NCEE 2011-4019May 2011

## 3.2. Identifying SIG Schools and Intervention Models

### 3.2.1. Defining persistently lowest-achieving schools

To determine SIG eligibility, SEAs must define and identify the persistently lowest-achieving schools, the definitions of which vary across states. ED's Guidance on School Improvement Grants requires states to use three common elements to identify the lowest-performing schools: (1) a school's overall academic achievement level, (2) whether there is a "lack of progress" in the school, and (3) for high schools, whether the school has a graduation rate below 60 percent.7 However, within each of these elements there is variation across SEAs.

Schools' academic achievement level. All 50 states and the District of Columbia used student assessment results in reading/English language arts and mathematics to determine whether a school is persistently lowest-achieving (no other content areas were used). To determine the achievement level of SIG-eligible schools, states analyzed 2.7 years of data on average. Exhibit 1 lists the number of years of data used by each state. Twenty-seven states and the District of Columbia reported using three years of achievement data to identify the SEA's persistently lowest-achieving schools. Ten states reported using one year of data, and eight states reported using two years of data. One state, Florida, used seven years of achievement data.

Lack of progress. To determine if a school is eligible for SIG, SEAs are required to determine whether the school demonstrated a lack of progress on the reading/English language arts and mathematics assessments in the "all students" group. Each SEA developed its own criteria and thresholds for lack of progress. Thirty-eight states used either two or three years of data to determine lack of progress in schools. Another six states used either four or five years of data, and one state, Florida, used seven years of data (see Exhibit 2).

Eleven states used a student-level growth measure to determine whether a school had made progress. For example, the Colorado SEA developed school level growth scores from the states' student growth percentile (SGP), which is assigned based on how a student's performance compares to the academic progress of the student's peers.8 For schools, growth scores were aggregated at the grade level within each school by taking the median of all SGPs for students in a given grade across three years (2007, 2008, and 2009) separately for each content area resulting in the median growth percentile (MGP). The MGP was then combined with the school-level percent proficient, with extra weight given to the growth measure resulting in a standardized performance index score. Using the index, the lowest five percent of eligible schools were identified.9 The remaining 39 states and the District of Columbia took a somewhat different approach focusing on school-level improvement over time. For example, in New York, a school must have failed to make at least a 25 point gain for the "all students group" between 200506 and 200809 on the English language arts and mathematics assessments to be designated as "not making progress."

Graduation rates. To determine a school's graduation rate for the purposes of SIG, SEAs used between two and seven years of data, averaging the graduation rates across years (see Exhibit 3). Forty states and the District of Columbia used two or three years of data. One state, Florida used seven years of graduation rate data. Four states (Maine, Ohio, Vermont and West Virginia) did not specify the number of years of data used. In Hawaii there are no Title I eligible high schools in improvement, corrective action or restructuring.

Three states expanded the graduation rate definition for SIG-eligible high schools. For example, Nebraska raised the SIG threshold from 60 to 75 percent. The Vermont SEA had no schools with graduation rates below 60 percent, and thus included schools with higher graduation rates. Rhode Island also had no Title I-eligible high schools with a graduation rate below 60 percent.

Each of the three states with expanded graduation rate definitions devised its own system for determining schools' eligibility for SIG. For example, Rhode Island used the following point system for determining SIG eligibility among high schools:

• 2 points were assigned when the school's graduation rate was more than 1 standard deviation below the overall state average of 73.9 percent.
• 1 point was assigned when the school's graduation rate was between the overall state average and 1 standard deviation.
• 0 points were assigned when the school's graduation rate was higher than the overall state average or when the school proficiency rates for math or reading were above state averages of 52 percent and 68 percent respectively.10

Using the cumulative criteria described above (overall achievement level, lack of progress, and graduation rates), states ranked schools within each eligible pool to identify Tier I, Tier II and Tier III schools.

Top

### 3.2.2. SIG-eligible schools by Tier

SEAs defined the SIG-eligible schools by Tier in the state applications.

• Tier I Schools. Tier I schools are selected from the pool of Title I schools that are in improvement, corrective action or restructuring. Title I high schools with a graduation rate below 60 percent are also eligible for Tier I. Five states and the District of Columbia weighted a school's ranking based on the school's improvement status. For example, Tennessee first ranked all Title I "high priority" schools based on their mathematics and reading/language arts assessment scores for all students. If a Title I school had failed to make AYP for six years or more (and thus was in restructuring status), then the school's rank was multiplied by six to determine a school's priority in the list of persistently lowest-achieving schools.
• Tier II Schools. Tier II schools include secondary schools that are eligible for, but not funded by Title I, that are also among the state's persistently lowest-achieving schools. On the SIG applications, SEAs are required to define "secondary school" for the purposes of identifying Tier II schools. Twenty-three states defined secondary schools as a high school or a school serving 9th through 12th grades. Twenty-seven states and the District of Columbia defined secondary school as including both middle and high school levels, or those schools serving 6th through 12th grades. Eighteen states included newly eligible schools as part of the Tier II eligible schools.
• Tier III Schools. Tier III schools include all Title I schools that are identified for improvement, corrective action or restructuring and that are not included in the list of Tier I schools. SEAs developed different criteria for awarding SIG funds to Tier III schools, including commitment to implementing one of the intervention models, schools in greatest academic need, a school's improvement status under ESEA, schools in LEAs with Tier I and Tier II schools, schools with the highest score on the LEA application, and schools that are feeder schools for Tier I or II schools (see Exhibit 4).

"Newly Eligible" Schools. When identifying persistently lowest-achieving schools, SEAs are required to indicate if they are including a school that is newly eligible to receive SIG funds. According to ED guidance, the Consolidated Appropriations Act of 2010 modified the SIG program to allow "SEAs and LEAs to use SIG funds to serve certain newly eligible' schools (i.e., certain low-achieving schools that are not Title I schools in improvement, corrective action, or restructuring)."11 (See Exhibit 5.)

SEAs are authorized to include newly eligible schools in each of the tiers. Of the twenty-one states that identified newly eligible schools, five states included newly eligible schools in all tiers. The remaining states identified newly eligible schools in two of the three tiers (eight states) or one tier only (eight states). Twenty-eight states and the District of Columbia did not include any newly eligible schools, and one state did not report whether schools were newly eligible.12

Exclusions. Twenty-three states chose to exclude schools from the pool of eligible persistently lowest-achieving schools. Reasons for excluding schools varied, but the two most commonly cited reasons concerned small school size or schools that were primarily alternative schools designed to transition students back to a home school. Sixteen states excluded schools with low enrollment. For example, Kansas excluded schools that had less than 30 students in the "all students" category in the most recent administration of its state assessment. The exclusion of schools based on low enrollment ranged from a school size of 100 students in California to 10 students in South Dakota, Tennessee, and Wisconsin. Ten states excluded alternative schools from the eligibility list. For example, Colorado excluded alternative education campuses (AECs) that serve special needs or at-risk students because these schools are focused on re-engagement of students to transition back to their base schools.

"Start Over" and "Schoolwide Eligibility" Waivers. In their applications, states had the option to apply for waivers to allow Tier I and Tier II schools to "start over" in the ESEA school accountability timeline. Forty-seven states (all but Montana, Tennessee, and Vermont) and the District of Columbia applied for a waiver to allow Tier I and Tier II schools to "start over" in the improvement timeline if they implemented a turnaround or restart model (ESEA, section 1116(b)(12)). Through this waiver, schools that previously had been identified for improvement, corrective action, and restructuring under ESEA would no longer be so identified, thus, these schools would no longer be required to offer supplemental educational services, Title I choice, and other accountability requirements associated with Title I, Part A, of ESEA. A school that receives this waiver in the 201011 school year would not even be eligible to enter into the first year of improvement until the beginning of the 201213 school year. Another waiver option gave SEAs the option to waive the 40 percent poverty eligibility threshold to permit LEAs to implement a school-wide program in a Tier I or Tier II Title I participating school that does not meet the poverty threshold. Forty-four states (all but the District of Columbia, Hawaii, Montana, Pennsylvania, South Dakota, Vermont, and West Virginia) sought to waive the 40 percent poverty eligibility threshold in order to implement a school-wide Title I program.

Top

### 3.2.3. Intervention models authorized by states

The Guidance on School Improvement Grants from ED specifies the four intervention models to be implemented in schools that receive SIG funds. However, because of state policy or context, some models are not allowable. The turnaround, transformation and closure models are authorized in all 50 states and the District of Columbia. The restart model is authorized in 46 states and the District of Columbia (all but Mississippi, North Dakota, Vermont, and West Virginia).

In the four states in which the restart model is not an available option for LEA SIG grants, charter schools are not authorized by state statute, or there are no charter management or education management organizations in operation. For example, the Vermont state SIG application explains:

In the restart model, an LEA would close a school and reopen it under a charter school operator, a CMO, or an EMO. This option is not currently available in Vermont because no charter entities are available to work to provide these services. The low population, small schools, 94% white demographic and relatively high educational outcomes overall makes Vermont a poor location for supporting charter and/or education management organizational services (p.52).

Four states (Alabama, Mississippi, Texas, and Virginia) modified aspects of one or more of the federal intervention models by adding or focusing requirements within the models to align with state priorities. For example, the Texas SEA enhanced and customized both the transformation and turnaround models by focusing the goals of the models and prescribing specific supports (referred to as the TEA Designed Model for Turnaround and the TEA Designed Model for Transformation). The SEA's modifications to the transformation model are intended to align with three principles: 1) improving student achievement, career, and college readiness; 2) improving school climate; and 3) supporting districts to transform schools. To accomplish this, the model requires staff to participate in training on data use, develop an evaluation system, and conduct a needs assessment. The support to be provided through the model includes the assignment of a case manager, online professional development, and partnerships with the regional support network and other community stakeholders. The Texas SEA's transformation model includes a two-year leadership program in partnership with institutions of higher education. This program is intended to build LEA and school-level capacity through the establishment of a talent pool for the recruitment, selection, and development of highly qualified and effective leaders; mentoring and coaching principals to develop the knowledge, skills, and resources to accelerate and sustain increases in student achievement; and integrating research-based best practices in turnaround efforts.

In two states, the SEA added a model that is designed to enhance the transformation or turnaround models in Tier I and Tier II schools, or to serve as the improvement strategy in Tier III schools. The Alabama SEA added the "Alabama Transformation Model" to the list of potential models, which integrates the SEA's Response to Instruction13 (RtI) framework. LEAs selecting this model are required to hire a "District Grant Coach," who will be trained by the SEA in the school improvement process and RtI. The District Grant Coach will work directly with LEA personnel to build instructional capacity in Tier I, II and III schools. Virginia, too, has added a "state transformation model" for Tier III schools. The Virginia model requires LEAs to hire a coach who will work with each Tier III school on the area(s) associated with the school's chronically low performance.

Top

### 3.2.4. State priorities for awarding funds to Tier I, Tier II, and Tier III schools

In their SIG applications, all SEAs were required to specify how they would prioritize SIG grants to LEAs if the SEA did not have sufficient funds to serve all eligible schools for which each LEA applies. The federal rules require that all qualified Tier I schools be served before serving Tier II schools, and that both Tier I and Tier II schools must be served prior to awarding funds to Tier III schools. Additionally, the rules require states to describe how they will prioritize Tier III schools, anticipating that states may not be able to fund all Tier III schools.

According to state SIG applications, all states and the District of Columbia planned to fund Tier I schools (as required), and 46 states planned to fund Tier II schools. Of the four states and the District of Columbia that were not anticipating funding Tier II schools, Colorado and Rhode Island indicated that they would serve Tier II schools only if funds were available; the District of Columbia, Hawaii, and Montana did not identify Tier II schools. Twenty-five states indicated that they would serve Tier III schools as well, and twenty-one states and the District of Columbia planned to serve Tier III schools only if funds were available. Four states reported that they would not fund Tier III schools.

SIG funding extension. In their applications, SEAs had the option to apply for waivers to extend the period of availability of FY 2009 SIG funds for two additional years, through September 30, 2013. Forty-nine states (all but Montana) and the District of Columbia applied for a waiver to extend the period of availability of school improvement funds (General Education Provisions Act, 20 U.S.C. 1225(b) section 421(b)).

Top

7 U.S. Department of Education. (2010). Guidance on School Improvement Grants Under Section 1003(g) of the Elementary and Secondary Education Act of 1965.
8 According to the February 2010 Colorado state SIG application, "Colorado has developed its own measure of student academic progress, the Colorado Growth Model, which has been approved for use in the AYP growth pilot. This growth model assigns each individual a student growth percentile (SGP)[and] are reported on a scale of 199, with 50 being typical growth representing a year's worth of academic progress in a year's time...In order to calculate a growth percentile, a student must follow a traditional grade progression and have test scores for at least the two most recent years. Additional prior years of test scores yield better growth estimates, and are used whenever available." (p. 37).
9 For more information on the Colorado SEA's lack of progress indicator, see pages 37 and 38 of their state application at http://www2.ed.gov/programs/sif/summary/coapp.pdf.
10 Rhode Island Department of Elementary and Secondary Education, Application for School Improvement Grants, p.23.
11 U.S. Department of Education (2010). Guidance on School Improvement Grants Under Section 1003(g) of the Elementary and Secondary Education Act of 1965. Washington, D.C.: Author.
12 In the South Dakota state application, there is a link to a website for the list of persistently lowest-achieving schools. The list did not have a column for "newly eligible."
13 Alabama's SIG application refers to Response to Instruction not the more commonly known Response to Intervention.