Chapter 4 : Accountability

States approach accountability in various ways. Massachusetts, a state with a centralized accountability system for its charter schools, employs an experienced inspection team in the charter renewal process. As each school comes up for renewal, a team of seasoned educators evaluates the charter school?s progress and renewal application through observations and interviews at the school site. The team prepares a report, including recommendations to the state (which, in this case, is the charter school authorizer). The conclusions of the inspection team serve as evidence for either granting or denying the school?s renewal application. In focus groups conducted by SRI International, charter school leaders who had undergone this inspection reported that the process was a tremendous learning experience.

It is important to note that the Massachusetts inspectorate model is not the norm for either states or charter school authorizers. As the following findings indicate, states and charter school authorizers have many approaches to holding charter schools accountable.

The charter school movement is coming of age during a time when increasing demands are being made for public school accountability. Hence, even though charter schools may be designed to "break the mold" and "think outside the box," these schools are usually held to the same or greater outcome standards as other public schools. Requiring charter schools to participate in state assessments and to submit the same financial reports as other public schools and school districts is the way that states usually enforce this expectation. Most survey respondents argued that charter schools should also be held to the goals set in their charter, in addition to the traditional measures.

State charter coordinators were asked to identify appropriate goals for charter schools, how they should be measured, and overall expectations for charter school accountability. The majority of respondents indicated that their state laws dictated broad expectations for charter schools, usually by requiring that charter schools establish measurable goals for student performance, in addition to participating in assessment programs and meeting state standards. In states that require specific goals to be stated in the charter application, the goals are typically reviewed by the charter school authorizer and, if necessary for approval, revised or fine-tuned by the applicant. Some state coordinators cited the need for additional goals, such as dropout rates, attendance rates, parent involvement, and parent satisfaction.

Most states also require reports from charter school authorizers as part of, or in addition to, the reports required of charter schools. Exhibit 4-1 displays what information charter school authorizers are required to report to state entities. Please note that states often require charter schools to prepare their own reports that cover this information. The incidence of direct reports from charter schools to states is not reflected in Exhibit 4-1.⁹(For more information about the relationships between state reporting requirements, the age of state charter legislation, and other issues, see Appendix N.)

Like states, most charter school authorizers reported that measurable goals existed for all of their charter schools. As indicated in Exhibit 4-2, the most frequently cited goal area was academic achievement. Student attendance, student behaviors, parental involvement, and student promotion/graduation were also important goal areas.

Other types of measurable goals that were reported included goals for staff performance and attendance, parent satisfaction, student retention, course completion in high school, community service/service learning, and efforts to reduce racial, economic, and ethnic isolation.

Reports from state coordinators regarding who is responsible for holding charter schools accountable show a varied picture, with multiple levels involved. As Exhibit 4-3 demonstrates, in over half the states (22), state coordinators reported that multiple agencies were responsible for holding charter schools accountable, including state educational agencies, state boards of education, the charter school authorizer (if different from the state), and other state entities such as the state auditor and state legislature.

An examination of the type of entity permitted to charter in each state compared with the responses for who is responsible dfs zfor holding charter schools accountable suggests some interesting tensions about locus of control. For example, in states where the local boards are the only type of authorizer in the state permitted to charter?a seemingly decentralized model for charter-granting authority?a majority reported that both the charter school authorizer and the state agency were responsible for holding charter schools accountable. Moreover, in two of the local-board states, only the state agency was mentioned as the responsible entity.

Although state charter laws and state-level infrastructures establish many of the broad expectations for charter schools and their sponsors, the complete accountability story requires an understanding of the relationships between the agencies that actually award and monitor charters and the charter schools themselves. The Year 1 sample of charter school authorizers adds considerable detail to the state-level accountability story and further illustrates the point that accountability involves multiple levels, parties, and mechanisms. The role of charter school authorizers in ensuring accountability will be discussed in the findings that follow. In future years, with the added perspective of charter schools, the study team will fill out a fuller picture of the respective roles and responsibilities of each entity and how charter schools respond to the various levels of accountability.

Almost all states indicated that charter school authorizers were responsible for reviewing, negotiating, and overseeing the terms of the charter and for monitoring a charter school?s student performance. Only six states reported that charter school authorizers were not required (or were required only in some cases) to review and monitor the student performance of their charter schools. (In these cases, the state board or other state body was responsible for reviewing and monitoring charter schools.)

Responsibility for other activities, such as budget and personnel administration and provision of facilities or services, varied by activity. A majority of states did not expect charter school authorizers to administer budgets (59 percent) and personnel functions (56 percent). In contrast, states were more likely to assign the responsibility of providing services such as special education to charter school authorizers; almost 70 percent of states said that some or all charter school authorizers were responsible for providing these services. Exhibit 4-4 indicates the degree to which state coordinators reported that charter school authorizers were responsible for these functions. Further details on these responses are found in Appendix O.

The charter approval process is usually the first step in the accountability process because it is where initial expectations are laid out by schools and charter school authorizers. Charter school authorizers reported on the importance of particular program areas at various stages of this approval process: reviewing the charter for approval, requiring changes to the charter, and denial of a charter. Charter school authorizer representatives reported that the factors of most importance to them when reviewing a charter application to determine whether to issue a charter were finances, curriculum, and accountability provisions. In contrast, an applicant?s personnel policies or requirements, targeted population, and student discipline policies were considered least important. Exhibit 4-5 illustrates the importance of various program elements when charter school authorizers are determining whether or not to issue a charter.

In addition to these elements, 11 respondents reported additional program areas they considered important when determining whether or not to issue a charter. These self-reported criteria included ways that schools addressed the following: special education, language needs, insurance, parent involvement, racial diversity, projected enrollment, transportation, and student recruitment.

Charter school authorizers reported that, in some cases, they require charter applicants to make changes to their application or program during the application review process. Not surprisingly, many of the same criteria important in deciding whether to issue a charter are also important in deciding whether an applicant is asked to modify the proposed charter: curriculum, assessment, and finance were cited as the top areas in which charter school authorizers requested that changes be made. Exhibit 4-6 displays those areas in a charter in which charter school authorizers most frequently require changes to be made, as well as the areas that commonly cause charter school authorizers to deny applications. Again, reasons most often cited for application denial?curriculum, finance, and management concerns?were similar to those cited in other steps during the charter authorizing process.

In addition to the responses reported in Exhibit 4-6, approximately one-third of the surveyed charter school authorizers described changes that they recommended in other areas, including special education (n=3), language needs (n=2), transportation (n=4), and the number of signatures on the charter petition (n=2). Notable among the other areas mentioned by single authorizers were insurance, parent involvement, racial diversity, and collective bargaining.

Once the schools have been approved, charter school authorizers reported monitoring the following areas, whether they were included in a charter school?s goals or not: student achievement, financial record keeping, enrollment numbers, compliance with federal or state regulations, and other student performance indicators, such as attendance rates. These areas are consistent with the monitoring activities most frequently reported by charter schools in the National Study of Charter Schools (Nelson et al., 2000, p. 50). Exhibit 4-7 indicates the program areas monitored by chartering agencies and whether these areas apply to all, some, or no charter schools. Exhibit 4-7 also shows the mean scores for those areas monitored by charter school authorizers.

Respondents also reported monitoring the delivery of special education services, test administration, maintenance of facilities, insurance coverage, health and safety, employee rights and qualifications, adhering to orientation and mission of the charter school, student discipline, and meeting curriculum standards

* The mean scores were based on converting responses to a 3-point scale with "all" equal to 3, "some" equal to 2, and "none" equal to 1.

Although the Year 1 sample of charter school authorizers did not allow the SRI team to draw generalizable conclusions, the sample was varied enough to allow the team to determine trends and patterns across different agencies. Specifically, Year 1 analyses looked at variation between the type of authorizer and the numbers chartered along these areas: existence of written policies, changes required to the charter, reasons for denial, and organizational capacity. In looking across authorizers, it is important to bear in mind that the type of charter school authorizer appears to be closely linked to the number of schools chartered¾ i.e., local agencies in our sample chartered fewer schools than other (nonlocal) agencies.

The Year 1 data from the charter school authorizer survey indicate that the charter school promise of increased flexibility in exchange for increased accountability may be occurring more frequently in charter schools that are sponsored by nonlocal agencies than in those sponsored by local agencies. In fact, local agencies not only are less likely to allow flexibility at the school site, but also are less likely to have established written policies or procedures regarding charter schools. (The existence of written policies is an important indicator of charter school authorizers? formal processes for working with charter schools.) One interpretation is that charter schools sponsored by local agencies are monitored through traditional accountability systems based on inputs that are used for regular public schools, rather than through systems emphasizing accountability for results. Future data collection will investigate these differences in flexibility and accountability based on the type of charter school authorizer.

Most charter school authorizers reported having established written policies or guidelines for the charter school application and approval process and for monitoring and revoking charters. Of the 48 charter school authorizers surveyed, only 9 reported that they had not developed any written policies on charter schools. All of these authorizers were local agencies and had chartered five or fewer schools. In contrast, nonlocals were more likely to have written procedures (especially for granting the charter, revoking the charter, and imposing sanctions). Similarly, those charter school authorizers that had sponsored larger numbers of schools (typically nonlocal agencies) were more likely to have written policies. Exhibit 4-8 indicates the variation in whether a charter school authorizer reported having established its own written policies, procedures, or guidelines in a number of areas. In sum, 100 percent of nonlocal agencies reported written policies in one or more areas. The comparable figure for local agencies was 74 percent. (Put another way, a quarter of the local charter school authorizers in the Year 1 sample had no written policies on charter schools.)

As discussed previously, charter school authorizers often require applicants to modify a charter before it is granted. In examining changes required according to the number of schools chartered, a trend emerges: generally, higher-volume charter school authorizers require applicants to make more changes than lower-volume charter school authorizers. This finding suggest that higher-volume charter school authorizers may have learned from experience to be clear about their expectations at the beginning of their relationship with charter schools. On the other hand, some local agencies who have chartered very few schools reported working very closely with schools as they developed their charter and even writing the charter themselves.

A closer look at the data also reveals some differences in reasons for denying charters, based on type of authorizer. For example, the fact that seven local agencies reported that charter school applications were denied because of facilities (compared with only one nonlocal agency) indicates that facilities may have been a more pressing concern to local charter school authorizers than to nonlocal agencies.

The variation in charter school authorizers? organizational capacities provides some context for their abilities to oversee and monitor charter schools. (For more detail on state infrastructures available to support charter schools, refer to Appendix P.) The charter school authorizer survey included items designed to gauge the existence of a monitoring infrastructure, including whether a separate office or staff is assigned to charter schools, what proportion of the respondent?s time is devoted to charter schools, and whether written policies exist on charter schools. As Exhibit 4-9 shows, the likelihood of having a separate office or staff devoted to charter schools increased when the charter school authorizer was a nonlocal agency.

Another indicator of the organizational capacity to monitor and supervise schools is the percentage of time that the charter school authorizer respondent spent on charter-related work. In most cases, team members spoke with the person at the charter school authorizer charged with overseeing charter school-related efforts. Overall, the average percentage of time spent on charter-related work by these individuals, according to respondents, was 38 percent. This figure does not include the time that may have been spent by others at the agency. Like the existence of a separate staff for charter schools, the amount of time spent on charter-related work also varied by type of charter school authorizer; Exhibit 4-10 illustrates this variation.

In sum, nonlocal agencies are more likely to have policies in place, to have staff dedicated to charter schools, and to have chartered a greater number of schools. How interactions and negotiations of roles and responsibilities play out in practice will be a subject of further exploration as SRI gathers school-based information and additional information from a larger sample of charter school authorizers.

If flexibility is the carrot of the charter movement, the threat of charter revocation is the stick. Many argue that the most direct way of holding schools accountable is to have the power to close the school if it cannot deliver on the performance goals in its charter. Charters are usually granted for a term of 3 to 5 years and either renewed or not at the end of that term, on the basis of the school?s performance (Berman et al., 1998). In a few states, renewal is not anticipated for many years, since the charter law allows for terms of up to 15 years (e.g., Arizona, which has the largest charter school population in the country).

At present, the charter school accountability story is at a formative stage. Only 29 percent of the states with charter school legislation have had any schools come up for renewal, and in those states, almost all schools seeking renewal have been successful. Renewal activity has been most heavily concentrated in five states: California, Colorado, Massachusetts, Minnesota, and Wisconsin. In each of these states, more than 10 schools have come up for renewal.

Whereas "renewal" of a charter seems to have common meaning across states (though it may occur at varying points during the term of the charter), individual states have developed their own terminology for the negative consequences that charter schools may experience as a result of the accountability process. For several respondents, "revocation" was too harsh a term for what took place. For example, sometimes charter schools have been placed on probation and ultimately ceded their charters voluntarily before the point of charter renewal. Similarly, "closure" seemed to be a term that states preferred, rather than "revocation."

Survey respondents in 20 states (53 percent) reported that one or more charter schools in their state had been subject to some corrective action (e.g., nonrenewal, revocation, or probation). Within each of those states, the number of charter schools affected is very small?generally only one or two. Of the 18 states that reported no corrective actions taken so far, 6 do not have any charter schools in operation yet.

According to the states, the primary reasons for requiring corrective actions have been related to money and management. Other reasons mentioned included facilities issues, lack of enrollment, failure to adhere to reporting requirements, and governance issues. Some "closures" are simply a matter of failure to open because a founder?s group is unable to implement its plan or contract.

The 48 charter school authorizers surveyed had chartered a total of 837 schools, 71 of which (about 8 percent) had come up for renewal. Of those 71 schools, 76 percent had been renewed. Several schools were still undergoing the renewal process. Only 5 schools (7 percent) had definitely not been renewed. The reasons for nonrenewal of the five charters included financial problems, management or leadership issues, and student performance, among others.

A total of 27 charters had been revoked or otherwise terminated before the renewal cycle by 14 (6 state agencies and 8 local agencies) of the 48 charter school authorizers surveyed. As with the state data, the tendency to take corrective action is associated with the number of schools chartered by an authorizer. Revocations were implemented primarily by authorizers that had chartered six or more schools. Reasons for revocation again emphasized fiscal mismanagement and leadership issues.

Charter school authorizers also use probationary status as a means of helping charter schools resolve problems before revocation becomes necessary. Nine of 48 charter school authorizers had employed this strategy with a total of 15 charter schools. Once again, authorizers that had sponsored more schools were more likely to put a school on probation.

Few charter school authorizers have revoked or not renewed charters because of student performance problems. The main reason is that most charter schools have not yet reached the end of their renewal cycles (typically 5 years). The data from the Year 1 charter school authorizer survey indicate that one authorizer has not renewed charter schools because of problems relating to student performance and two authorizers revoked charters before the end of the renewal cycle because of failure to meet student outcomes specified in the charter. Also, one charter school authorizer reported placing a school on probation because, among other things, it had failed to maintain its scores on the statewide assessment. As more charter schools come up for renewal, the study team will be able to analyze in greater depth the reasons why charter school authorizers revoke charters.

The relatively low incidence of corrective actions can be interpreted in many ways. Does the fact that closure is rare indicate that charter schools are succeeding as planned; does it signify that charter school authorizers are hesitant to close schools, regardless of performance; or does it simply indicate, as some suggest, that the movement is still young? Analysis of the Year 1 data yields the following:

• Among the 20 states in which some corrective action has taken place, almost all are somewhat older states in the movement (95 percent passed their charter legislation before 1997). In contrast, half of the states that have had no incidence of accountability measures are newer participants (i.e., 9 passed legislation in 1997 or later).
• Of the states that reported no corrective actions yet, 72 percent have 10 or fewer charter schools. In contrast, 85 percent of states that have taken some corrective actions have more than 10 charter schools. Four of the five states with more than 100 charter schools (Arizona, California, Florida, and Texas) reported more than five occurrences of a corrective action taken.
• States that permit multiple but independent chartering entities (see definition of "multiple entities" in Appendix L) account for 33 percent of all permanently closed charter schools (and 58 percent of the country?s total number of operating charter schools). States that have multiple interdependent chartering entities account for another 31 percent (and 30 percent of the total universe of charter schools). States that allow only local entities to charter account for 22 percent of closures (and 9 percent of the charter schools nationally). In contrast, states where a state entity is the only charter school authorizer account for 14 percent of permanently closed charter schools, despite having only 3 percent of the country?s total number of charter schools.

In sum, the consequences for charter schools that do not meet the terms of their charters are just beginning to be understood as the older charter schools face renewal in the context of increasing state and local accountability requirements. At the same time, accountability expectations for charter schools have been expressed strongly since the beginning of the movement. Both parts of the charter school equation?increased flexibility in exchange for increased accountability?must be examined closely before judging the success of the charter school reform.

⁹ See Nelson et al. (2000), pp. 52-53, for information on reports made by charter schools.

Chapter 3

|

Table of Contents

|

Chapter 5