Analysis of the 2004-05 Budget Bill Legislative Analyst's Office February 2004

Federal Accountability And Assessments

Accountability

The federal No Child Left Behind (NCLB) Act requires the state to develop an integrated state and federal accountability system. In summer 2003, the state submitted its consolidated state NCLB plan. The state plan created a new accountability system that measures school and district "adequate yearly progress" (AYP) that operates parallel to the existing state system measured by the Academic Performance Index (API).

The state and federal systems are based mainly on Standardized Testing and Reporting (STAR) assessment results. However, the two systems measure school performance differently. The federal accountability system focuses on the percentage of students at a school that meet a certain level of achievement. The state accountability system measures the growth in school achievement from year-to-year—regardless of the level of student achievement at the school.

Low-Performing School Intervention Programs. Accompanying the two measurements of school performance are two systems of interventions and sanctions for schools that fail to meet accountability targets. The state programs include the Immediate Intervention for Under Performing Schools (II/USP) and the High Priority Schools Grant Program (HPSGP). Generally, II/USP targets schools with below-average API scores that also failed to achieve their API growth scores, and HPSGP targets the 10 percent of schools with the lowest API scores. Both state programs are voluntary, providing fiscal incentives ($200 per pupil for II/USP and $400 per pupil for HPSGP) to support the cost of developing and implementing a school improvement plan.

In contrast, the federal intervention and sanction system known as Program Improvement (PI) relies on districts to intervene in schools using existing Title I funding. Figure 1 shows the interventions for Title I schools that fail to meet AYP targets. A school that does not make its AYP target for two consecutive years enters PI (Year 1). Each year a PI school fails to make AYP results in additional sanctions. After four years, federal law requires a major restructuring of the school. To leave PI, schools must make AYP targets in two successive years. 

Figure 2 shows the number of schools in California in PI in 2003-04. More than 1,200 schools are at different stages of PI. An additional 1,155 Title I schools failed to make their AYP for the first time in 2002-03. If their 2003-04 assessment results do not improve, these schools will have failed to make AYP for two consecutive years, and therefore will enter PI. Over the next several years, we expect that most of the 5,469 Title I schools will face interventions unless the state amends the AYP system (discussed below). 

Integrate State and Federal School Intervention Programs

We recommend the State Department of Education report at budget hearings on its proposal for the integration of state and federal intervention programs in order to inform the Legislature of how it intends to bring California into compliance with federal law.

Federal law requires states to develop one system of accountability and intervention. The State Board of Education (SBE) approved California's integrated accountability program, which uses both the concepts of AYP and API. The board has not approved any policy for integrating the state and federal intervention systems. As a result, the state continues to operate the II/USP and HPSGP programs—which are oriented to school-based accountability—even though the federal system encourages states to focus on district accountability.

In the Analysis of the 2003-04 Budget Bill, we provided a framework to integrate the state and federal accountability systems. Figure 3 outlines the principles we recommended the Legislature use to design these systems. We would place the state's focus on intervening at the district level and the lowest-performing schools. Interventions in the remainder of schools would be the responsibility of school districts and the federally-mandated network of county office of education assistance called the "statewide system of school support." 

Under our principles, schools participating in II/USP would be transitioned to the federal system as quickly as possible. The HPSGP, which targets schools in the first decile, would continue and become the only state program directly intervening in schools. Finally, federal and state outcome expectations on student assessments would be aligned with passage of the high school exit exam.

In approving California's NCLB plan, the federal government reminded the state that the plan was incomplete, as the state had not included a plan for the integration of state and federal intervention programs. At the time this analysis was written, the board had not approved such a plan nor had the State Department of Education (SDE) issued a formal proposal on the issue.

Therefore, to inform the Legislature of how California will come into compliance with federal law, we recommend SDE report to the budget committees on its proposal to integrate the state and federal intervention programs.

Federal Law Requires District Interventions

We recommend the State Department of Education and State Board of Education provide the Legislature with a detailed plan on how the state will meet the intervention requirements for school districts whose schools are failing to improve. 

The federal system also requires states to establish performance targets for school districts, and requires the state to intervene in districts failing to meet those targets (see Figure 4). Thus, for the first time, the state will hold school districts responsible for the academic outcomes of their students. We believe this relationship makes sense and aligns the accountability system with the finance system—the state provides funding to school districts and not schools.

By investing more resources to directly assist districts, the state can help districts build and sustain the capacity to assist low-performing schools. Without effective district involvement in school-level reform, change is much more difficult. In the recent Evaluation Study of the Public School Accountability Act, the American Institute of Research found that districts have a significant influence on the effectiveness of direct state intervention in schools. The report found that district influence was not always positive, and that for some districts, the impact of the district on the school actually reduced the effectiveness of the reform. These findings suggest that improving district administration and governance is a prerequisite to improving the performance of individual schools.

No school districts are facing state intervention in 2003-04. The district intervention feature of NCLB is new and a school district must fail for two years to enter NCLB PI. The SBE has not determined specific performance targets for districts, and therefore it is not known how many districts will require state intervention. Even if the numbers are relatively small, the state will need to build capacity fast, and determine how to ration limited funds to assist districts with the greatest need. As such, we recommend SDE and SBE provide the Legislature with a detailed plan on how the state will meet the intervention requirements for school districts whose schools are failing to improve.

Federal Funds to Assist Schools and Districts Going Unused

We identify unallocated Title I funds available for district and school interventions to assist low-performing schools, $13 million of which must be spent before October 2005 or the state will lose it. We recommend the Legislature require the State Department of Education and the administration to provide a comprehensive plan on how the funds will be used to assist low-performing schools and districts.

Along with the new intervention requirements of NCLB, federal law requires states to set aside a portion of the main Title I grant to intervene in districts and schools not making AYP. Figure 5 shows the amount and uses of these resources over the last two years and the administration's 2004-05 proposal.

Under federal law, states have 27 months to spend federal funds or they revert to the federal government. Currently, assuming all 2002-03 and 2003-04 appropriations are spent, the state will have used $29.5 million by the end of 2003-04. This amount is slightly greater than the amount of federal funds received in 2002-03 (which must be spent by October 2004). Therefore, the state will not lose any federal funds in the current year.

In contrast, we estimate that the proposed Governor's budget could result in the state returning over $13 million to the federal government in October 2005. There are several high-priority uses for these available funds:

• District Accountability. The first PI districts will be identified in 2004-05. Since the state has little experience assisting low-performing districts, a portion of these funds could be used to support improvement programs for districts that fail to meet state performance targets. 
• PI Schools. As discussed above, a large number of schools will likely enter PI in 2004-05. The Legislature may want to consider using a portion of the intervention funds to assist districts that have a significant number of schools in PI.
• Evaluation. The state should support an in-depth evaluation of its district intervention program in order to learn what assistance is critical to strengthening district capacity to assist schools.

To avoid losing federal funds, we recommend SDE and the administration provide the Legislature with a comprehensive plan for spending at least $13 million—and up to $79 million—in Title I intervention funds that would assist schools and districts with the difficult task of turning around low-performing schools. We recommend that these plans focus around district interventions because (1) state intervention in districts supports the current governance structure, and (2) federal law requires states to intervene in failing school districts.

Legislature Should Set Realistic Expectations For Federal Accountability

We recommend that the Legislature amend the Public Schools Accountability Act to define "proficiency" for purposes of the federal No Child Left Behind Act as passage of the high school exit exam for grades 10 through 12, and being on track to pass the high school exit exam for grades 3 through 8.

As part of the definition of AYP, SBE opted to maintain the current definition of academic proficiency. For grades 3 through 8, students must score at the proficient or advanced levels on the California Standards Tests (CST) for English language arts (ELA) and mathematics. For grades 10 through 12, SBE selected a score on the California High School Exit Exam (CAHSEE) that corresponds to proficiency levels on the CST. This score is higher than the score required to pass the CAHSEE.

Figure 6 shows that on average a little more than one-third of students are proficient as measured by the CST in math and ELA in grades 4 and 8. However, for English learners and economically-disadvantaged students, as few as 14 percent and at most 34 percent, score at the proficient or advanced levels depending on grade and subject. For special-education students, the scores are even lower. Over the next 12 years, the schools will have to improve until all students are proficient. It will be very difficult for all students to reach this standard and many schools will be identified for PI. In light of these consequences, we believe the Legislature should reconsider the definition of proficiency, to a level more consistent with student expectations established in state law.

The SBE designed the proficient and advanced achievement levels to correspond to students who are on track to attend the University of California or the California State University. While the state expects all students to aspire to these proficiency levels, SBE did not establish these performance levels as a requirement for all students. The only performance requirement for all students is passage of CAHSEE. The SBE set the minimum passing score for CAHSEE at a level that reflected a performance expectation for all students, and this level is lower than what would be considered proficient on the CST.

Since schools are under pressure to meet AYP, schools have an incentive to concentrate on students near the proficient level—those who need only a little extra attention to reach the proficient level. Such a local strategy could ignore students scoring at the below-basic, and far below- basic level. We believe that creating such an incentive is contradictory to recent legislative efforts to focus more attention and resources at the lower- performing students and schools.

We recommend that the Legislature change the definition of proficiency to make passage of CAHSEE the ultimate goal for all students. The definition of proficiency for grades 3 through 8 could be defined at a level commensurate with being on track to pass the CAHSEE. While this recommendation will create a lower standard than what SBE approved for the definition of AYP, it will provide a more consistent message for what the state expects of schools and students. It will also slow the rate that schools enter NCLB PI.

Assessments

State Risks Losing Federal Funds

We recommend (1) the Department of Education report at budget hearings on the status of federal Title VI spending and whether the state is likely to return a portion of these funds in 2003-04 or 2004-05, and (2) the Legislature adopt trailer bill language to appropriate $8 million in Title VI funds in the current year for the California English Language Development Test to ensure the state does not lose any of the 2002-03 or 2003-04 federal grants. This also would save $8 million in General Fund support for this program in 2003-04.

The Governor's budget proposes spending $32 million in federal Title VI funds for the support of state assessment and data collection programs, an increase of $4 million from the current-year revised amount. This net increase results primarily from proposals to augment STAR by $2.4 million, increase funding for the CAHSEE by $6.3 million, and reduce funding for development costs of the longitudinal student assessment data base by $4 million.

Title VI funds are intended to pay for state and local costs of assessment and data reporting activities required under NCLB. Federal rules governing the Title VI funds require states to spend—not just obligate—each year's grant within 27 months from the beginning of the state's fiscal year in which they are received. Any funds that remain unspent must be returned to the federal government.

The state received $29 million in federal Title VI funds in 2002-03. It appears that the state will spend roughly this amount by October 1 of this year. If, however, the state spends less than $29 million by that date, it will have to return some federal funds. The state could be in a similar situation come October 1, 2005.

To provide the Legislature with better data on the status of these federal funds, we recommend SDE report during budget hearings on actual and projected expenditures of Title VI funds in 2002-03, 2003-04, and 2004-05. (The department was able to provide estimates of when 2002-03 appropriations would be spent, but could not supply the same data for the current or budget years.) Because the Legislature needs this information to assess the state's Title VI spending plan, we recommend SDE provide a quarterly accounting of Title VI spending over the three years.

We also recommend the Legislature adopt trailer bill language to appropriate $8 million in unspent Title VI funds for the California English Language Development Test (CELDT) in 2003-04. As we note above, part of the problem placing the state at risk of losing federal funds is the large amount of unspent carryover that has developed over the first two years of the grant. By spending a portion of this grant on a 2003-04 program that spends funds relatively quickly, the state could guarantee that no federal funds would be lost in either year. The CELDT test is a good candidate because it is supported primarily by the General Fund and is administered in the fall (and therefore spends most of its appropriation relatively early in the fiscal year). For these reasons, we recommend using $8 million of unspent Title VI funds for current-year CELDT costs. This would replace a like amount of one-time General Fund monies, resulting in savings the Legislature could use to meet its priorities in K-12 education or other parts of the state budget.

Our recommendation would leave about $7 million in Title VI funds unspent. We believe this reserve should be maintained until the costs of the longitudinal student assessment data base become clear. The department expects to begin the development of a feasibility study in 2004-05 and select a vendor for the construction of the data base in 2005-06. Keeping a reserve of federal funds would ensure that funds are available to support this high-priority data system and meet other unexpected assessment and data needs that may develop over the next two years.