As governments around the world are struggling with the dilemma of providing an education system that will meet the future social and economic needs of its country the charter school reform is gaining momentum. The Wall Street Journal described the movement as "the most promising of reforms" because of the potential to resolve the school autonomy struggle in a way that traditional site-based management has not (Bierlein and Mulholland, 1994).

The growth of the charter school concept in countries such as Canada, the US, New Zealand and Australia has resulted in much debate and discussion. Each country has approached the charter school movement differently but in all cases the introduction has been linked to the political ideology of the 'new right', which places the majority of research in this area in an interesting situation. If findings from a research study are positive, the researcher can be accused of bias and be seen as supporting the policies of the 'new right'. If the findings do not support the charter movement, the researcher may be viewed as being critical of the 'new right' thinking on education. An example of this is the Wells et al. (1998) research report Beyond the Rhetoric which was viewed by the Institute for Educational Reform as slamming California's charter school movement. The Canadian Teachers' Federation report, Ten Charter School Myths (1999) is critical of all research to date into school charters for what appear to be ideological reasons. The linking of the charter concept to a political platform often means any possible advantages of such an approach are overlooked.

A review of literature identifies a confusion of language, definitions and contexts, which make interpretation of, and discussion on, school charters problematic. Taking a non-political view, this paper seeks to establish a typography to provide a framework for more informed discussion and debate on charters. Research in Victoria, Australia, New Zealand and the United States is reviewed to establish what impact the charter school movement has had and the issues that emerge in relation to competition, community ownership, management, student learning and accountability.

The Larousse Dictionary defines charters as the articles of incorporation of an organisation or company, which usually include a statement of principles or objectives. The charter confirms that a company is qualified to practise and lays down its legal jurisdiction, within which it has great freedom to operate but outside of which it is barred from practice. The Macquarie Dictionary defines a charter as a written instrument or contract, a written document granted by a legislature giving privileges, rights, the benefit of a new invention.

The definitions are appropriate when applied to schools. School charters define the area of operation of the school and articulate what the school offers, how the school will operate and the outcomes it will seek. The charter is more binding than school aims and objectives. Where school aims, objectives or desired outcomes can be changed at a whim, changes to school charters require the agreement of all parties involved in formulating and approving the charter.

The term charter can also infer hire. In this context, the school community is the company hired by the government to provide an agreed service at an agreed price. Funds provided are in effect a fee for service and the customer paying for the service, in this case the government, has a legal right to demand quality for service and to hold the governing board accountable for the school's achievements. The broad charter concept has been modified to meet the purpose and cultural context in each country.

In New Zealand the charter was defined as a contractual arrangement with schools. The charter was a centrally prescribed document which required a description of the community (provided by the trustees), guiding principles including the overall aim of education in New Zealand, the goals for curriculum, equity, educational opportunity and the Treaty of Waitangi. The charter included details of review and codes of conduct for the trustees, the principal and teaching staff. The document was considered to be a legal agreement between the school board and the government. Opportunity for schools to add additional school goals and objectives of their own if they so desired was included, although in reality the centrally prescribed goals limited individuality. The prescription of the charter ensured uniformity between all schools. The emphasis on equity which included gender, disability and the Treaty of Waitangi, were viewed as safeguards for the national education system which reflected government policy for promoting change in these areas (Mansell, 1992). The charter and the processes for policy writing were to regularise what had been an individual and varied approach to policy writing in schools.

The Ministry of Education meets its requirements for the provision of universal education by contracting out the service to educational providers. While traditionally the program had been applied to former government operated schools, a number of private schools have 'opted in' to the charter program, reducing the difference between public and private providers while at the same time creating an option for private enterprise to also become involved. The clear contractual brief was to ensure that the content and quality of outcomes is controlled and monitored by the government.

In the United States a charter school is essentially a public school created and operated under a charter or contract negotiated between the organisers who manage the school and the sponsors who oversee the provision of the charter (Bierlein and Mulholland, 1994/5). The concept of charters was to free the schools from restrictive legislation, by providing what Bierlein and Mulholland call an 'institutional bypass' around the status quo. In general charter schools are given exemption from most state and school district rules with the exception of health, safety, civil rights, audit, student assessment, special education requirements, insurance and pupil assessment.

The charter is negotiated for a set time and will only be renewed if the school achieves its goals, has not violated laws or grossly mismanaged its affairs and continues to attract students, parents, and teachers. The length of charters varies between states and is generally up to five years. Charter schools are legal entities with the ability to hire and dismiss, sue and be sued, award contracts and control their own finances. Funding is based on student enrolment with 80-90 percent of funding coming directly from the state allotment to the local school district, freeing the school from state and district regulations.

The charter school movement in the United States is limited by state controls where the number of schools who may become charter schools is limited. For example the number of students in charter schools in Massachusetts cannot exceed 0.75 of 1 percent of public school students. Only two states, Hawaii and Wyoming, have no limitations on the number of charter schools that can be established.

In Victoria the charter was to define the area of operation of a school; to articulate what the school offers, how the school will operate and the outcomes it will seek. The school charter required the agreement of all parties (the school, the school council and the Minister of Education) involved in its formulation and approval. In this context, the school council became the company hired by the government to provide an agreed service at an agreed price. Funds provided were, in effect, a fee for service. The customer paying for the service, in this case the government, has a legal right to quality for service and to hold the school council accountable for school achievement.

The development of the school charter was based on information contained in the preliminary document, Schools of the Future (Hayward, 1993). Requirements were distributed throughout the paper and included an agreement between the minister and the school (Hayward, 1993). The charter process was to involve parents in deciding how the charter would contribute to achieving the school's objectives, allowing them, as members of the community, to determine the future of their school (Hayward, 1993). The charter was to also build on the statewide framework for quality provision, foster attributes of good schools, support continual improvement and form the basis for school accountability (Hayward, 1993):

The emphasis of the charter was to be on planned coherent and continuous improvement. Schools are considered to be self-managing but part of a state-wide government system of education. Hence it is required that school charters not only reflect local needs but the policies and priorities of the government and Directorate of School Education. Schools in Victoria are required to organise their charters around eight key concepts. The Profile describes the ethos, expectations of students, schools beliefs and values, the context in which the school operates, the facilities it has to offer, its programs and future plans. Goals describes the outcomes the school is striving to achieve, the processes of implementation, the criteria for measuring success and the method of reporting on the school's performance. Priorities were set which defined the school's improvement targets in a three-year time frame. A Curriculum Profile lists the curriculum offered to the students including the after school activities. The roles and responsibilities of members of the school community are set out as Codes of Practice and a student Code of Conduct articulates student behaviour and approved sanctions.

The contents of charters are similar in the three contexts. Charter documents are considered to be legal documents in New Zealand and the United States while in Victoria they are considered to be a formal understanding between the school council and the funding body. The legal status of the New Zealand charter is open to challenge. Unlike states within the US, if a New Zealand school fails to achieve the goals in its charter the school is not likely to have its charter withdrawn and be closed down, although intervention from government authorities may occur. The charter models used in both Victoria and New Zealand are best described as quasi-charters, where final responsibility for the delivery of services tends to rest with the funding body, in these cases the government of the day. Formal contractual arrangements where a direct legal contract operates between the school and the funding body are the common form of arrangement in the United States. Failure to comply or deliver the contracted services to the agreed standard in the US results in withdrawal of the charter and the right to practice.

The approach to the charter documents appears to take two forms. The first type consists of a centrally developed agreement containing common goals and outcomes sought which is provided to the school community with some opportunity for individualisation, although this is usually limited. This approach ensures consistency and uniformity in delivering educational services across a school district, state or country. The chartering of schools in New Zealand in 1989 is an example of this approach. The approach can be described as a prescribed approach.

The second approach provides a broad conceptual framework identifying areas that must be addressed but leaves the content to the school community to articulate within government and department policies. Schools develop their own identities and can specialise. This approach is found in the state of Victoria (Australia), Canada and the majority of American States. This type of charter document can be described as a framework charter.

Governments who have introduced charters have either taken a systemic approach and converted all schools within a system to charter schools or have established independent schools.

Where schools are given independent management, the schools contract for delivery of services as detailed in their charter. Once the charter is granted, the school is answerable to the government for the outcomes it delivers and is granted exemption from district rules and regulations and, in some cases, state legislation. Examples of this approach include New Zealand, Alberta (Canada) and Massachusetts (US). Schools operating under this model may be established schools that elect to become charter schools or new schools that are established by community groups, unions or other organisations, or private citizens.

The second approach, systemic management, means that charter schools are founded and operate within the current organisational structure. That is, they are answerable to the education authority, department or district and/or state or district regulations in which they are located. Examples of school systems where this operates include Victoria (Australia) and Los Angeles Winford School District (US). Schools adopting charters managed by this approach are usually established schools or new schools that have been planned by the central authorities.

While the establishment and operational procedures for charter schools vary, all governments sought to encourage innovation, empower local communities to take greater responsibility for education, increase educational choice and to improve student learning outcomes.

Using the themes of competition, community ownership, school management, student learning and accountability, the next section of this paper reviews the achievements of charter schools to date.

Research in relation to competition and school charters needs to be considered in two groups: competition that occurs in systems where all schools are converted to charter schools and competition in systems where charter schools are schools established in addition to currently existing schools.

In New Zealand, all schools were converted to charters at the same time that schools were dezoned, allowing parents to send students to the school of their choice. However, the structure of school charters was designed to ensure uniformity across the system and, as schools were similar in what they offered, choice was simply based on their location. Lauder et al. (1994) studied the impact of these changes on four schools in Auckland. The findings of Lauder et al.'s research suggest that where choice is offered, student numbers are more likely to fall in working class schools than in middle and upper class areas. The result is competition for places in preferred schools and a struggle for survival for working class schools. The response for schools with falling numbers was to try to become more competitive by improving their image through changes to discipline, introduction of uniform and general promotion (Marginson, 1997). The introduction of market competition into New Zealand schools suggests that the competition is not between schools per se but rather between schools representing different socio-economic districts. Marginson (1997) suggests that these findings meant that a level playing field was impossible and the outcome of intended choice and competition had in fact resulted in "market competition (that) would ensure that inequities would increase" (Marginson, 1997, p. 179).

The introduction of charter schools in a number of American states allowed for the formation of new schools, which placed charter schools in direct competition for students with already established schools (Metcalf and Tait, 1999). Consequently an analysis of charter schools demonstrated a more diverse range of offerings than where systems had converted all schools to charter schools (Office of Educational Research and Improvement, 1999). New types of schools emerged including vocational and technical schools (Skills for Tomorrow School operated by the Teamsters Union, and Livingston Academy Lowell); schools for students who have 'dropped out' of mainstream schooling (City Academy San Diego); schools for students with special needs (Vaughn Next Century Learning Centre for students with limited English and Boston Renaissance responsible inclusion program); and innovative teaching such as ungraded classes found at Renaissance School, Douglas County (Office of Educational Research and Improvement, 1999). Some indication of the popularity of charter schools with parents is evident in The Centre for Education Reform's 1996-1997 survey of charter schools that stated that 65 of the charter schools had waiting lists that averaged 135 students (Centre for Education Reform, 1998-99; RPP International, 1997).

Emerging research indicates that the competition created by charter schools can take many forms. Millman and Hess (1999) and DeSchryver and Funk (1999) report that charter schools have made a positive impact on traditional school districts across America. Millman and Hess (1999) distinguish between the possibility of charter schools and the reality of charter schools. Where charter schools are proposed they found school districts undertook low-cost reforms such as informational campaigns and teacher in-service training. The presence of charter schools, however, has resulted in high-cost reforms such as full-day kindergarten programs or administrative changes especially in low performing districts (Millman and Hess, 1999). DeSchryver and Funk (1999) provide a wide range of examples from 12 states that illustrate the ripple effect of charter schools on surrounding schools and school districts. The loss of students to charter schools, for example 5,000 students in the Mesa Public School system and 30 percent of the students from the Queen Creek School district in Arizona, has resulted in school districts taking action to attract back the students. This has included major advertising campaigns, a focus on parents to encourage them to make the right educational choices for their students and the retraining of teachers in skills such as phonics. Horn and Miron (1999) report that the introduction of the Michigan Public School Academies had an impact on their local school district by applying pressure on traditional public schools to become more accountable.

Los Angeles Unified School Districts Charter School Evaluation (1998) suggests that charter schools have influenced district reforms by heightening awareness, initiating dialogue and focusing attention on issues related to decentralised decision making and budget flexibility. Rofes (1998), who found that 25 percent of districts showed significant responses to charter schools, recorded a similar finding.

The Massachusetts Department of Education Report (1997) found that district schools were mirroring the reforms of charter schools in a bid to compete with the charter school initiatives. Aud (1999) suggests that the overall impact of the charter school program will be considerable in Massachusetts with six districts having lost more than 5 percent of their operating budgets to charter schools. Aud's (1999) study found that districts losing the highest number of students responded strongly to the market pressure, making modifications and taking action to attract students back to the school. Aud (1999) found that the impact of market pressure was greatest when the decline in student numbers was greater than 2 percent.

Hoxby (1997) reports that public schools react to competition from charter schools by offering better schooling and reducing costs. Hoxby suggests that increased competition results in significant improvement in student test scores and educational attainment. Wells et al.'s (1998) study of charter schools based on interviews with 17 schools in ten districts contradicts this statement as their research found that surrounding schools felt no pressure to change their practices because of the introduction of charter schools. Wells et al. (1998) report that public school educators believe that charter schools had an unfair advantage as the charter laws allow many things (for example, innovation) that non-charter schools cannot do. The Los Angeles Unified School Districts Charter School Evaluation (1998) challenges this finding as this research suggests that the majority of innovations adopted in charter schools could have been initiated in non-charter schools. The research to date suggests that there is likely to be less overt competition in systems where all schools are chartered than in systems where new schools are established to compete directly with charter schools.

Community ownership of charter schools reaches across a continuum from private ownership to quasi-ownership through contractual agreements and legislation where schools are governed by school councils (Victoria), boards of trustees (New Zealand) or school site boards (US). Governing bodies usually consist of parents, staff representatives and members from the surrounding community.

American research indicates a diverse range of people have established charter schools including business, unions, ethnic groups, parents, and teachers (Schmidt, 1994; Centre for Educational Reform, 1998-99; RPP International and the University of Minnesota, 1997). Los Angeles Unified School Districts Charter School Evaluation (1998) suggests that in the Los Angeles Unified School Districts parents reported feeling more welcome in charter schools and indicated that charter schools are more responsive to their concerns than their previous schools. Reports from Texas, however, suggest that despite legislation to enshrine parent participation, current involvement was mainly at fund-raising level (Texas Education Agency, 1999).

Some newly established charter schools are highlighting the importance of parent participation in schooling by specifying their requirements of parents at enrolment through a signed contract with a stated sanction for non-compliance (Wells et al., 1998). Tasks required of parents include supervising reading, checking homework and supporting the school's behaviour policy. Other requirements include set hours of voluntary work which may include grounds development, maintenance and fundraising activities. Reports on closures of charter schools indicate that where community support is not present, charter schools will not survive (DeSchryver, 1999).

In systemic charter schools increased community ownership usually means the participation of parents in the governance processes of the school. This is exemplified in New Zealand where it was the intention of the government to transfer the operations of schools to the local community, thus dismantling the state education system. The government envisaged boards of trustees taking on the policy making role, while the principal and staff would be responsible for its implementation. Mansell (1992) reports that in practice the lines were more blurred. Each school appeared to work out a process suitable for its environment with the work of policy writing generally being taken up by school administration and staff. Some schools reported that effective collaboration processes with parents had been established for policy development. In other schools, although parents were invited to participate in policy writing, few had taken up the option. Parents did not appear concerned that teachers or administrators were usurping their intended role as policy developers. In fact most schools appeared to adopt a sharing of the workload that best met the needs of their particular community (Mansell, 1992). Despite the limited participation, partnerships between parents and teachers had been established, with communities demonstrating their willingness and ability to govern schools (Ministry of Education New Zealand, 1993).

Victorian research conducted as part of the Cooperative Research Project (1997) suggests ownership and participation varied between schools, but principals on the whole had remained confident that the goals of the Schools of the Future program relating to improved parent participation and community ownership would be achieved. Female principals were reported as more confident that goals relating to community ownership and increased participation would be fulfilled than male principals. Likewise new principals (less than five years experience) were more confident than more experienced principals of community goals being met. Principals of small schools and primary schools were more confident of achieving participation and ownership goals than principals of large schools and secondary schools. However, community involvement and more cohesive staff and community relationships were ranked by principals as least likely to be realised (Cooperative Research Project, 1997).

Bradly's (1999) study of the interrelationships which characterise the daily life of charter school communities in Victorian schools suggests that a positive attitude by the school is essential for active parent participation in decision making. Schools were more likely to be successful in this area when "conscious sustained efforts to involve parents in school decision making and curriculum programs and to promote partnership relationships" occurred (Bradly, 1999, p. 9). This included publication and access to policies, responsiveness to parent concerns and a principal who nurtures and supports community relationships and maintains a high professional standard and behaviour.

Minimum research has been conducted to identify the differences that occur in parent involvement in different types of charter schools. Mansell (1992) identifies a difference between primary and secondary schools with trustees and teachers in primary schools being more aware of the way the charter was being used and what policies were in place than those in secondary schools. No evidence was provided to suggest that parent participation in primary schools was greater than in secondary schools.

While tentative, the research to date demonstrates that parents can successfully operate schools and have indicated a willingness to do so, as demonstrated by the transfer of the responsibility to parent controlled boards of trustees to operate New Zealand schools (Mansell, 1992). The type of participation varies from actively establishing schools (as in some states in America), to roles in decision-making and school operations including the hiring and firing of staff (New Zealand Boards of Trustees), to a more removed but supportive role of fundraising. Emerging data seems to suggest that the ways in which parents are involved with a school may be influenced by the school leadership and staff and their attitude to parent involvement.

School management covers a broad range of issues including: vision, financial management, staff management, long term planning and leadership. For clarity and consistency throughout this research both the preliminary discussion of the concepts and the findings are presented under the above categories and detailed below.

Research to date suggests that charters have provided the means for a school to develop a clear purpose and direction that can be communicated and shared across the school community (Caldwell and Hayward, 1998; Dunstan, 1994). Mansell (1992) reports that while the charter was given high status by principals and most teachers in New Zealand, trustees were more tentative in their support of the importance of the charter, feeling that the charter should offer guidelines rather than be definitive. The majority of principals in Mansell's (1992) study considered that the school charter simply codified what was occurring previously in their school and had little or no impact on the school. Principals reported that similar practices for policy development and consultation were already in place and at best the charter was seen as a very general basis and starting point for policy development. However Mansell found that few examples of this role could be given.

The degree to which the vision has been implemented appears linked to the ability and skills of the school leadership. Wells et al. (1998) suggest that school leaders played a crucial role and were often the catalyst for charter reform. Where whole systems have converted to charter schools the findings of the Cooperative Research Project (1997) suggest that the role principals play in a state-wide reform may be connected to the level of job satisfaction of the principal. The Cooperative Research Project (1997) found that principals with low job satisfaction also reported lower ratings for the realisation of the benefits of the reform (Cooperative Research Project, 1997).

Wells et al. (1998) identified complex fiscal, personnel or operational issues that principals were required to address in the first few years of charter operations. Business savvy, especially ability to problem solve, a sound knowledge of budgets and school finance and an ability to maximise existing and acquired resources, were identified as important for charter school administrators (Los Angeles Unified School Districts, 1998). Problems encountered by principals during the introduction of charters have included the ability to manage resources while at the same time leading positive educational reform, and managing the workload of principals and staff (Wells et al., 1998; Cooperative Research Project, 1997).

A common theme for newly established schools was the difficulty principals and charter school directors perceive in relation to obtaining adequate funding for facilities and operations (Hassell, 1999; Dolan et al., 1998). All schools however express concern about the amount of funding for building maintenance, repair and staffing (Texas Education Agency, 1999; Cooperative Research Project, 1997). Hadderman (1998) reports that while charter advocates recommend that schools control all pupil funds, in reality they rarely receive as much as other public schools.

Hassell's (1999) analysis of funding of new schools in Wake County North Carolina (US) indicates that district schools receive start up grants of $750 per pupil compared to start up grants of $195 per pupil for charter schools. In addition district schools were provided with district office planning support prior to the school's opening as well as buildings and finance.

Wells et al. (1998) highlight the variance in perceptions between funding bodies and principals in California (US). Funding bodies asserted that charter schools received the same amount of money as non-charter schools while charter school administrators universally believed they received less funding. Perceptions of less funding appear to emerge from funding formulas used in all systems that allocate funding according to grade level, with funding increasing as students move through a school. In Californian schools, access to additional funding arrangements varied greatly between schools, depending on the status of the charter school as dependent or independent and the eligibility or not for funding of facilities or capital expenses.

Directors of charter schools in Texas (US) who had to find their own buildings stated that limited financial resources made it difficult to find and maintain suitable space to offer classes, hire faculty and staff and to acquire teacher facilities (Texas Education Agency, 1999; RPP International, 1998). Dolan et al.'s (1998) analysis of charter school costs identifies the provision of facilities as high cost items for charter schools. For Victoria and New Zealand, where the responsibility for the provision of facilities remains with the Government, evidence relating to funding issues is not so prominent. While schools reported that the contracting for maintenance at school level enabled schools to be made more secure and aesthetically pleasing (Texas Education Agency, 1999), the introduction of the global budget was viewed by principals as enhancing a school's capacity in planning and resource allocation.

The fiscal viability of charter schools is intimately tied to the school's connection with resources and the skill of the administration in managing those resources. Research into closures of 30 charter schools in the US indicated that 38 percent were for reasons of poor or mismanagement and in 45 percent of the cases financial management was cited as the problem (De Schryver, 1999).

Staffing flexibility in order to make operational enhancements is commonly cited as a reason for school administrators to seek charter status. Research findings in relation to staffing appear to be influenced by the previous staffing arrangements in a school or system, that is, whether the school employed its own staff or whether staff were employed at a district or state level. In schools where staff were employed locally the impact of staffing flexibility was immediately obvious. Teachers were attracted to charter schools for their philosophy, the opportunity to work with like-minded people, small classes and prescribed methodology or curricula. Evidence suggests that charter school staff had a strong esprit de corps and a sense of being in the elite of school reform (Texas Education Agency, 1999; Bosetti, 1998).

Los Angeles Unified School Districts (1998) school charter evaluation suggests that increased control over staffing and funding flexibility motivates and fosters a sustainable school improvement culture. The findings identified a sense of being in charge and a clear sense of empowerment and efficacy supported the improvement culture. The evaluation found that just the perception that certain degrees of freedom and responsibility exist appears to foster a school culture for improvement, and teachers' acceptance that their only limitation is the teachers themselves leads to a 'no more excuses' mindset.

Los Angeles Unified School Districts study reported that the major differences between charter and non-charter schools were considered to be class sizes and the changed relationship between parents, teachers and students. While teachers valued being a part of the decision making processes in a school, concern was expressed about the workload for administration staff, principals and teachers and the time demanded for out-of-school meetings and planning. Staff generally reported tiredness and fear of 'burnout'. In Texas a high turnover of staff on an annual basis was reported. As school staff numbers in American charter schools are generally small, high demands were placed on teacher time, with teachers concerned at the lack of subject expertise (Los Angeles Unified School Districts Charter School Evaluation, 1998; Texas Education Agency, 1999).

Fears expressed at the beginning of the charter school movement in the US in relation to employment of non-credential staff do not appear to have materialised as a major problem. The National Study of Charter Schools found that only 10 percent of converted public schools report any difficulties with unions or with collective bargaining (Office of Educational Research and Improvement, 1999, 1998).

The Victorian experience highlights the difficulties involved in transferring from central employment to local school-based employment. Where central systems have been working through a transition phase to local employment the results of staffing flexibility are less obvious as staffing flexibility has been tempered by the need to reduce the number of over-entitlement teachers. Consequently the original promises have been a long time in coming to reality. Teachers have been reluctant to move to contract arrangements, with only 200 teachers accepting this arrangement in the six years since the program was introduced. New Zealand, on the other hand, was more direct in the transfer of staff from central employment to school employment, simply enacting a transfer on a set day. New Zealand has maintained a central wage fixing system. Although at odds with the overall government approaches and bemoaned by principals in conversation, it is the accepted practice in all schools.

The more successful principals demonstrated an ability to draw diverse constituencies together (Wells et al., 1998; Los Angeles Unified School Districts, 1998; Bosetti, 1998). Effective interpersonal skills enabled principals to motivate staff and parents (Wells et al., 1998) and network effectively outside their school. These skills were considered essential especially for new schools and were particularly valued by parents and teachers.

An outcome of the influence of NPM in the delivery of government-provided services such as education has been that success in student learning has moved from being viewed as the amount of resources, programs and services offered (that is, inputs) to success being viewed as the outcomes achieved by the users of the service. In the case of education this is the amount of student learning that has been achieved (Yergin and Stanislaw, 1998; Osborne and Gaebler, 1992). The dilemma, however, is in deciding how student learning will be measured given the many variables that impact on student learning. As Manno et al. comment:

Rothstein (1998) suggests that the dilemmas that beset charter schools in demonstrating their achievements include challenges to the reliability of school performance on the grounds of the validity of test items, reliability of tests and test scores due to emphasis of different tests and lack of standardisation of test conditions. Consequently the ability to compare charter schools with other schools either within a district or across districts is limited. In addition, no account is taken of the impact of social and economic status on students and school performance (Rothstein, 1998; Camilli, 1996).

School systems such as the Massachusetts Department of Education in their 1997 evaluation report recognise that "standardised tests do not tell them everything about a school" and state that other measures of schools' success will be used to judge the effectiveness of charter schools. Although in the process of replacing their testing regime, the Department believes that the current standardised test results will offer some valuable information about the initiative. The report cautions that the findings should be considered as suggestive rather than authoritative. While the limitations of the methodology used in making judgements on student learning is acknowledged, reports of student achievement are beginning to emerge.

The Massachusetts Education Department's preliminary report based on their standardised tests indicated that 80 percent of charter schools which opened in 1995 have enrolments that are at average or below average in academic performance. Six out of the eight schools were able to demonstrate that their students made academic gains. The division of charter schools Texas Education Agency in the Texas Open Enrolment Charter Schools Second Year Evaluation (1997-98) reports on charter schools that in the main exist to serve students at risk of dropping out before graduation (Texas Education Agency, 1999). Texas charter school students performed at lower levels on standardised testing than traditional public schools. Texas accountability ratings show that only 40 percent of charter schools have an acceptable or higher rating compared with 91 percent of Texas public schools in general. Three charter schools outperformed the state average and one was recognised by the state for its achievements.

Morgado and May (1999) report on the achievements of charter schools in 12 states of the US from 1994 to 1998. The report cites examples across the states that demonstrate improvements in the performance of students in state tests. Morgado and May (1999) report that the Vaughn Next Century Learning Academy (California), providing education for a Hispanic population, has increased student achievement on test scores 330 percent over four years, bringing students more in line with the national mean. The Chicago International Charter School holds the highest percentage of students at national norms. Michigan's charter schools made significant gains on the 1998 statewide achievement tests including a 24 point increase on Math scores. Results released among the sixty charter schools who administered the Stanford Achievement Test showed that 43 percent of schools improved reading scores in over 50 percent of their grade levels.

Separating and identifying the factors that have influenced student performance in charter schools is problematic. It is difficult to distinguish whether improvement is a result of greater flexibility in management, changed teaching techniques or a greater focus on areas of improvement. Rothstein (1998) challenges the results achieved by the Boston Renaissance School. The school teaches reading by contracting for a program called "Success for All". About 750 low-income non-charter schools have purchased this program. Further research comparing these schools' success rates would assist in establishing similarities and differences in outcomes and allow more substantive evidence of the factors impacting on student success than is currently available.

In Victoria, evidence of the impact of chartering schools and the subsequent focus on student outcomes is in embryonic form. Gurr's (1993) analysis of school charters demonstrated schools were focusing on improving student performance with 64.8 percent of goals and 85.2 percent of improvement priorities of school charters directly concerned with student learning. Over the period of charter intakes in Victoria, principals' perceptions had remained constant in relation to improved student outcomes with over 80 percent indicating they believed this outcome was being achieved (Cooperative Research Project, 1997). In a case study of four schools in Victoria by Wee (cited Caldwell, 1998) results suggest that where schools claimed improvement in student learning had taken place, teachers were able to cite evidence of improvement.

The Los Angeles Unified School Districts Charter School Evaluation (1998), addressing a similar concern, was unable to establish a causal relationship between increased flexibility provided by chartering and student outcomes. Charter school achievement on district standardised tests was found to be comparable to and sometimes slightly higher than student performance in non-charter schools. Schools that began with high average scores have maintained these scores. Schools that began with very low average scores that also have lower income have generally shown improvement over time that is comparable to and sometimes stronger than improvement in non-charter schools.

While a range of improvements is reported, overall one would concede that improved student learning is yet to be demonstrated. In the meantime, research on student learning in charter schools is embryonic in nature, with reports lacking the benchmarks and comparisons that would be necessary for research findings to have validity. Likewise, especially in the US in places such as Texas, the use of charters to provide for cohorts of students not currently engaged in schools or not experiencing success in schools suggests that other measures such as participation and completion of studies may be more relevant in judging the success of these schools.

The charter concept is encompassed in the idea of autonomy for accountability (Rothstein, 1998; Manno et al., 1997). Essentially schools are able to make their own decisions in relation to organisational structure, curriculum, educational emphasis and staffing profile, with exemptions from some regulations in return for being accountable for student outcomes. Research into the accountability methods and their evaluation is very limited. Manno et al.'s 1997 analysis of research conducted in 14 states and 60 schools in the US is one of the few specific studies in the area. Manno et al. identify the common areas of accountability as being: academic results, success in meeting school's own stated goals, use of public money and compliance with state laws as agreed in the charter. The analysis indicates that the latter two are easier to implement than the former.

Who is accountable to whom and for what underpins the issue of accountability. Charter schools are accountable for ensuring funds are used properly, or, in Cuttance's (1995) words, accountable for "probity and legality". Research shows that schools in the US have been closed for not meeting these requirements.

The New Zealand experience provides an overview of approaches to accountability in charter schools over the last decade. The New Zealand government desired to increase school accountability and created a high profile by implementing an independent Education Review Office (Mansell, 1992). Up to and including 1997 the task of the Education Review Office focused on assurance relating to compliance with policy and effectiveness in relation to student performance (Mohi and Hill, 1998). It was not until 1993 that National Education Guidelines were introduced. Data considered by the reviewer included the school's own charter and expectations of its students' performance.

In 1998, reviews in New Zealand were renamed "accountability reviews" and are based on an external reviewer assessing the risk in relation to the crown as owner and purchaser of educational services and to the students as the primary beneficiaries. Mohi and Hill (1998) reported that data on student performance is not collected at a national or district level and, as schools set their own criteria for assessment, like school or national data is not available with which a school can compare its performance. Schools in New Zealand currently cannot lose their charter to operate for any reason, although government advice and support may be given to schools in crisis. Consequently the purpose and value of the charter in these circumstances appears reduced in importance (Dale, 1992).

The New Zealand boards of trustees did not consider that they had an active role to play in monitoring and evaluation of policy implementation, with most boards considering this to be the role of the principal (Mansell, 1992). Mansell reported that trustees appeared happy to rely on the reports provided by the principal in the absence of parent complaints. The right of the government to override charters was supported, based on the need to safeguard national standards in regard to equity, the needs of all students and the performance of boards (Mansell, 1992).

In Victoria the accountability process is well documented, but research on its impact and effectiveness is limited at this time, further underscoring the need for research in this area. Gurr (1993) analysed the contents of the school charters which had been adopted. While not directly concerned with accountability per se, Gurr's report did demonstrate the data available to governments and education departments through the charter process. Gurr's analysis showed that using charters as quasi-contracts with schools would allow the tracking of the implementation of government initiatives and, over time, would track changes in school management and schools' relationships with parents, students and teachers.

The Cooperative Research Project (1997) suggests that principals are generally comfortable with the concept of accountability and the sharing of their results with the local community. The Cooperative Research Project identifies some ambiguity among principals in respect to the achievement of students. Principals believe that the Schools of the Future program will improve student outcomes but are not satisfied with the student achievement measures used in the annual reports. In general, principals reported that they were more accountable to their community because of the program.

Although the body of research on the impact of charters on schools is small and dominated by work in the US, it ranges over all areas of school operations from governance to effectiveness, from improvement to outcomes. While the rhetoric in literature is strong, research to support the suppositions in the concept is still emerging. One can assume that the debate on the concept of school charters will always be vigorous because of the association with political ideology, but, taking a more pragmatic view, what can the charter school movement offer the profession of teaching and the delivery of education to students?

The research suggests that charters offer an element of professional freedom, the right for teachers or communities to have their views on learning recognised and funded in return for improved student outcomes. Although contestable, early research in the US also suggests improved learning outcomes for many students and a re-entry into education for many previous dropouts.

Charter schools are said to be more responsive to consumer needs. Early research suggests that a number of charter schools are meeting the needs of specific groups of students. These are: students of ethnic origins, students with special learning needs and those who have dropped out of schooling but are finding a way to re-enter through charter schools. Charter schools also meet the needs of parents who are seeking alternative structures such as multi-age grouping or alternative philosophies such as Waldorf or Montessori approaches. The research to date suggests that newly established schools, in most cases in the US, are more likely to achieve these results than converted systemic schools such as in New Zealand and Victoria.

Evidence is scant in relation to whether charter schools actually create competition. A number of factors impinge on this. These include the size of local area population, availability of schools in the area and the uniqueness of the product offered by the school. There is emerging evidence to suggest that in some areas in the US charter schools have had an impact of reducing student populations in traditional school environments, and that schools are likely to respond through innovative practices and marketing. In New Zealand the competition appears to be based on access to schools in more socially desirable areas. The question then arises about who controls the access to over-subscribed schools, the school administrators or parents. Whether competition reduces costs is questionable, although there is evidence to suggest that in the US,` charter schools may have less access to funds than the local schools which are funded through the school board. In New Zealand and Australia, funding tends to be on a formula basis which takes into account socioeconomic status and issues such as isolation and special needs, so is more likely to be consistent across all schools.

The issue of whether communities, private individuals or groups can successfully operate schools, on deeper investigation, appears almost trivial, as history clearly indicates that it was communities who established and operated schools in the first place. Schools have been successfully operated by private interest groups, churches and elected school boards. The development of regulated systems of schooling emerged from the rapid increase in demand for schooling, especially post Second World War. In addition, education was recognised as essential to improve the economic status of western countries as they rapidly industrialised through the 1950s and 1960s. The ideological approach of the times suggested that education was essential for all and what was taught should be regulated. Hence centralisation allowed regulation as well as inferred economies of scale. Interestingly, the early research on charter schools challenges the economy of scale view espoused across Australia, clearly demonstrating that smaller schools are not only sought by parents, but also appear economically viable.

The American research suggests that financial and resource management skills as well as networking are an advantage for leaders in charter schools. Whether charters have reduced the timeframe or grid-lock in decision making is still to be judged. Likewise the role of principals or directors in charter schools is still to be defined and the manager/leader debate is unlikely to be resolved easily. Further research needs to be conducted to establish if the principals' or directors' roles in charter schools vary greatly and in what ways from principals in traditional system schools.

Whether charters will prove to be a useful tool in demonstrating improved learning outcomes is still to be researched and is perhaps the most important component of efforts in school reforms. While it is acknowledged that the majority of reforms implemented under the banner of school charters could be accomplished in any school, the process of introducing school charters appears to provide a clear and focused approach for schools to achieve set outcomes. Early US evidence, while contentious, is demonstrating improved outcomes in some schools, or at least results similar to the results students were achieving before they entered the charter school. New Zealand is yet to establish the means to report on the impact of student performance and Victorian data is only beginning to emerge.

The area of accountability appears to be the most difficult for schools and school systems to address. While charter schools are viewed as receiving increased autonomy for student outcomes, recognition that student outcomes may involve more than academic attainment is still to be addressed. The role of external agencies as a means of enforcing accountability needs further research. The history of the Review Office in New Zealand demonstrates the changing views of governments towards accountability.

The role of the school charter in accountability has not been researched in any depth. Consequently, the charter document itself may contain answers sought by funding systems, but not recognised as such. Research reported in this study may provide an insight into the charter as an accountability tool in itself.

Public services, such as the delivery of education to the community, will constantly change. The concept of charters, however, has the potential to enable schools to 'map' or document the change. While not providing all the information required, charters enable schools to start off on 'new voyages of discovery' with some confidence, enabling the schools to communicate to others what they are trying to do and map the challenges as they occur. Such activity allows the building up of the equivalent of educational atlases that record the journeys and the achievements. Just as the atlases of old were constantly found wanting by new discoveries, so will the way in which education is delivered. However, such documentation allows each new explorer to build onto the foundation stones that exist and to share new discoveries. The use of the charter to document the directions and proposed actions of a school ensures all members of the community are plotting a course headed in the same direction and assists in building internal and external confidence in operations of schools.

Like early maps, charters are not perfect, however they do appear to individualise educational programs to the needs of the local students within government policy and guidelines. While the journey in accountability is just beginning, there appears to be potential for schools to articulate their own goals and practices, be measured against them through self-evaluation and external validation procedures, and for funding agencies to be more fully informed than ever before about what is occurring in schools for the funding provided.