and centralization.On the one hand, decisions about curriculum and academic performance measures were determined externally and nationally.On the other hand, schools were given discretion over school budgets and physical operations, with responsibility to purchase services and supports previously provided by school districts.This shift created tensions in how schools functioned and how teachers and headteachers (principals), in particular, defined their goals and purposes.
Among the first changes that occurred was a shift in the funding from and functions of school districts (called local education authorities, or LEAs) under the banner of "locally managed schools" (LMS).With the aim of making schools more accountable to the resources received from the national government, school funding was devolved, giving schools an independent budget and linking the number of pupils on roll to the funding they received (Whitty et al., 1998).With the expansion of parental choice, schools had to compete against each other to attract students and to direct efforts at "marketing" themselves to parents.
Another core feature of the reforms was the establishment of a National Curriculum in 1992, now followed by all state-run and assessed through a program of national testing.It has been widely acknowledged that the underlying ideology of the Conservative Government, under which the National Curriculum was implemented, stressed the need for greater accountability from local education authorities (school districts), a common structure, and a return to "real" academic knowledge (e.g.Basini, 1981;Helsby & McCulloch, 1997).Rejecting the local curriculum standards developed collaboratively by schools and districts under a previously decentralized system, the Conservative government sought to define through the National Curriculum, the "core" subjects that all students were required to take.Currently, the National Curriculum requires students to follow a curriculum comprising religious education and the following subjects: English, mathematics, science, design and technology, information technology, history, geography, art, music, physical education, and foreign language (Docking, 1997).
Following the implementation of the National Curriculum, a program of more frequent student testing was established for primary and secondary schools, along with a set of nationally established student attainment goals that would be measured by assessments administered at ages 7, 11, 14, and 16.As part of the National Curriculum, schools are required to administer standard assessments tasks (SATs) as four "key stages: At age 15, students are "entered in" (a decision made by teachers) to take exams in three core subjects (English, mathematics, and science), with the option of taking exams in other subjects within the National Curriculum.Students receive scores on an eight-point scale (A*, A, B, C, D, E, F, G), with "A*" being the highest, "C" considered passing, and "G" failing.Students who receive a "G" are recorded as "U" for unclassified and do not receive a certificate.
According to the government's national target, all students should aim to earn at least a "C" on at least five GCSE exams.While the GCSE is not required for a student to graduate from secondary school, his or performance on these exams has a direct bearing and influence on a student's admission into university and access to education beyond secondary school.
Although the majority of higher education institutions are autonomous bodies which set their own admissions policies, applicants are generally required to have obtained at least three GCSE "passes" at grade "C" or above and two General Certificate of Education Advancedlevel (commonly known as "A-level) passes, advanced exams that typically are taken at ages 16-18, in the final two years of secondary schooling.This stage of further education is considered post-compulsory and commonly known as "sixth form." Another aspect of the reforms was the development of accountability mechanisms aimed at increasing public accountability and expanding parental choice.In England, schools are judged on the percentage of students who achieve scores that meet national performance targets.The requirement that schools make public their results subsequently led to the annual publication of school rankings by national newspapers in what are known as "league tables."The combination of performance targets and league tables, critics observed, was that they encouraged schools and the public to value only those young people who achieved the national standard, or the top three GCSE grades (Lovey, 2000).
A second mechanism aimed at increasing school accountability was the centralization of local school inspection in 1992.Previously viewed and used by schools as an innovative model of peer support and advice, the process came under the control of the national government, following the creation of OFSTED (Office of Standards in Education).OFSTED inspection emphasized a more high-stakes and external model of accountability.Standards for inspection were developed at the national level and school inspection reports are publicly available, another tool for promoting parental choice.
During this period of reform, the authority of schools and LEAs to make key decisions about curriculum, assessment, and instruction was dramatically reduced under the increasing control of central government.Docking (1996) observed that "whatever view people take about education policy since the Conservatives won the general election in 1979, they are agreed that the changes have been radical, pervasive, and controversial" (p.x).Although England's government changed from a Conservative to a Labour-led government in the 1996, following the election of Tony Blair, the main features and structures of England's accountability system have continued to remain firmly in place.(Parsons, 1999).e All of these figures are from DfES (2002).

The Rise in Exclusion
In the decade following the implementation of these sweeping reforms, the incidence of school exclusion in England's schools increased steeply throughout all sectors of schools.
In the case of permanent expulsions (the most severe form of student exclusion in which a student is officially removed from school and either transfers to another school or receives alternative provision), the numbers of exclusions from primary and secondary schools increased 400 percent between the 1990-91 school year and the 1996-97 school year, from 2,910 students to 12,668 students, as shown above in Table 1.Although exclusions have decreased slightly since their peak in 1997, they have not fallen below their levels, pre-1988 reforms.
Alarm over the steep increase in the numbers of students being removed from school led the new government, under Tony Blair, to form a number of national committees and task forces to investigate the rise in exclusion rates and to formulate proposals to reduce exclusion (Commission on Racial Equality, House of Commons, 1998;Social Exclusion Unit, 1998).
With growing public concerns over the short-and long-term adverse social consequences of exclusion --the increased costs of providing alternative provision, and the greater risk for juvenile crime, imprisonment, unemployment, and poverty --the government commissioned large-scale, quantitative studies to examine the causes and explanations.These studies found school expulsion rates were strongly correlated with student demographic variables, in particular, race, class, and gender (OFSTED, 1996;Parsons, 1996Parsons, , 1999)), and showed that certain groups of children were being disproportionately excluded, including students from Afro-Caribbean background (Commission for Racial Equality, 1996), students from lowincome families, and students with special educational needs (House of Commons, 1998).
At the same time, these reports offered little explanation as to why schools might be resorting increasingly to exclusion.Gillborn's (1996) analysis of exclusion figures and school reports revealed that students were being excluded for relatively minor incidents, challenging the perception that students were becoming increasingly aggressive or violent.Significantly, Gillborn (1996) found that most decisions to permanently exclude students arose from incidents of disobedience and disruption, and not from physically threatening behavior, which actually accounted for only one in four permanent exclusions: The Education Department's figures indicate that: Disobedience in various formsconstantly refusing to comply with school rules, verbal abuse or insolence to teachers -was the major reason for exclusion (DfE, 1992, p.3 cited in Gillborn, 1996).Osler & Hill (2000) also concluded that while boys represented the majority of exclusions, the rate at which girls were being excluded was rising more steeply.Similarly, Parsons (1996Parsons ( , 1999) ) pointed to rising numbers of primary school exclusions.Large-scale, quantitative studies revealed another important dimension to the problem of exclusion: significant variance between individual schools' exclusion rates that could not be explained by student background factors (Parsons, 1999;OFSTED, 1996).Earlier studies on exclusion found that differences between school discipline policies, the style and approach of teachers and the ethos of schools also explained differences between the expulsion and suspension rates of schools with similar student intakes (Rutter et al., 1979;Galloway, 1982Galloway, , 1995)).
Largely missing, however, were any studies or theories about the influence of the wider policy context as contributing to the increase in student exclusion rates.Although government reports acknowledged the increased demands on schools to implement reforms, and testimony from teachers and local school officials implied that schools might be resorting increasingly to exclusion to cope with students whose academic and behavioral struggles were difficult to address amidst such pressures, no study had examined the relationship between the impact of these national policies and the rise in exclusion.The lack of research in this area led one team of researchers to conclude: It is clear that further direct investigation about what is happening in schools and what subtle processes might be in operation is needed.It is likely that qualitative studies …of teachers' styles and attitudes, and schools systems are more likely to reveal more meaningful findings (Gersch & Nolan, 1994, p. 36).
The evidence that schools were resorting more frequently to exclusion pointed to an increasingly intolerant environment in classrooms; yet there was little discussion of the policy factors that could be contributing to this changed climate.The lack of empirical research in this area thus formed the central motivation for this study, which sought to examine whether pressures from accountability pressures had affected teachers' and principals' beliefs, expectations, and actions in ways that could explain the rise in exclusion or the different patterns across schools.

Conceptual Framework Theorizing the Wider Context of Exclusion
In this study, the national policy context under which schools operate, and the local school context in which teachers perceive and respond to their students is central to understanding how and why exclusion occurs.This theorization of exclusion --as a complex, systemic phenomenon reflective of local school decisions and influenced by external factors such as national policies --suggests a different view from studies that have examined exclusion within a behaviorist framework (e.g.York et al., 1972).Rather than focus narrowly at a student's behavior, personality, or social characteristics to investigate the causes and dynamics, I argue that the context of exclusion is much wider, and that national policies as well as a school's organizational context critically influences how teachers respond to their students' behavior and academic needs.
For the purposes of the study, the term "exclusion" refers specifically to the permanent removal (known as "permanent exclusion") or temporary suspensions (known as "fixed-term exclusion") of a student from the school where he or she is officially enrolled.This definition is adopted from the current legal, procedural, and statutory framework in England.
While statistics on exclusion are regularly collected, such data do not reveal the wider context in which exclusion occurs, or the sources of trends.To assess whether the impact and interaction of policies within schools may have created new pressures to exclude students, this study developed a conceptualization of exclusion that theorizes its causes and dynamics to include the factors that affect how teachers and schools respond to student's behavioral and academic needs.

The Embedded Contexts of Exclusion
Defining exclusion as a complex, systemic, and policy-related phenomenon does not mean that an individual's behavior and background are unimportant in explaining how and why exclusion occurs.Rather, the study draws on the notion of "embedded contexts," a conceptualization adapted from McLaughlin & Talbert (1993) that theorizes the practices of teachers and schools as being influenced by multiple policy contexts and mediated by the school organizational context, which can comprise such factors as a teacher's professional communities, level of staff collaboration, and opportunities for professional development (See Figure 1).Studies that have examined the impact of various aspects of educational policies by focusing on schools' and teachers' responses to policy have utilized variations of this conceptual framework, drawing upon institutional theory to study the interaction between policies and practices (Spillane, 1998;1999) and teachers' implementation of educational reforms (Coburn, 2001).
As Figure 1 shows, this study views the context in which exclusion occurs as comprising of multiple layers, which includes: policies at the national-, state-and local district-level; the professional contexts of higher education and associations; the local community and school districts; and the school's organizational context.As result of these complex interactions, policies can exert multiple and interacting influences on the decisions and actions of schools, as well as teachers' perceptions and practices.
In theorizing that the demands and pressures of the wider national-and local context influences the practices and actions that occur in schools (McLaughlin & Talbert, 1993;Spillane, 1998), this study views teachers and principals as professionals who operate in a highly contested arena and whose decisions are influenced by the context in which they work and interact (Rosenholtz, 1991).This view of teaching also draws from a conceptualization of a teacher as a kind of "dilemma manager, a broker of contradictions," in having to make "dichotomous choices … between promoting equality or excellence; building curriculum around children's interests or around subject matter, fostering independent creativity or maintaining standards and expecting everyone to meet them" (Lampert, 1985cited in Fang, 1996, p. 53).As "reflective practitioners" who require autonomy to creatively solve problems in response to the unique needs of each situation (Schön, 1983), teachers' and principals' explanations of the pedagogical tensions and conflicts they experience in relation to the demands of specific policy mandates and requirements can thus provide critical insight to the unintended consequences of educational reforms, including accountability.

Figure 1 The Embedded Contexts of Exclusion: The Interaction between National Policies and Local School Practices
(Adapted from MacLaughlin and Talbert, 1993) A further argument for taking a broad view of exclusion draws from social constructionism, which views social phenomenon in terms of the actions, perceptions, and interactions between individuals and institutions (Burr, 1995).The application of social constructionism to explain the impact of policy can been found in the work of educational historians such as Tyack and Cuban (1995) who view teachers as actively mediating norms, belief systems, and practices, while socially constructing and reconstructing their practices in order to fit within their own context (Coburn, 2001).The conceptual framework for this enquiry draws also from organizational theory, which views human behavior as being influenced, in large part, by social structures and institutions (Morgan, 1997).In other words, this study suggests that exclusion cannot be attributed to a single cause or understood by looking solely at the behavior of an individual.Rather, the causes and dynamics of exclusion might be better viewed as a reflection of a wider set of influences, including how schools and teachers interpret and respond to students' needs and the pressures that influence those capacities.In short, the aim of this study's conceptual framework is to move toward a richer understanding of the wider school and policy context in which exclusion occurs.

Study Design & Methods
To investigate whether and how the pressures from national reforms may have contributed to the rise in exclusion, the study examined how teachers perceived the causes of exclusion, and assessed whether and how they linked school conditions and policy considerations, including accountability, to exclusion.This article discusses how staff responded to several key questions: What policies were associated with having aggravated or helped to reduce exclusion?What kinds of pressures did teachers, headteachers, and other staff feel from these policies, and what effects did these pressures have on students who were having difficulties in schools?
The data collected for this investigation drew primarily from interviews with 44 staff (teachers and principals) from four secondary schools located in high-poverty neighborhoods in Southeast and Northern England.As shown below in Table 2, two of the schools (School R and School S) had high rates of exclusion, whereas the other two schools (School L and School M) were low excluding.The selection of these schools was based on their exclusion rates, their level of social disadvantage, and the opportunity for access through a project evaluation commissioned by the LEA in which two of the schools were located.This selection of schools with differing rates of exclusion rates provided a way to investigate why exclusion rates might vary in schools with similarly challenging student populations, a comparison of staff capacity to resist and prevent exclusion, and a exploration of whether and how the pressures of national policies might be interpreted differently across schools.
• School L was a large, ethnically diverse secondary school of 1,922 students with high levels of social disadvantage located in an inner city area of Southeast England.44 percent of the students were eligible for free school meals (equivalent to "free lunch") and 30% were reported to have special educational needs.At the time of the study, School L had not permanently excluded any pupils since 1993.
• School M, another low-excluding school was a small, ethnically diverse secondary school of 454 students with similarly high levels of social disadvantage located in an urban area of North England.Half of the students were eligible for free school meals, and 44 percent of students had special educational needs.Like School L, School M had a low incidence of exclusion and permanently excluded fewer than five pupils between 1998 and 2001.
• School R was a medium-size secondary school enrolling 614 students and located in the same town as School M, with similarly high levels of social disadvantage, but with only 11 percent of students from ethnic minority backgrounds.Approximately 63% of the students were eligible for free school meals and 43% of the students had special educational needs.According to the reports made available by the local education authority, 33 students were excluded between 1997/98 and 2000/01.
• School S was a medium size school of approximately 620 students located in the Southeast east region of England.Approximately 43 percent of students were reported to be eligible for free school meals.Over 50 percent of the students came from ethnic minority backgrounds, and the school had the highest percentage of Afro-Caribbean students of the four schools.According to the figures provided by the school, 24 students were permanently excluded between 1997/98 and 2000/01.
Table 3 below provides a description of the interview sample for each of the four schools.Individuals who would be particularly knowledgeable about the process of exclusion, the needs of students with behavioral and academic problems, and the implementation of testing and other policies were selected on the basis of their role in the school.In England's secondary schools, these roles typically comprise headteachers (principals), special needs coordinators (SENCOs), and heads of year (senior teachers with oversight for an entire grade).The total interview sample thus included teaching and administrative staff from a range of subjects and disciplines (including mathematics, English, history, science, physical education, and vocational arts) and a range of experiences and positions.As indicated in Table 3, four to five visits were made to each school during the 1999-00 and 2000-01 school years.Interviews were conducted on the school site, tape-recorded and transcribed, and included a variety of formats: individual interviews, group discussions, and informal conversations such as those that take place in the hall or during a lesson.To protect the identity of the teachers in each of the schools, the teachers and principals are not described by their roles, but are identified as "staff members" of their particular school.

Number of teachers interviewed
The interview data collected from each of the four schools was treated as a whole sample, and then analyzed within and across school types (higher-and lower-excluding).In analyzing this data, attention was aimed mainly at constructing the themes that emerged from individuals' views, and generating theories about the interaction of policies.A consideration in the analysis of this data was whether staff in the higher-and lower-excluding school differed in how they viewed and described the causes of exclusion, and the influence of school context.Thus, in addition to interviewing staff in these schools, ethnographic case studies were conducted in each of the four schools to confirm perceptions and generate additional theories about influence of school organizational factors on exclusion.

The Pressures to Exclude
The findings reported in this section draw largely from the data collected through interviews with the teachers and principals in all four schools.The section is organized around first, a discussion of the overall pressures and dynamics that teachers and principals perceived in relation to specific national policies, and second, an analysis of the differences between higher-and lower-excluding schools about their capacity to resist the pressures to exclude.
To illustrate the various themes that emerged in relation to each of the policies, quotes have been selected from across all four schools to illuminate the nature of the pressures generally felt and experienced by teachers.However, as will be discussed later in the paper, analysis of this interview data, in combination with data collected from the ethnographic case studies of each of the schools, found that the perceived effects of these policy pressures were more pronounced in the higher-excluding schools, thus pointing to the influence of school context on teachers' capacity to cope with these pressures.

Pressures from national policies
The study found wide agreement from staff within and across each of the schools that a combination of national policies had created greater incentives to resort to exclusion; however, a comparison between the higher-and lower-excluding schools revealed differing levels of capacity to resist these pressures in the context of their own school.The increased incentives to exclude students were linked to four national policies: 1) constraints from the National Curriculum; 2) pressures from national assessments and performance targets; 3) effects of school ranking and monitoring systems (league tables and OFSTED inspection); and 4) dynamics of parental choice and school competition.

a. Constraints from the National Curriculum
Across all four schools, teachers described constraints from the National Curriculum: less flexibility to make decisions about instruction, being driven to cover content due to the demands of testing, and intense pressures to meet national levels of attainment.The impact of these constraints was viewed as making it more difficult for schools to adapt and differentiate the curriculum to fit the needs of individual students, especially those viewed as low-performing.These constraints on schools and teachers were described as being frustrating for both teachers and students, and leading to behavior for which students were more likely to become excluded.For example, one teacher said: Certainly, throughout the country, the numbers [of exclusion] have gone up.I think one reason put forward is the introduction of National Curriculum, which is actually very strict, and gives very little room for movement…it does mean that [for] children who find school very difficult … teachers have very little leeway with them.
Therefore when [teachers] feel they can't get deal any longer, then the answer is exclusion.(Staff member, School L) Others similarly viewed the National Curriculum as prescriptive and rigid, describing the pressures to "get through the material" in order "to cover the topics that would be tested."Another teacher, in describing the challenges of accommodating students who did poorly on national assessments and attended classes irregularly, explained how the prescription of the National Curriculum made it difficult to accommodate their specific needs: You follow a syllabus which is tested and so the children have to be there [for the lessons]… and also for the national assessments.The teacher is trying to get on with those commitments under that sort of pressure … you have only two years to get through this and it makes it critical that you can get through the work (Staff member, School R).
Other staff linked exclusion with increasing levels of frustration from students who needed more time and in-depth instruction, but explained how the pressures of keeping up with the "sheer pace" of the curriculum made it "impossible for some pupils to keep up," and "difficult to stay on top of all the coursework."For example, one teacher explained: The curriculum can be very rigid, and for certain pupils it doesn't work and once they come up against the curriculum, frustration sets in then you see the different types of behaviours and attitudes come out.Ones that can overcome and make it …will succeed and achieve.But there are a significant number of others who don't.(Staff member, School S) Another teacher, in describing the constraints and challenges of balancing the individual needs of students with the pressures to cover curriculum for the whole class, noted the decreased tolerance for disruptive behavior and addressing individual behavioral problems, particularly with the competing demands of preparing the rest of the class for GCSEs: In Although some teachers acknowledged that the National Curriculum did "offer guidance" and helped to define "what should be taught," the curriculum's rigidity and the pressures to cover material created a set of unintended consequences that were linked to exclusion.Even in the lower-excluding schools, the overall pressures to move quickly through topics were believed to discourage teachers from "spending too much time," "taking risks," and "being creative" -effects that could be viewed as having a particularly discouraging impact for students who needed individualized strategies.For example, one geography teacher described the conflict between following the National Curriculum's syllabus and pursuing topics that he believed were of interest to his students: I teach geography … I think there is a case to be made that some students are alienated from what they are doing [in the National Curriculum]… I think in a lot of cases there is a lot of leeway to follow students' interests, [but] you have to move on and say we are doing this today.In a school where we have got a high proportion of children who are weaker academically, they find it very difficult to follow these sorts of courses (Staff member, School M) According to other staff, the pressures on students "to reach the levels" of attainment established nationally by the National Curriculum, discouraged teachers from spending too much time on a particular topic and pursuing strategies that did not fit directly within the scope and sequence of National Curriculum.As one teacher explained: I think the National Curriculum had had a bad effect [on exclusion] because all the children are under pressure to reach the levels.There is pressure on the staff as well to teach that work.I was teaching before the National Curriculum was introduced across Britain and in that particular school, lessons…were far freer.You did not have the restraints of having to do that work in a set time.You could actually teach the children without frustration (Staff member School M).
Another teacher, in describing the boundaries of the National Curriculum described how pressures to get through the material had overshadowed the social aspects of schooling and changed how teachers responded to students' behavior: The National Curriculum has so many boundaries on the work that has to be done.The work seems to overtake the social aspects of the school.People feel they are under pressure to get the work done and they deal with behavior in a different way (Staff member School M).
The perceived rigidity of the National Curriculum's scope and sequence, and its shift from a student-centered approach to a content-driven focus, was viewed as having reduced the flexibility to develop instructional strategies that might help students engage with the curriculum.Rather than expanding the opportunities for students, the constraints of the National Curriculum had a narrowing effect on what teachers felt they could do to address the needs of student with particular difficulties.When students are disengaged or alienated from the work they do in class, they become disaffected and disenfranchised from school (Riley & Rustique-Forrester, 2002), increasing the likelihood for disruptive behavior and increasing levels of frustration.

b. Pressures from testing and targets
The combined pressures of testing and performance targets were also associated with having aggravated exclusion.As explained earlier, the pressures are particularly pronounced for teachers teaching students at age 15 (Year 10) and age 16 (Year 11), when students prepare for and then undertake the GCSEs in different subjects.Secondary schools are then judged by the percentage of students who achieve a highly proficient score (A* to C) for at least five GCSEs.
Across the four schools, the pressures from testing and targets were felt to have created a climate less tolerant of students with behavioral problems.For example, one teacher, in describing the stress and pressures of Year 11, the final year of secondary school when GCSEs are administered, noted the frustration expressed by students and the lowered capacity of teachers to respond: In The belief that pressures from high-stakes assessments, specifically GCSE exams, contributed to a climate in which schools and teachers less tolerant of disruptive behavior was expressed by another teacher, who said: "Because of exam pressure, if you have students who are continually disruptive and are stopping able students from doing well then the easiest way is to ship them on elsewhere!" While staff in all four schools described pressures from testing and targets, some individuals indicated having strategies for meeting these increased demands.One teacher, for example, acknowledged that the emphasis on school performance and progress had some positive benefits for the school, but had to balanced by a continued focus on individual students' needs: Whilst we have responded to recent external requirements for target setting, we value highly our own processes for identifying individual potential and promoting individual achievement.The process of monitoring and analyzing individual and whole school performance and progress has been significant in promoting individual achievement.The fact that we are a small school has enabled us to analyze by potential and outcome more easily.(Staff member, School M) A teacher in another school similarly emphasized the importance of helping individual students not to feel daunted or discouraged by the pressures of testing.During a schoolwide assembly she explained to Year 10 students: We know how hard it is to do well on exams and meet those targets…but at the end of the day, it is much harder to go out in the world and to be a good person … [to be] someone who does good things for the good of the world (School L).
The belief that the external demands from national policies needed to be balanced with other goals of schooling was a theme that featured far more strongly in the lower-excluding schools, a finding that will be discussed later in the paper.The implications of this finding suggested that although the pressures were felt across all four schools, the incentives to exclude students could be resisted the school's ethos and processes.

c. Pressures from league tables and national inspection
Teachers also linked exclusion to the impact of league tables and national inspection, and the pressures of being ranked and compared with other schools.In England, students' raw test results are compared with other schools in the country through "league tables," or annual rankings published by national newspapers.In the higher-excluding schools, staff expressed particularly negative views, describing feelings of failure and discouragement in consistently being ranked at the bottom of the league tables, year after year.
One teacher, in describing the impact of league tables, viewed the practice of comparing and ranking schools as having more damaging than motivating effects on the school, reinforcing the perception among students as "failures."She explained: I think league tables have been used to brow beat schools, and a school like this now sees itself as always being a failing school.We were happy ten years ago knowing that the children that we got, we did a very good job with.
[But] there has obviously been a mistake [with] league tables [which] puts pressure on children terribly … because that label [of failure] is felt by everyone … staff are under pressure to get results.(Staff member, School R).
Another teacher similarly described how in her school, the pressure to raise exam results resulted in a decision to begin tracking (called streaming and setting) pupils, which "worked fine for high-achieving kids, [but] was completely disastrous for the low-achieving kids, who felt labeled and worthless."Though few teachers would admit that exclusion had been used in their own school to boost test scores; several suggested that pressures to improve a school's ranking on league tables had increased the likelihood that students with behavioral problems were removed from class to minimize disruption and preventing other students from performing well during times of exam preparation and inspection.One Head of Year who had oversight over the preparation of students for GCSEs, offered his observation about the pressures on principals to improve achievement and provide results: Schools want to be seen as performing.No school wants to be associated with low performance.So the school tries as much as possible … and the only way we can do that is to get rid of those who in one way or another … are not allowing [improvement] to happen (Staff member, School R).
The decreased tolerance for students with behavioral challenges was also revealed in teachers' descriptions of the pressures of national inspection, known as OFSTED.Teachers testified that this policy created incentives to exclude because schools had "no or low tolerance" for pupils whose behavior would be disruptive during an inspection.For example, one teacher said, There's another angle…because when schools near inspection, they get rid of those kids who are causing problems…I've been at other schools as well…where, even if it's just a week or two to get rid of those who are causing problems…then you have a nice rosy picture when the inspectors come.You can send these pupils to a certain location.And just hide them for a day.Or you can ask them to stay home.And it's done at a number of schools.And officially, whether or not schools call them expulsion, it really is a type of exclusion.(Staff member, School S).
According to another teacher, disruptive students could be "liabilities," by "preventing the school and other students from performing," during the high-stake periods of exams and inspection.
Although the impact of these school ranking and inspection systems was generally felt as having increased the likelihood that students who were low achieving and who had behavior problems would be excluded, the staff in the lower-excluding schools expressed different views about resisting these incentives.For example, one vice-principal explained how his school had managed to prevent and resist exclusion, with the support of parents and teachers: We haven't excluded anybody permanently for many years … I think it's about four years now…we don't have a non-exclusion policy [and] will exclude if necessary.But we prefer to find alternatives and we've been successful at doing that so far.The governors [local school board] are full behind that approach, and it's very evident that the entire teaching staff [are] behind that approach as well.(Staff member, School L) While this individual believed that having an "inclusive ethos" could help resist exclusion, the staff in the higher-excluding schools did not express similar views.In contrast, staff in the higher-excluding schools described more acutely the effects of being judged by student test performance and the pressures of inspection as aggravating the need for exclusion.

d. Pressures from school choice and competition
A final set of pressures described as having lowered the tolerance for low-performing students related to the impact of school choice, which had caused schools to become more selective in their student admissions processes and created incentives that encouraged schools to exclude certain students in order to attract a "balanced" student intake.One senior administrator offered a detailed explanation of the impact of parental choice and competition between schools, contributing several insights, which directly linked school choice with greater incentives to exclude: I think nationally, the reason for the increase in exclusions has been parental choice and competition between the schools.That's been the real reason.When the government introduced parental choice, the idea was that parents could send their children to the school of their choice.The government introduced this idea of parental choice with a view to raising standards.Presumably the best schools would get better, and the poor schools would be disgraced and go out of existence if no one chose to go to them.But it's not quite as simple as that because schools don't actually go out of existence.But certainly, the good schools got better, and the poor schools generally got worse.(School S).
He described exclusion as a form of "purification" a mechanism for improving the "status" of a school and its attractiveness to parents: Exclusions, as I saw it, certainly working in a school outside an inner city, was a tool which heads were able to use to purify their schools… in the worse form of purification.So if there were difficult children who were causing discipline problems in the school, this was an easy way out for them.If a school was seen by the parents as one where they just excluded difficult children, their status went even higher.There's no question in my mind that schools used this as a ploy to become more popular… because they've gotten rid of themselves of difficult children.Now I believe that the number of exclusions went up dramatically with choice because it wasn't in schools' interest to have too many difficult children.Now, no one would ever admit that, but that's what actually happened (School S).
Although the depth of this individual's analysis was not typical of other views expressed among staff in the other schools, others similarly described the dynamics of increased parental choice as having encouraged the use of exclusion as a strategy for influencing student enrollment.One teacher, for example, described the "marketization of schools" as having "caused schools to be more selective … turning away and also excluding" those students "who did not contribute positively to the school's image."Another teacher perceived an increased "unwillingness amongst schools to offer a place to students [who were] previously excluded."Others observed that with school choice, exclusion had become a "transfer of pupils" from "good" schools to "bad" schools.One teacher noted that "better schools had no incentives to retain or keep pupils" who were "challenging" or "had special needs" because "such students represented an additional drain on school's resources and teacher's time."Another teacher said, "There's less time and little interest on the part of a school to devote valuable resources to a pupil who is not performing."Among the unintended consequences of these dynamics was that schools, which were underenrolled, were labeled as "sink" schools.One teacher explained that, "Sink schools don't have a choice…but to take those pupils who were excluded," offering his explanation about how the acceptance of students from other schools added to the pressure of producing results within his own school: We accept a lot of pupils who come from other schools who may have been excluded or banished from other schools.We take on here and it can cause us problems.Some of these schools have not actually have worked hard with … the children before excluding them.That is a pressure from these policies; it is a pressure from having to produce results having to appeal to parents ... that is certainly there.(Staff member, School S).
The decreased tolerance for disruptive behavior was also linked to the dynamics of choice and greater incentives to be more selective with students.According to one teacher: "The solution lies with the pupils we take into this school … we need to be more selective if we want to reduce disruptive behaviour and exclusions."Another teacher (in the same school) similarly concluded: "If you want to concentrate on raising achievement … pupils who you can't help, you have to let them go."Staff in the lower-excluding schools did not indicate that this had occurred in their own school.However, the effects of school choice were generally acknowledged as having discouraged in other schools, the admittance of students who were unlikely to contribute to a school's performance, were felt to provide another explanation for the rise in exclusion.
The overwhelming response across all four schools was that national policies had contributed negatively to exclusion, through a confluence of pressures on schools and teachers.No individual felt that the national rise in exclusion could be explained by a single policy, cause or factor; rather, the universal view was that exclusion was a complex problem, influenced by multiple factors.This finding was critical to the study, reinforcing the utility of conceptualizing exclusion as a phenomenon of multiple dimensions and confirming the limitations of defining the causes and dynamics exclusion in terms of behavior.
Although staff in all four schools felt similarly about the increased incentive to exclude students with academic and behavioral difficulties, the effects of these pressures appeared to be more pronounced in the higher-excluding schools.Although staff in both the higher-and lower-excluding schools similarly concluded that the impact of national policy had contributed negatively to the national rise in exclusion, staff in the lower-excluding schools appeared to indicate that pressures could be resisted by developing systems and structures for supporting students within their school.The different rates of exclusion between the higherand lower-excluding schools points, therefore, to the influence of school-level factors in mediating the effects of national policies.

The Influence of School Organizational Context and Capacity
Analysis of the interview data pointed to several differences between the views of staff in higherand lower-excluding schools, and in how teachers interpreted and responded to the pressures they felt from accountability and other policies.In assessing the effects that the pressures of policies had on exclusion in their own school, stark differences emerged first, in how staff explained their role and responsibility in relation to exclusion; and second, in how staff assessed their individual capacity to respond to the needs of students, particularly those with behavioral and academic difficulties.Higher-excluding schools indicated a lower capacity while staff in lower-excluding schools showed a greater capacity.These differences in capacity appeared to be related to the organizational differences between the schools.
Staff in the lower-excluding schools, while acknowledging the increased demands and constraints from a combination of policy pressures, pointed to a combination of efforts within their school to focus on students' individual needs and to build their capacity to resist exclusion.Indeed, in the case of the lower-excluding schools, School L had not permanently excluded a student for five years, and School M had resorted to permanent exclusion on only three occasions.The explanation offered by one staff member in School L, pointed to the role of school ethos as helping to resist exclusion: There's very much an ethos that this is an inclusive school … we say there is room for all… We try to make sure that every child gets some kind of opportunity.We very much try, when we have a special occasion or an important visitor... to find new youngsters to get involved….Some of those youngsters identified with emotional and behavioural difficulties, we make a point to welcome visitors and to give them a tour.We also try to pick out the other youngsters who may be shy or very quiet ….We try and give them an opportunity to shine.(Staff member, School L) Deliberate efforts to encourage within the school, an overall ethos and classroom practices aimed at supporting students at risk of exclusion, particularly those with behavioral difficulties, was also described by a staff member in School M: [To prevent exclusion] we have a lot of steps in the process and … a forgiving nature of the staff.If there is a behavioural problem in that department, we will expect [the teachers in the department] to put in a number of steps themselves.They may make contact with home with a simple letter, saying they are not happy with behaviour.(Staff member, School M) In contrast, staff in the higher-excluding schools described their own role and the responsibility of the school in much different terms.A teacher in School S, for example, explained: Quite frankly these days, the role of the teacher is not to spend time bringing youngsters around to fitting into the curriculum.The role of the teacher is to increase achievement.Increase GCSEs.Increase entry into GCSEs.That is their role, not to spend too much time with disaffection, attitude, or behaviour.Their role is to teach….If the ethos of the school is solely academic, high achievement, teachers need to be a bit more strict.You cannot be allowing too many resources or time to disaffected students.If your aim and ethos is achievement, then that is what you must go for.It needs to be made clear.If teachers or schools are given clear instructions what's required of them, that would make it easier to do their job.(Staff member, School S).
In School R, staff explained that while there were "occasional and small" efforts to "reward students," the lack of systems to support teachers and a positive school ethos made it difficult for to provide individual help to students who were struggling: What we really should be looking at is what are we going to do to make the kids feel like they're achieving [as individuals]...When the kids do a beautiful piece of work, they are given certificates in assembly, but it's all very desultory.The kids don't actually want to go up and get them because there's no culture of achievement.It's never been built in.(School R) In the higher-excluding schools, a lack of collaboration and communication between teachers, staff distrust and cynicism toward the schools' leadership seemed to exacerbate the pressures that teachers felt from policies.Without the structures and systems to help mediate the demands from the wider policy context, staff had a lowered tolerance and capacity to help students who were struggling in school.In contrast to the staff of School M and School L, the views and attitudes expressed by the staff in School R and School S was that little could be done to avoid and prevent exclusion: In a school where there wasn't such a high concentration of misbehaviour, you'd be able to tackle those problems in school -you bring in therapy, or you bring in remedial action.And you'd be able to punish and reform without exclusion.But in a school like this, you don't have enough time to do all that.And until this school is more stable, then exclusions will be high.(School S) In stark contrast to this conclusion, the staff in the lower-excluding schools pointed to a set of school-level factors that enabled them to prevent and resist exclusion: a headteacher who encouraged a collective sense of responsibility within the school, by delegating responsibilities to staff and promoting teachers to leadership roles; an extensive network of academic and social support structures for struggling students; a closely-knit and highly collaborative staff culture, and a school ethos aimed at valuing students' as individuals.The alignment between these structures, systems for supporting students and teachers, and a strong set of guiding principles were attributed by teachers as having enabled them, as individuals, and as a school to cope with the increased demands from policies and to resist the temptation to use exclusion as a quick fix for challenging issues.
A comparison between the views and attitudes of higher-and lower-excluding schools revealed two contrasting levels of staff capacity to resist and prevent exclusion: ?High-capacity staff.The teachers and senior administrators in these low-excluding schools conveyed, through their descriptions, an empowered way of responding to the increased demands of accountability.Although the constraints on schools and the increased incentives to exclude were felt to pose a negative and unwanted set of pressures, staff members attributed a higher capacity to resist exclusion to specific features and structures within their school, which enabled them to focus and respond effectively to individual students' needs, in spite of the pressures.
? Low-capacity staff.Staff in these higher-excluding schools conveyed a view of exclusion as inevitable and unavoidable.As with the staff in the lower-excluding schools, teachers and senior administrators also acknowledged the pressures of national policies.However, staff in these schools described a lower capacity for responding to the increased demands, and appeared to have very few systems, mechanisms, or structures in place that would enabled staff to communicate and respond more effectively to the challenges and difficulties posed by students with behavioral and academic difficulties.
The differing exclusion rates between the higher-and lower-excluding schools suggest that the type of school organizational culture that enables teachers to prevent exclusion is one that encourages teachers to organize their practices around the internally-developed goals of the schools and the individual needs of the students, rather than have their actions driven by the external measures, indicators, and requirements defined by national policies and accountability framework.This conclusion resonates with Elmore's (2003) observation about the role of school capacity in US accountability systems: Not surprisingly, schools and school systems that do well under external accountability systems are those that have consensus on norms of instructional practice, strong internal assessments of student learning, and sturdy processes for monitoring instructional practice and for providing feedback to students, teachers, and administrators about the quality of their work... High internal agreement is the best defense against uniformed external pressure (p.9).
These differences between the capacity of schools suggests that teachers' efficacy and attitudes within their school plays a key role in their capacity to resist the incentives and pressures associated with exclusion.Evans' (1999) study of teachers' attitudes toward disruptive behavior similarly found that teachers who believe strongly in the effectiveness of the strategies and systems used in their school were more likely to tackle directly, within their classroom, the problems related to behavior, rather than to refer problems to management, which increased the likelihood for exclusion.By contrast, teachers who lack support within a school are more likely to have a decreased capacity --a lowered willingness, confidence, and capability --for coping with the uncertainties of classrooms (Raudenbush et al., 1992).

Discussion & Analysis
The complex pressures of exclusion A major theme that emerged across all four schools was that a combination of policies and changes had contributed to the rise in exclusion.Across subjects, schools, and roles, teachers and principals linked policy mandates they experienced with the "increased pressures" and "greater demands" on schools.These pressures were conceptualized as being externally generated from national policies, which teachers frequently associated with "the government" or also referred to as "the system."Of the forty-four teachers and principals interviewed, the overwhelming majority acknowledged a confluence of multiple pressures as having created greater incentives to exclude.
Although teachers across all four schools suggested that exclusion was a necessary process to have in schools, a view supported by previous studies on teachers' perceptions of exclusion (e.g., Kinder et al., 2000), no respondent believed that exclusion -as an incident, process, or national trend -had a single cause.As one principal explained: [Exclusion] is a complex issue… there aren't any simple answers…a lot of people will say, well, it's the pressures on schools…the league tables…trying to recruit a balanced intake…pressures from the LEA to improve results, from OFSTED [inspection] and that has led schools to exclude the students who are preventing them from reaching the targets that have been set.And all of that's true.This principal's acknowledgement of the interrelated nature of these complex policy interactions was a theme that also emerged in how other teachers described the influence of policy on exclusion.In analyzing the impact of England's educational reforms, Whitty et al. (1998) concludes, "It is virtually impossible to separate out the specific effects of any one of these policies" (p.9).For example, the local management of schools (LMS) was introduced around the same time that other national policy reforms were being implemented.Similarly, the implementation of new requirements for national testing and exams overlapped with continuous changes to the national curriculum and the shift to nationalized school inspection.
The systemic nature of England's accountability reforms thus offers one reason why teachers and principals did not attribute the rise in exclusion to any one policy, but identified the combination of curriculum prescription, testing pressures, high stakes, and competition as having discouraged instructional practices that would benefit low-achieving students.
However, an important aspect to this finding was that individuals' perceptions varied in the extent to which they felt they could resist these negative and unintended consequences, and whether exclusion could be prevented within their own classroom and school context.
Though one might argue that the high exclusion rates are due to failures and limits of teachers' and school practices, and that schools need even more accountability measures to prevent staff to prevent schools from using such practices as exclusion to improve performance.That strategy, however, was in fact, tried by the government.Between 1998 and2001, changes in national and LEA policies and guidelines regarding exclusion directly affected how schools were expected to view and use exclusion.
According to one county education official in the Northern England LEA (where School R and School M were based) the confluence of national policy changes and LEA pressures on schools explains, in part, the decline in school exclusions between 1998 and 2000, and followed by rise in exclusion between 2000 and 2001.(See Table 1).
With the formation of the Social Exclusion Unit in 1998, new government policies to reduce exclusion were being implemented through Circular 10/99, requiring major changes to exclusion procedures at the LEA and school level.Examples of these national policy changes included: 1) stricter procedures for using and reporting exclusion; 3) £3,000 fines to discourage schools from excluding pupils, and 4) financial incentives called "pupil retention grants" to encourage schools to accept students excluded from other schools and to improve support.According to the LEA's exclusions policy officer, these combined pressures, and the introduction of Circular 10/99 "sent a strong message to schools not to exclude".
However, by the end of 1999/2000, the different ways that schools responded to these pressures became apparent through the data.According to the LEA's exclusions policy officer, schools reduced their permanent exclusions by using longer, fixed-term exclusions.This provides one explanation for the 61% increase in the total days lost for fixed-exclusions between 1998/99 and 2000/01.However, at the same time, Circular 10/99, required schools to call a governor's meeting for exclusions over five days, which some schools avoided by either excluding "unofficially" or excluding pupils for only one day, but more frequently.This provides a possible explanation for the increase in fixed-term exclusions between 1998/1999 and 2000/01.
The LEA's exclusions policy officer also explained that the introduction of penalties and £25,000 pupil retention grants in 1999/2000 were also viewed by and used differently in different schools.Rather than be discouraged from using exclusion, some schools began to "allocate" part of their budget and ironically, the grants towards exclusion, setting aside money for the fine that would be imposed when they excluded a pupil.Other schools, however, which had "strong head" and "strong systems already in place" used the pupil retention grants to strengthen and increase their support within the school, increasing staff and assistance.Still other schools used the grants to set up off-site units or to send students to programs based outside the schools, which "didn't necessarily change how staff viewed and used exclusion".Following an OFSTED inspection of the LEA in 2000, which recommended that the LEA began to devolve funds that were previously used centrally to support schools to offer more choice.In response, the LEA reduced its Pupil Support Service team, which previously worked with students that schools had difficulty supporting, asking schools to "buy back" the teacher assigned to their school.According to the LEA exclusions policy officer, this policy change worked favorably for schools that had a strong and positive working relationship with the pupil support teacher.However, schools that did not have a strong teacher "lost out" because they not only spent their money on support they did not want, but they also faced the additional penalty of being fined if they excluded a pupil.

Teachers and the capacity to prevent exclusion
A final theme in this study's findings relates to the dilemmas and conflicts that teachers are likely to experience in the face of growing pressures to raise achievement and increasing demands that accompany a high-stakes accountability system.Here the study pointed to major conflicts that teachers experienced in the context of accountability, including dilemmas about whether to depart from the curriculum to address individual needs and risk not covering material that will be tested; uncertainties about how to cope with a student whose individual behavior disrupts or distracts from the needs of other students more likely to perform well on assessments; and finally, debates about how best to improve student performance.Difficult to capture through the quotes of teachers was the dismay and distress that teachers frequently displayed in explaining the rise in exclusion and the frustration they felt from the pressures of assessment, inspection, and targets.Several teachers described exclusion as a "failure" on their own part, offering some insight and explanation about why the pressures of accountability might have a demoralizing and deskilling effect on teachers.

Conclusion
Exclusion is a complex phenomenon that requires a view that extends well beyond the individual behaviour of an individual student.Understanding the cause and dynamics of exclusion requires a view that takes into account not only how schools view the demands of the wider policy context, but also the internal school context that shapes teachers' capacities to cope with the challenge of low-performing students.While the findings reported in this study are based on perceptions, the use of this approach provided insights that would be difficult to capture without allowing teachers the opportunity to construct their views.The explanations offered by teachers about the national rise in exclusion, points also to the complexity with which practitioners are able to analyze the context in which they operate and teach, and shows how keenly aware teachers can be, if given the opportunity to reflect, of how various policies influence their practices, beliefs, and actions.
The study's main conclusions are: ?Exclusion is influenced by multiple factors, which includes student and social background, school context, and national policies.The study found that teachers did not assign the causes of exclusion to any single cause or policy, but identified a wider set of interactions and factors at the school-and national policylevel.
? Accountability increases the pressures and incentives to exclude students who are low-performing.In the case of England, test-based accountability, in combination with the constraints of the curriculum and the dynamics of choice and competition created dilemmas for teachers --competing tensions in raising achievement and balance such pressures with the needs of students.?School organizational context influences how accountability policies are implemented, how their pressures are felt and mediated, and the extent to which incentives to exclude can be resisted by teachers.The study found that internal structures and systems can exert a positive influence in helping teachers to resist the negative consequences of accountability.The effectiveness of these structures and systems was stronger in the low-excluding schools and limited in the higher-excluding schools.?Teacher capacity plays a role in whether schools can minimize the potentially negative consequences of accountability and respond to students who are likely to be marginalized by the increased demands and pressures.
The study found that the extent to which teachers were able to resist the incentives to exclude depends on their capacity to respond effectively to the needs of students at risk of exclusion, specifically those with academic and behavioral difficulties.
Implications for the U.S.
The study's findings offer evidence of the unintended consequences likely to result from a high-stakes approach to accountability.Whether teachers are able to minimize these consequences and the incentives to exclude depends not only on skills and abilities of teachers, but more crucially on the decisions and structures made internally by the school, which can either hinder or strengthen the staff from developing the internal capacity to respond to those students with the greatest risk and needs.
Under what circumstances, conditions, and policies can schools and teachers resist the pressures likely to accompany high-stakes accountability systems?The question for policymakers to consider is whether increased pressures and higher standards for performance will make it even harder for at-risk students to do well in school, especially if teachers are constrained by the prescription of curriculum and the demands of testing.McDill, Natriello, and Pallas (1985) warned, nearly two decades ago, in considering the effects of standards-based reforms on dropouts, that the potential negative effects of higher standards, without providing students additional assistance, would be greater academic stratification, fewer student options made available to them, and increased time requirements on the part of schools and teachers, which would conflict with the demands on students.The suggestion that "alterable characteristics in schools" could minimize the risk of these unwanted effects is one confirmed by this study, and the features of lower-excluding schools (p.157).
While the intent behind England's reforms may indeed have been about "raising standards," the construction of a high-stakes accountability system, layered onto the dynamics of parental choice, does not appear to have provided a more equitable entitlement for students, and is now being called into question (Olsen, 2004).Rather, there is evidence that instead of raising expectations for all students, England's system has ignored those at the bottom and highest levels, and has encouraged schools to focus their efforts on those students most likely to achieve the target -those right below the bar -a strategy that enables a school to improve their rankings on league tables (Olsen, 2004).One set of studies has found evidence of schools increasingly resorting to tracking and streaming over the last decade (Boaler, Wiliam, and Brown, 2001;Ireson, Clark, & Hallam, 2002).Boaler et al., (2001) concluded that the "traditional British concern with ensuring that some of the ablest students reach the highest possible standards appears to have resulted in a situation where the majority of students achieve well below their potential."This construction of failure, the polarization of lowachieving students, and the likelihood that schools in the US will resort increasingly to tracking and ability grouping to improve overall test scores (a practice that has been found to have disproportionately negative effects for Latino and African-American students (Oakes, 1987)), is a scenario that policymakers might consider, as increasing the risks for exclusion.
What are the lessons for educators?As the pressures of increasing accountability bear down upon states, districts, and schools, much will depend on the capacity of teachers to respond in ways that will minimize the effects on low-performing students.It will be important therefore for practitioners to build the capacity of teachers to meet the increased demands of accountability, and to strengthen internal practices and processes that will help prevent students from dropping out and becoming discouraged by increasingly high performance targets.For example, districts and schools might consider alternative or complementary ways of assessing students, expanding the support and resources available to teachers to identify and address students' individual needs, and expanding the support structures for students struggling in school.
What are the implications for future research?The study's findings point to the need to understand the complex interactions that accountability policies will have with other aspects of a state's educational system, especially with regard to policies on testing, graduation, and choice.With respect to future studies on exclusion, the study points to the role of school context in mediating how accountability policies are implemented and in potentially minimizing the negative consequences of accountability and its pressures.While large-scale studies might look to macro-level trends in student achievement, dropout, graduation, and other such indicators as evidence in judging the effectiveness of various accountability systems, further investigation at the school-and teacher-level will help either to confirm conclusions, or reveal a different picture.The perspectives of students could also offer an important perspective on how pressures from high-stakes policies affect their motivation, confidence, and capacity to do well in school.
Where is England now?After a decade of a high-stakes approach to accountability, the wisdom of its post 1988 reforms is being called into question.Suggestions that the government might revamp the tests and raise the bar even higher in response to the rapid test gains of 11-year olds, which have now reached a plateau, have led to the National Union of Teachers threatening to boycott national tests and exams (Olsen, 2004).Following Scotland's devolution and creation of an independent Scottish Parliament, Scotland immediately abolished its policy of publishing student performance results in 1999.With the establishment of the Welsh Assembly, Wales ended the practice of league tables in December 2003, concluding that the practice had long damaged schools and teachers.A series of reports have proposed that England follow suit (Audit Commission, 2003;IPPR, 2003).A review of the National Curriculum has resulted in proposals to reduce the amount of testing (QCA, 2000) and to incorporate more local, teacher-led assessments and OFSTED, under new leadership, in trying to move away from its externalized inspection system appears to be shifting toward school self-evaluation as a way to motivate school-based improvement.
The findings from this study provide not only important insight to the dynamics of exclusion but more crucially, to how and why accountability, in combination with the pressures of testing, curriculum, and other mechanisms and measures used to judge schools, is likely to create incentives to exclude students who are low-performing.The cautionary tale from England is that while a high-stakes approach to accountability might motivate schools and teachers to pay greater attention to the measures by which they will be judged and the curriculum that is tested, such a system, by its very nature, creates a climate of teaching and learning that profoundly changes how low-performing students, who pose an increased risk to a school's performance, are viewed.As states across the US pursue accountability and assessment systems that will greatly increase the pressures on schools and teachers to raise levels of student achievement, the research and policy community should look carefully at the impact that such pressures, and the increased stakes attached to them, will have on students as well as their teachers.
While an education policy might set out in theory to ensure that no child is left behind, however, what students learn and achieve does not occur through policy alone, but depends fundamentally on the capacity of schools and teachers.Academic rigor and college readiness -the current mantras of high school reform in the United States --will not result through the rhetoric and will of policymakers, but will come about because of the efforts of schools and highly skilled teachers to deliver instruction and curriculum in ways that will enable each and every student to engage in an process of meaningful learning, within a caring and personalized environment.
g., influences from national educational goals, reforms, and policy initiatives) PROFESSIONAL CONTEXTS (e.g., influences from professional associations, teacher education programs) HIGHER EDUCATION INSTITUTIONS (e.g., influences from standards for admission and achievement) PARENTS AND COMMUNITIES (e.g., influences from social class and culture) LOCAL SCHOOL DISTRICTS (e.g., influences from local educational authority) SCHOOL ORGANIZATIONAL CONTEXTS (e.g.influences from local school practices) " Key Stage 1 (age 7); Key Stage 2 (age 11); Key Stage 3 (age 14) and Key Stage 4 (age 16).The final stage, Key Stage 4, is assessed through a culminating series of exams, called the General Certificate of Secondary Education, known as the GCSE, which comprises a series of content-specific examinations that are somewhat similar in purpose and theory to high school graduation exams.

Table 2 School Sample Profile of Low- Excluding Schools Profile of High- Excluding Schools
the past you could accommodate the needs of the individual and the rest of the group.But now with the constraints of the curriculum, teachers find it affects what they are doing and the learning of the other pupils in the class ….You have got balance.So if you have a child in Year 10 [equivalent to junior year in American high schools], where children are starting to get involved in exam courses, you have got to balance the needs of a child who has a problem and causing a fuss with the wider audience in the classroom.It is something I am trying to tackle.Pupils have behavioural problems, which need addressing, but I see the wider needs of the group (Staff member, School R) Year 11 you see that tolerance diminished rapidly.Pupils are stressed.Teachers feel the pressure.No one likes it.At our school, we even call it "exam confusion" because students get confused with all the deadlines, the coursework, and the schedule for preparing for exams.Everyone gets aggressive, panicked, and upset . ..and it comes out in pupils' behavior and attendance.Pupils, who can't cope, they act out, they bunk off, they don't come to school.We try hard not to exclude, we really do, and we try to sympathize with what the pupils must go through.(Staffmember, School S) This teacher's account of what happens in his classroom because of the increased demands of national assessments, illustrates and points to the particularly negative effects that testing can have on students who are struggling in school.In other schools, staff offered similar explanations of the effects on students who did not perform well on tests and who felt deeply discouraged by being labeled by their test results.One teacher explained that "the fear of failing GCSE exams made some [students] want to give up … and so then … exclusion becomes an option for everyone."Anotherteacher felt that the national target of achieving 5 A*-C's was particularly daunting for students who were struggling academically, particularly those with special educational needs: "[The national target] makes it hard for [pupils with special educational needs] to feel confident and positive about taking exams."Pondering the link between exclusion, behavior, and the pressures of testing, one teacher wondered aloud whether students' behavior had changed, and then concluded that the increased demands on schools had changed what teachers were willing to tolerate: Over the years, I wonder if[student's]behaviour has got more difficult, or if their behaviour has changed over time.I think perhaps it hasn't, but I think there is a greater expectation over what we are expected to do.This is to produce results to improve A-C numbers on the GCSE exams, to increase your entry of GCSE results … that is true.(Staff member, School L)