Introduction

Imagine that you are currently a student with a disability or you come from a low-income minority household. You attend an urban school in New Delhi, Johannesburg, Bogota, London or New York. You have been targeted as a member of a group of students at-risk for failing school. You and your group must meet grade/age level expectations for academic achievement as measured on a standardized test such as PISA or TIMMS.Footnote 1 You take the test, along with other students in your group, and do not achieve a passing score established for your school district or local education agency. As a consequence, your school is labelled as failing, and your entire school is penalized. Now imagine you are a teacher of the students in one of these schools. You know that your students have made excellent progress, are motivated and have responded successfully to the adapted and culturally relevant curriculum that you worked hard to develop. Yet, according to the educational standards set by government policy in your country, your students have been marked as “failures”. This scenario has been repeated for tens of thousands of students and their teachers in countries with high-stakes testing—the United States among the most prominent. In this paper, we argue that it is government policies that have failed. The No Child Left Behind (NCLB) education law enacted in the United States in 2002 and legislation like it in other countries: (a) assumes that all students must meet the same standards; (b) fails to recognize individual differences, talents and achievements; (c) promotes a culture that blames, stigmatizes and excludes students and their teachers; and (d) establishes mechanisms that all but guarantee segregation, retention or dropping-out of school (Box 1).

Box 1 The market-based view of standardized test scores
Full size table

The consequences of this market-based view have undermined the goals of closing the achievement gap for under-performing students that accountability measures were meant to accomplish. In the following sections of this paper we critically analyse the consequences of this view, highlighting the United States’ education policy as an example. We then contrast and compare an alternative view of education based on the social justice goals of inclusive education. From this alternative view, we then present a school-community framework for action, with examples from selected countries that have achieved equity with excellence. This alternative approach highlights the ways in which inclusive education can close the achievement gap among students in schools where educational high-stakes testing policies have failed. While these contrasting approaches to education are presented here as two distinct views of schooling, our purpose is to clarify the different types of claims that can be made for school success and failure. We recognize that these two approaches can be interpreted in various ways and ultimately may not be mutually exclusive. However, providing transparent contrasts between the two views is needed in order to clarify what education stands for and whom education claims to serve (Slee 2007: 165). The case of NCLB in the United States and comparisons/contrasts with other countries hold lessons for many other countries that have adopted market-based educational reforms.

Assumptions driving high-stakes testing

The goals and mandates of high-stakes testing are driven by a market-based view of education that values competition, productivity and individual achievement. A market-based view of education uses a system of accountability that reflects a shift from inputs and processes to outputs and performance. Several assumptions concerning schools underlie this view of education. These assumptions influence the behaviour of teachers and administrators in schools, and result in particular consequences. These are summarized below in Table 1 and are accompanied by a discussion concerning the negative consequences of these assumptions.

Table 1 Market-based assumptions regarding schools
Full size table

Market-based assumption 1

The measure of productive citizens, according to high-stakes assessment policies, is achieved by obtaining proficiency levels in academic subjects of reading and mathematics. The standardized tests that have been developed to measure this proficiency rely on multiple-choice tests, with a lack of higher-order thinking and problem-solving measures. However, in the most recent PISA assessments, the United States ranked nineteenth out of forty countries in reading, twentieth in science, and twenty-eighth in mathematics (on a par with Latvia), being outscored by nations like Canada, Finland, Hong Kong (China), Japan, the Netherlands, the Republic of Korea, Singapore and Sweden (Darling-Hammond 2007). In contrast to the United States, these countries focus curriculum and testing on students’ abilities to apply what they know to new problems and situations. This alternative focus on curriculum and testing is supported by what is known about current workforce requirements. According to the United States Department of Labour, workers in a global economy will need the following skills: (a) interpersonal skills, including participation in teams; (b) acquiring and using information; (c) understanding complex interrelationships from a systems perspective; and (d) working with a variety of technologies (Levin 1997: 4–5).

Further, the higher scoring countries on PISA are characterized by a focus on the following substantial inputs to support achievement (Darling-Hammond 2007):

  • providing high-quality universal pre-school and health care for children.

  • funding schools centrally and equally, with additional funding to the schools with the greatest need. (In the United States the wealthiest school districts (local education agencies—LEAs) spend nearly ten times more than the poorest, and spending ratios of 3:1 are common within different states in the United States (Kozol 2005: 244–246, 321–325; Darling-Hammond 2004: 6).

  • intensive support for a better prepared teaching force—funding competitive salaries and high-quality teacher education, mentoring and on-going professional development at government expense. By contrast, the United States is undergoing a crisis in the teaching profession. Low salaries and poor working conditions in many schools are causing high teacher absenteeism, significant numbers of new teachers who leave the field and critical shortages of qualified teachers in the urban schools that need them most (Darling-Hammond 2007).

Market-based assumption 2

According to Levin (1998), there is a weak relation between test scores and adult earnings. In fact, each additional grade/age level completed in school is associated with four times as large a gain in earnings post-school as an additional grade/age equivalent test of basic skills. Completion of the last year of secondary school is associated with ten times the increase in annual earnings of an additional grade/age level equivalent of test-score gain. Significant evidence from research studies points to a narrowing of the curriculum and teachers who are abandoning effective instructional practices to teach to the test (White and Rosenbaum 2008).

Market-based assumption 3

According to Fusarelli (2004), a report by the American Federation of Teachers found that only twenty-nine states in the United States have clear and specific standards in core academic subject areas. In addition, there is growing evidence that these states are lowering their testing standards and manipulating proficiency scores in order to avoid failure to meet annual yearly progress (AYP).Footnote 2 For example, the state of Arkansas, with considerably lower standards than other states, has no “failing” schools, whereas the state of Michigan reported more than 1,500 such schools in 2003. Michigan’s response was to redefine AYP by lowering the percentage of students required to pass state-level exams from 75% to 42%, reducing overnight the number of failing schools from 1,500 to 216 (Fusarelli 2004: 82).

Several states in the United States have also abandoned tests that require demonstrating mastery through authentic assessment, preferring machine-scored multiple-choice tests that are less expensive to score (Toch 2006). All of these behaviours make test results virtually meaningless and have led to lowered standards of performance.

Market-based assumption 4

Reports from several states in the United States indicate that higher numbers of failing schools are reported every year, mostly in urban areas (Swanson 2008). Qualified and experienced teachers are migrating out of these schools, creating a fundamental and inequitable lack of opportunity for students to achieve. There is also a fundamental problem that 100% proficiency is impossible on norm-referenced tests (when by definition 50% must score below the norm).

Students identified as disabled receive special education services precisely because they have not been achieving to expected levels of academic performance. Yet, in the United States, these students must also become proficient at grade/age level in order for a school to achieve the status of sufficient annual yearly progress. NCLB legislation in the United States makes no distinction between a school that barely misses AYP (and therefore is considered “failing”) and schools that fail dramatically on multiple measures. An increasingly common cause of failing schools is the failure among students with disabilities. For example, Fusarelli (2004: 78) reports that a school in the state of Connecticut failed because eight special-education students out of forty-five in their school failed to attain proficiency on the mathematics test.

High levels of school drop-out rates for low-achieving students have been reported (Swanson 2008). The United States has not improved secondary school graduation rates for twenty-five years, and graduation rates are now declining as requirements for an educated workforce are going up. Specifically, Darling-Hammond (2004: 21) reports that the graduation rate of 71% for African-American students in 2002 dropped to 59.5% in 2003. In this same report, individual states showing the steepest increases in test scores have the highest retention and school drop-out rates. The “Texas Miracle” was accomplished when a 9th Grade class (age 14) of 1,000 students decreased to fewer than 300 students by the 12th Grade (final year of secondary school). The miracle is that not one school drop-out was reported (Darling-Hammond 2004: 21).

At the same time, there has been a large jump in the number of students identified with disabilities. Specifically, one state reported three times as many 3rd Grade students (age 8) and six times as many 6th Grade students (age 11-12) have been classified as in need of special education since accountability policies were put in place (White and Rosenbaum 2008: 102). In identifying students, a new operational definition for learning disabilities (LD)Footnote 3 has emerged: students are labelled as LD if they have such low test scores that they seem unlikely to meet the established proficiency level by the time of the next test (White and Rosenbaum 2008: 104).

All of these assumptions about schools have corollary assumptions about society. These assumptions are provided in Table 2 below, along with evidence of their faults.

Table 2 Corollary market-based assumptions regarding society
Full size table

Market-based society assumption 1

Inequalities in United States schools are widespread. For instance, the differences in the states of Michigan and New York are typical. Kozol (2005: 321–325) reports that in the urban school areas of Michigan, one local education agency spends US$12,825 per student, with only 2% of their students classified as low-income. By contrast, public schools in the inner city of Detroit spend US$9,576 per student, with 59% classified as low-income. School funding in the New York City area includes the school district (LEA) of Manhasset, which spends US$22,311 per student compared to US$11,627 per student in New York City itself. The proportion of low-income students in Manhasset is 5%, compared to 83% in New York City. Race is also affected. Manhasset has a population consisting of 9% Latino and African-American students. New York City has a 72% population of Latino and African-American students. In the Detroit area, Bloomfield Hills LEA is 8% African-American/Latino, and Detroit is 95% African-American/Latino (Kozol 2005: 321–325).

Market-based society assumption 2

According to a recent study conducted by the RAND Corporation (2005), a substantial percentage of children in the United States are disadvantaged in terms of resources available for healthy physical and mental development. One-fifth of children under age 6 live in poverty and nearly half of all children face one or more risk factors associated with inadequate school preparation. These disadvantages translate into low academic achievement, poor social behaviour, lack of educational attainment and, eventually, higher rates of unemployment and criminal activity.

Market-based society assumption 3

According to Skrtic (1995), a “machine” bureaucracy treats schools as business assembly lines by imposing educational policies to establish conformity and control over the schools. However, he suggests negative sanctions only serve to increase the teachers’ ability to “play the machine bureaucracy game”. While this may satisfy policies and standards, there is little or no focus on serving the students, even though the education of students is commonly believed to be the ultimate reason schools exist.

Market-based society assumption 4

The United States Office of Civil Rights (OCR) and the Department of Education have been tracking over-representation of minority children in special education programmes for many years. In 2002, in response to widespread concerns, the Civil Rights Project at Harvard University undertook an extensive study that produced the following findings (Losen and Orfield 2002: xv-xvi):

  • In the 1980s, African-American children constituted only 16% of total school enrolment, yet they represented 38% of students in special education programmes for students labelled as educationally mentally retarded.

  • In 2000, African-American children constituted 17% of total school enrolment, but 33% of those labelled as mentally retarded.

  • During the same period (1980–2000), students with emotional disturbance (ED) were over-represented in special education programmes, and the rate of identification of both ED and specific learning disabilities (SLD) grew significantly for African-Americans.

Market-based society assumption 5

Nieto and Bode (2008) suggest that practices involving tracking and segregation of students can have devastating consequences on children. Foremost, once a student is placed in a lower track, it may prove virtually impossible to move out of that track—there is often little or no mobility between such groupings throughout an individual’s entire educational career. In addition, students are aware when they are being tracked into lower- or higher-level ability groups. Consequently, students may begin to believe that the labels associated with such groups, like “smart” or “dumb”, give an accurate picture of their personality and ability. Lastly, students who are tracked into lower-level academic classes and programmes often have the newest and least academic-content-prepared teachers; whereas more experienced educators are often given advanced-placement classes in which discussion and higher-order thinking skills are emphasized.

Summary

The assumptions inherent in this market-based view of education treat schools as “black boxes” isolated from societal influences, i.e. poverty and lack of adequate funding. Further, schools are expected to behave like machine bureaucracies, providing standardized and homogenized curricula and producing equally proficient “productive workers”. This view of schools and schooling is illustrated in Fig. 1 below.

Fig. 1
figure 1

Machine bureaucracy (black box) model

Full size image

The machine bureaucracy “black box” model depicted in Fig. 1 represents the inputs, outputs and consequences of high-stakes educational policy impacting a variety of students, particularly those with disabilities.

In this model, the primary inputs of educational policy for accountability through high-stakes testing require schools to standardize tests, teacher qualifications and academic curricula. These requirements are passed down or “inputted” into the schools, which are represented in this model in the form of a “black box”. Essentially, there is little or no attention paid to actions occurring inside of the black box in a machine bureaucracy. Instead, the primary focus is on the inputs (policies and standards) and output (results). Within these outputs there are specific outcomes expected from the students, teachers and schools. The expected outputs include proficient students and productive citizens, which can be measured by 100% academic content mastery, test scores, graduation from secondary school and employment rates.

These expected outputs have some significant unanticipated consequences that take the form of both student and school failures. Student failure occurs when students lack a variety of basic academic skills. Consequently, students do not become “productive citizens” as defined by this type of system. An additional consequence of school failure involves educational policies tightening control at all levels of the education system. In addition, teacher decision-making may also be lost because teachers feel pressured to teach to the test in order to increase students’ test scores. This kind of high-stakes testing will inevitably lead to increased inequalities among schools. Ultimately, if schools and students continue to be forced to function in a machine bureaucracy “black box” education system, a failure of the school itself, the community and society at large may be the ultimate outcome.

Essentially, educational policies that use high-stakes testing for accountability are based on standardizing or “normalizing” students. These policies fail to recognize individual differences, talents and achievements of students—especially those from minorities and those with disabilities and special education needs. In addition, high-stakes assessments promote standardization of education for all students while encouraging a school climate that stigmatizes and excludes students (as well as their teachers), reinforcing an education system with the potential to increase segregation and ability tracking, as well as school drop-out rates. In this type of system the processes of teaching and learning that occur within schools are significantly neglected. Skrtic (1995) refers to this type of system as a “machine bureaucracy” in which schools are responsible for students’ academic achievement and teachers are expected to produce the desired outcomes. In this view of education, no attention is given to the processes of teaching and learning involved in obtaining such outcomes. Instead, a system of punishment functions to control and regulate those schools, teachers and students who are considered “successful” or “unsuccessful” in meeting pre-established standards for what counts as learning.

A twin-track quality, social-justice and equity approach

In contrast to the educational policies of high-stakes testing and its market-driven assumptions of schooling, inclusive education is a social-justice view of education that values co-operation, equity and diversity. According to UNESCO’s Guidelines for inclusion (2005), inclusion may be viewed as “a dynamic approach of responding positively to pupil diversity and of seeing individual differences not as problems, but as opportunities for enriching learning” (p. 12). While the “machine bureaucracy” has traditionally focused on the problems often associated with diversity, inclusive education emphasizes the potential value diversity brings when all students are provided with an equal opportunity to be educated appropriately. In a “machine bureaucracy” model students of different needs and backgrounds may be integrated into a school system in which students are expected to conform to the standards and behaviours of the majority. Conversely, inclusive education calls for the acceptance and respect of variations within a school community (Bourke 2009). This model not only implies that all students physically belong in the same learning environment, but that each individual has the right and opportunity to actively participate in a community of their peers. Essentially, the inclusive education model shifts the focus of educational policy and practice by placing the problems associated with diversity within the social institution, rather than blaming individuals for inherent variations in areas like ability, race, ethnicity and gender.

Bing (2008: 116) describes effective schools that promote inclusive education and display certain common characteristics for improving quality and equity, described in Box 2.

Box 2 Characteristics of effective schools for inclusive education
Full size table

An international comparative study in Latin America found that Cuba was the highest scoring country in both reading and mathematics (Bing 2008: 118). Cuba’s education system is characterized by a child-centred curriculum with a focus on instruction and a supportive social environment. An analysis of PISA 2006 standardized scores found that differentiation or tracking at an early age damages equity without improving quality. For example, Poland discontinued tracking and increased that country’s scores significantly between 2000 and 2006 (Bing 2008: 119). As a high-performing country in the PISA survey, Finland is able to achieve both equity and quality through comprehensive schooling that displays many of these same characteristics.

These studies and their findings provide numerous implications for how inclusive education can be put into practice. In order to accomplish inclusion for all, change in both the beliefs and practices of many current education systems will need to occur. According to UNESCO (2005): “It involves changes and modifications in content, approaches, structures and strategies, with a common vision which covers all children of the appropriate age range and a conviction that it is the responsibility of the regular system to educate all children” (p. 13). Ultimately, the implementation of inclusive education for all requires a strong commitment by school systems and society as a whole that every child has the right and ability to learn as a member of a community of their peers. Through collaboration and shared responsibility, social justice can reach diverse populations of children around the world that have historically been unable to access or benefit from educational opportunities.

The goals of equity and equality that drive a social justice view of inclusive schooling contain three basic principles (Slee 2007). First, a deliberative democracy confronts complex social issues that demand extensive and inclusive participation in decision-making. Second, schools bring students to the centre of institutional life, rather than the “ineluctable drive to differentiate students as projections of measurable outcome” that is central to exclusionary practices of market-driven schools (Slee 2007: 167). Third, inclusive education is not an end in itself, but a means for achieving an education in and for democratic citizenship. The explicit assumptions inherent in these three principles of inclusive schooling are described in Table 3 against those of market-based education (Table 4).

Table 3 Comparison of market-based and inclusive education assumptions regarding schools and learners
Full size table
Table 4 Comparison of market-based and inclusive education corollary assumptions regarding society
Full size table

A school-community framework for inclusive education

We propose a school-community framework for inclusive education that is especially effective for schools and that prevents the negative consequences of high-stakes testing policies. This framework for inclusive education recognizes students’ differences as an asset—not a problem. Essentially, we imagine schools of the future as places that provide a culture of inclusion and governments that are supporters, establishing policies to promote the success of students with disabilities and special education needs, their teachers and all students together.

The school-community inclusive education model shown in Fig. 2 may be seen as a strong alternative to the “machine bureaucracy” model of education. In contrast to a top–down approach to policy, the school-community inclusive education model provides flexibility and collaboration based on a culture of community. The design of this model shows how all of the parts work together through continuous movement within and between its parts.

Fig. 2
figure 2

School-community inclusive education model

Full size image

In this model the student is at the centre of the school. All students are included—especially those with disabilities and those who have been traditionally marginalized, labelled, segregated and tracked. This school culture celebrates diversity, recognizing all students as contributing members with specific strengths and talents.

The community circles around the school (shown by the schoolhouse picture in the centre). Any individual who is considered to be either directly or indirectly affected by the way in which a school functions is a school-community member. The responsibility for student, teacher and school success depends on the efforts of school-community members, including teachers, school administrators, parents, school council members, policy-makers and local business people.

In order to create a successful school community, all members of the community must be willing to work together with a common inclusive education vision of education, high expectations, equitable distribution of resources, co-operative learning, strength-based approaches to teaching and learning, culturally responsive teaching, rich curricula and authentic assessments. These actions and beliefs are self-renewing processes. If the members of the school-community work together to achieve common goals, then positive results become an expected outcome.

The school-community inclusive education model is based on individual and community accountability that involves much more than just high-stakes testing. Instead, accountability takes the form of the community’s sense of responsibility for achieving a common goal based on the shared interests of the entire community. Inclusive education schools need to be flexible and self-renewing, which can only be achieved by everyone working together. An important component of this inclusive education school-community model is that innovation may be required. Because the most effective strategies for achieving common goals may be different for each school, it is essential that community members constantly interact and work together to develop new ways of achieving shared goals.

Several successful school-reform movements provide examples of the basic assumptions in this model. To provide cross-national comparisons, we briefly highlight three specific reforms here, although there are many more that could be cited. In the United States, the Accelerated Schools Project (ASP) has experienced significant success since it was introduced in the 1980s. ASP goals were to “eliminate remediation by using all of the schools’ resources to accelerate the growth and development of all students in order to bring them into the academic mainstream by the end of elementary school” (Levin 1997: 390). ASP focuses on schools with high numbers of at-risk students, typically placed in special education programmes. The reform uses a set of practices based on a philosophy with three basic principles: unity of purpose; empowerment with responsibility; and building on strengths. A major component of ASP includes parental and community involvement. Evaluations of these schools “have shown substantial gains in student achievement, attendance, full inclusion of special needs children in the mainstream, parental participation, and numbers of students meeting traditional gifted and talented criteria” (Levin 1997: 392).

The second example of a school-reform movement that reflects the school-community inclusive education model of schooling is Finland’s comprehensive school. The Finnish education system is flexible, and streaming and homogeneous ability tracking do not exist. Finnish comprehensive schools are characterized by: teachers who differentiate their instruction in response to students’ needs and who co-operate closely with parents or guardians; available guidance and counselling for all students; services to support the physical health and psychological health of all students; remedial teaching for students temporarily lagging behind in their studies (Halinen and Jarvinen 2008: 90).

Students in Finnish comprehensive schools outperform all OECD countries on various measures of academic achievement. But, more importantly, these students have smaller variations in performance among students, achieving high quality and equality at the same time (Halinen and Jarvinen 2008: 78).

Finally, Escuela Nueva schools were created in the 1970s, beginning in Colombia. They are models of education targeting rural multi-grade schools in low-income communities. Escuela Nueva philosophy involves a grassroots approach to promoting the success of students and teachers in rural low-resourced schools. It has achieved phenomenal success due to: (a) its child-centred, self-paced, culturally relevant curriculum; (b) emphasis on communities of practice and community participation; (c) focus on schools and teachers as the unit of change; and (d) commitment to social justice. The curriculum and materials are adapted to the local context to meet the needs of the students and teachers. “The program was not meant to be static, but rather an ongoing exchange between students, teachers, and the community” (Kline 2002: 172). All of these participants have ownership over programme development. For example, students take leadership roles through a student government structure and make decisions regarding their school environment. This involvement teaches civic and democratic values through participation and action. Students interact with their communities and invite community members to the school to share their knowledge and expertise. “In this way, the Escuela Nueva model also helps to break down divisions between schools and the larger community, facilitating implementation and increasing effectiveness of the education reform” (Kline 2002: 173). Evaluations of this school model have shown that students in Escuela Nueva schools score significantly higher than students in traditional rural schools in core academic subject areas, as well as on measures of civic values and self-esteem. Significantly higher rates of school completion have also been accomplished. Several countries have adopted this model with similar successes (Kline 2002).

Conclusions

The obsession with high-stakes testing and their results has led governments to claim improvements by students overall, while inequalities and achievement gaps continue to manifest themselves with the sorting, tracking, segregating and excluding of certain racial, minority, low-income and disabled students. Governments that use high-stakes tests are characterized by high levels of centralization, low per-pupil funding, high numbers of at-risk students and low-income families. It is thought that high-stakes tests can raise achievement in the lowest performing schools. However, governments that have adopted high-stakes assessment policies have fared worse than those with no or low-stakes testing programmes. “Using the best external measures available, evidence exists that high-stakes tests do create negative, unintended consequences. […] The adverse consequences of high-stakes tests appear to outweigh what few benefits such tests may have” (Amrein and Berliner 2002: 48).

By contrast, governments that have achieved equity and quality are characterized by decentralized education systems that provide local autonomy responsive to community contexts with additional funding to the schools with the greatest needs. They have invested heavily in high-quality teaching, with a focus on differentiated instruction. The goals of schooling have focused on education for a democratic society and lifelong participatory citizenship in collaboration with communities, valuing all students in the process.

The poor outcomes and negative consequences of market-driven school systems under high-stakes testing mandates in the United States and other countries, as opposed to alternative successful comprehensive school-community based approaches, provide educational policy-makers with very different choices. These choices invite educators and policy-makers to imagine and consider the impact of different views of education. It will also be necessary to invite new stakeholders in the civil society into decision-making processes in order to create new directions for educational reform.

As modern technology allows for enhanced communication and information-sharing options, members of the global community have a greater opportunity than ever before to collaborate and share the responsibility for providing children around the world with the opportunity for inclusive education. However, in order for successful implementation of inclusive education to occur in both policy and practice, there must be a commitment to the belief that inclusive education can provide all students with a more equal opportunity to actively and productively participate in the communities to which they belong. The question that remains is: how will countries with market-based education policies and legislation currently driving educational reform heed the lessons of their failed policies—and respond?