How School Accountability Systems Work

The goals of accountability systems are to:

• Identify schools producing the lowest outcomes for students — particularly students who have been traditionally underserved by the school system, including students of color, students from low-income families, multilingual learners, and students with disabilities —and provide all schools with data that can inform resource allocation and continuous improvement efforts
• Enable all schools to make data informed decisions that address unmet needs of students and provide identified schools with to access additional state and federal supports and resources designed to increase the capacity of identified schools, improve practices
• Ultimately close longstanding opportunity gaps and improve student outcomes

Accountability systems are designed to improve student outcomes by:

• Allowing state officials and policymakers to prioritize resources and build the capacity of the schools and districts with the most need
• Providing schools and districts with clear goals for improving student outcomes through additional supports and resources for the lowest-performing schools
• Empowering families and communities with data, so they can demand action when schools and districts are not meeting the needs of their students

Government Roles & Responsibilities Under ESSA

Each level of government has key roles and responsibilities to ensure that accountability systems are designed and implemented in ways that promote equity and school improvement. State education agencies and state boards of education play the most substantial role under ESSA, but the federal government plays a small but essential role in enforcing, overseeing, and promoting civil rights and equity in education and incentivizing states and school districts to improve schools and student outcomes. Yet recent evidence suggests that both federal and state systems are failing to fulfill their responsibilities, limiting ESSA’s ability to facilitate school improvement.^{6 7}

Federal Roles & Responsibilities

• Provide supplemental federal funding to address the unique needs of student groups
• Review and approve state accountability plans
• Conduct regular oversight and enforcement of state education agencies (SEAs) for compliance with state plans and federal law
• Provide technical assistance and guidance on ESSA implementation and best practices to promote equity and school improvement

State Roles & Responsibilities

• Allocate state and federal funds to districts based on their specific needs
• Develop state ESSA plans that outline state accountability systems in consultation with a diverse set of stakeholders
• Collect and publish data on school performance
• Identify the lowest performing schools for support — Comprehensive Support and Intervention (CSI) schools based on overall performance and Targeted Support and Improvement (TSI) and Additional Targeted Support and Improvement (ATSI) schools based on the performance of one or more student group(s)
• Provide technical assistance and guidance for schools and districts as needed, including help and guidance on developing school improvement plans
• Approve plans for the lowest performing schools identified for CSI
• Allocate federal school improvement funds to districts with identified schools
• Conduct required Resource Allocation Reviews in partnership with districts with “significant” numbers of identified schools ⁸

District Roles & Responsibilities 

• Allocate federal, state, and local funds to schools
• Equitably allocate non-financial resources to schools
• Report school performance data to state education agencies (SEAs)
• Develop CSI school improvement plans in collaboration with identified schools
• Approve and support development of TSI and ATSI school improvement plans, including needs assessments, review of resource inequities, selection of evidence-based interventions
• Support implementation of CSI, TSI, and ATSI improvement plans and provide guidance and support

School Roles & Responsibilities

• Report accountability information to district/state
• Develop TSI and ATSI school improvement plans
• Implement evidence-based interventions to improve school performance and student outcomes
• Communicate school performance to staff, families, and community members

Principle #1: Design and use indicators in state accountability systems that keep student learning front and center

Measuring student growth

• States can do this by: Measuring student growth in ways that are high quality, are easily understood by stakeholders, and focus on growth toward grade level proficiency.
• Doing so means that: Schools maintain focus on — and get credit for — supporting all students in making growth toward proficiency.
• EdTrust recommends: Using a value table, which uses a simple calculation to show if schools moved students to the next performance level or made progress within a level.
• Our analysis revealed that: Just 13 states use the value table model. In contrast, 24 states use a student growth percentile model, which compares student growth to other students’ rather than against standards.
• Learn more about: Measuring student growth

Weighting academic growth and achievement

• States can do this by: Balancing the weighting of academic proficiency and growth.
• Doing so means that: Schools maintain focus on holding all students to high academic standards, while providing credit for improving skills of students not yet performing at grade level.
• EdTrust recommends: Having the weights of growth and proficiency within 5 percentage points, ideally with growth slightly outweighing proficiency.
• Our analysis revealed that: A plurality of states (13) weight growth and proficiency within 5 percentage points of each other. Only 4 states substantially (10+pp) weigh proficiency more than growth.
• Learn more about: Weighting academic growth and achievement

SQSS indicators

• States can do this by: Using School Quality and Student Success (SQSS) indicators to provide an expanded but clear definition of school quality.
• Doing so means that: Schools get credit for additional outcomes beyond academic performance, while still providing clear priorities for schools to focus on.
• EdTrust recommends: Including a small number — no more than 3 — SQSS measures per grade span, particularly for elementary and middle schools.
• Our analysis revealed that: Most states align with our recommendation, with just 3 states having more than 3 SQSS indicators across all school levels, including indicators that measure student engagement and performance in subjects beyond math and English language arts.
• Learn more about: SQSS indicators

CCR indicators

• States can do this by: Using SQSS indicators to measure how high schools are preparing all students for postsecondary success.
• Doing so means that: Schools are incentivized to provide high quality college and career pathways for all students.
• EdTrust recommends: Including both college readiness indicators (like AP exams or dual credit) and career readiness indicators (like CTE concentrations or industry certifications) that focus on student success, rather than just access.
• Our analysis revealed that: Of the 38 states that include college and career readiness indicators, more than half (21) include both college and career indicators — and just 3 states included indicators that only examined access.
• Learn more about: CCR indicators

Principle #2: Develop school ratings that reflect how schools are serving all groups of students

Including all students in school ratings

• States can do this by: Setting a minimum number of students required for a student group to be included in ratings systems (n-size) that balances inclusion and student privacy
• Doing so means that: As many students as possible are “seen,” so schools with small populations of certain groups of students are still held accountable for their performance
• EdTrust recommends achieving this by: Using an n-size of 10 students
• Our analysis revealed that: Only 13 states use an n-size of 10, while 30 states use an n-size of 20 or higher
• Learn more about: Including all students in school ratings

Incorporating student group performance into indicators

• States can do this by: Considering both overall student performance and the performance of individual student groups in school ratings
• Doing so means that: Schools, particularly those not at risk of being identified as low performing, are incentivized to focus on the performance of all students
• EdTrust recommends achieving this by: Including the performance of every individual student group when calculating scores for each metric that makes up a school’s overall rating
• Our analysis revealed that: Only 11 states include performance of student groups in ratings, with just 3 — CO, DC, NJ —outlining specific weights for each student group
• Learn more about: Incorporating student group performance into indicators

Principle #3: Establish criteria that honestly identify which schools need to take steps to improve overall or for one or more student groups

Setting clear and distinct identification criteria

• States can do this by: Setting criteria for identifying TSI in a way that acts as an “early warning” system for being identified as an ATSI schools—those with groups performing as poorly as the lowest 5% of schools overall
• Doing so means that: Schools are held accountable for addressing low performance of student groups and have access to supports unlocked by identification
• EdTrust recommends achieving this by: Setting criteria for TSI identification that is substantially more inclusive than ATSI criteria, but not so broad as to overwhelm state capacity to support identified schools
• Our analysis revealed that: Exactly half of states have TSI identification criteria that is meaningfully different than ATSI criteria
• Learn more about: Setting clear and distinct identification criteria

Using consistent standards for all student groups in identification criteria

• States can do this by: Setting criteria for TSI identification that looks only at student group performance and holds all students to the same high standards
• Doing so means that: Schools are explicitly held accountable for the performance of each student group and states don’t set different standards for different students. Results aren’t obscured by averages
• EdTrust recommends achieving this by: Setting criteria for TSI identification that avoids the use of schoolwide averages or use of different performance benchmarks for different student groups
• Our analysis revealed that: Two-thirds of states set TSI criteria in ways that honestly identify schools based on student group performance
• Learn more about: Using consistent standards for all student groups in identification criteria

Principle #4: Report disaggregated accountability information in a way that is understandable, easily accessible, and widely available

Summative ratings

• States can do this by: Providing clear information of school performance on school report card websites
• Doing so means that: Parents, families, educators, and school leaders have an accessible understanding of how schools are serving students
• EdTrust recommends achieving this by: Including a clear overall summative rating of school performance, ideally accompanied by information that explains how that rating is calculated
• Our analysis revealed that: Exactly half of states include a summative rating of school performance. A majority of these states (15) present their rating on a 1-100 scale
• Learn more about: Summative ratings

Publishing disaggregated and comparable data

• States can do this by: Publishing disaggregated data on performance by student group
• Doing so means that: All stakeholders have a clear understanding about how schools are serving each student group and allow leaders to consider which student groups need access to additional resources to succeed
• EdTrust recommends achieving this by: Including easily understood disaggregated data and providing users with the ability to compare the performance of schools—including by school type, size, and demographics
• Our analysis revealed that: All states include at least some disaggregated data, as required by law. Thirty states include a comparison function, but only 10 allow comparisons by school characteristics or include their disaggregated data in their comparison functions
• Learn more about: Publishing disaggregated and comparable data

Using asset-based language

• States can do this by: Using asset-based language in contextual information to frame school performance ratings
• Doing so means that: Students, families, communities, and educators have a clear understanding of what accountability data says about schools’ strengths and areas of growth and the purpose and uses of this data
• EdTrust recommends achieving this by: Highlighting the responsibility the education system has toward students, recognizing the impact of finite and inequitable resources, and highlighting school strengths without masking areas for growth
• Our analysis revealed that: Just four states include some form of asset-based framing language as a part of their online school report card
• Learn more about: Using asset-based language

Principle #5: Provide meaningful support to schools that need to improve

Setting exit criteria

• States can do this by: Setting criteria that allow identified schools to exit this status that is ambitious yet attainable
• Doing so means that: Schools no longer identified for support have made substantial progress in improving performance and have conditions to sustain progress over time
• EdTrust recommends: Setting criteria that require schools to have performance ratings substantially beyond the identification threshold, demonstrate multiyear growth in performance, and show evidence of participation in school improvement activities
• Our analysis revealed that: No state has exit criteria that meet all three of our recommendations, and just four states require performance to exceed the identification threshold. However, 30 states require schools to show growth to exit status, with 11 also requiring participation in school improvement activities
• Learn more about: Setting exit criteria

More rigorous supports and interventions

• States can do this by: Setting clear next steps for identified schools that fail to make enough progress to exit status
• Doing so means that: Identified schools that aren’t sufficiently improving get both the motivation (to avoid state intervention) and additional support and attention they need to improve and, therefore, ensure students are not stuck in schools that are not making improvements
• EdTrust recommends: Outlining a series of escalating interventions — including technical assistance and monitoring — that help schools progress and lead to substantial changes in school practices or control of school decision-making
• Our analysis revealed that: Most (37) states indicate that they provide a combination of additional supports and increased monitoring to schools that don’t exit status, while only six states outline clear, specific changes in school management or other state interventions
• Learn more about: More rigorous supports and interventions

School improvement funds

• States can do this by: Strategically allocating federal school improvement funds (required 7% set-aside of Title I funds) to identified schools to support school improvement activities and planning and publishing related information publicly
• Doing so means that: Identified schools have additional capacity to thoughtfully develop school improvement plans with stakeholder input and invest in evidence-based practices to improve performance
• EdTrust recommends: Developing a grant program to ensure all identified schools have access to improvement funds, creating clear criteria for receipt of funds that guide or require schools to carefully consider local data and invest in evidence-based practices, and publishing information on the use of funds, including how schools plan to use them
• Our analysis revealed that: Of the 37 states with publicly available information on school improvement funds, just four employ a hybrid approach that ensures wide access to funds, through a combination of formula and competitive grants or using a formula but requiring an application to ensure strong use of funds
• Learn more about: School improvement funds

Resource allocation reviews

• States can do this by: Conducting reviews of district resource allocation in partnership with districts that have a “significant” number of identified schools, setting clear criteria for which districts qualify for this support, and publishing information and guidance about their process
• Doing so means that: States and districts can work together to identify and address inequitable access to resources —including funding, high-quality educators, and a rigorous curriculum —that are contributing to low school performance
• EdTrust recommends: Setting criteria for which districts are eligible for resource allocation reviews (RAR), by setting percentage and number thresholds (to capture small and large districts) and looking beyond funding to other types of resources when conducting reviews
• Our analysis revealed that: Just 14 states have published information that includes a clear threshold for identifying districts for RARs, while only one uses both a number and percentage threshold. 15 states indicate they will consider providing resources beyond funding. Only 4 states say they do both
• Learn more about: Resource allocation reviews

Measuring Academic Growth

Most states now include some measure of academic growth when rating schools. Growth is an important aspect of school ratings because it can show whether schools are effectively serving students and helping them develop their knowledge and skills. Growth is also considered a fairer measure of school quality because, unlike academic proficiency, which is strongly correlated with student demographics and family income, growth is less influenced by how much access a student has to opportunities outside of school and more reflective of what a student has actually learned. ^{11 12} Yet, different ways of measuring growth answer fundamentally different questions about how schools are contributing to student success, despite falling under the same label of “growth.”

• EdTrust believes that a state’s measure of growth should be of strong technical quality, easily understood by a wide range of stakeholders, and focus on whether student performance is progressing toward grade level standards.

Of the measures used by states, a value-table measure best fits these criteria. Value-table systems credit schools for moving students within a performance level and/or from one performance level to the next from year to year. This way of measuring growth incentivizes schools to improve the performance of all students, regardless of their previous performance, focuses on moving students toward meeting grade level standards, and can be scored with a simple calculation. Unfortunately, just 13 states use this measure.

Eight states use value-added growth. While this measure is often viewed as strong because it uses advanced statistical modeling to isolate a school’s unique impact on achievement, the use of this measure is concerning because it often controls for student demographics such as race/ethnicity that can produce different growth targets for different student groups. What’s more, the complex calculations it employs may be difficult to explain to stakeholders and the public, potentially undermining trust in the results.

The widespread use of a measure known as student growth percentile (SGP) — which is used by nearly half of states (24) — is also problematic. SGP assesses the progress students are making in comparison to peers with similar previous scores on state assessments — rather than growth toward academic proficiency. Unfortunately, this measure can give a misleading impression of the progress schools are making with students, particularly now when schools are focused on addressing unfinished learning due to pandemic closures. A school in which students made little or no progress toward grade level proficiency can have a positive SGP score, if other students around the state performed worse. And this relative measure takes focus away from the goal of grade level achievement. Even more worrisome, four more states are using SGP than in 2019.¹³ When it comes to measuring student growth, states are moving in the wrong direction.

Weighting Academic Growth in Elementary & Middle Schools

Beyond how growth is measured, how much relative weight it has in a school rating system has serious implications for schools. If not appropriately weighted, it can skew overall ratings and undermine political support for accountability systems. Weights matter because they reflect where state leaders believe they should focus their improvement efforts and what is valued in the system.

Favoring growth gives schools more credit for the academic indicator that’s more in their control and, therefore, signals to school leaders and educators that their contributions to student achievement are valued by the state. Yet, in 2023, the Tennessee Department of Education slightly shifted its weighting in favor of achievement. While the impact on school ratings hasn’t been investigated, the change was made despite strong objections from the community and has led advocates and school leaders to view the state’s system as less fair. ¹⁴

• EdTrust recommends that states weigh growth and proficiency similarly in their systems to encourage a focus on grade level proficiency. This encourages schools to hold all students to high standards, while still giving schools credit for improving students’ skills, even if they’re not yet up to grade level. Ideally, growth measures should slightly outweigh achievement.

Encouragingly, a plurality of states (13) has weights for achievement and growth that are within five percentage points of one another, while another eight states weigh growth over achievement by 5 to 10 percentage points. By contrast, just four states weigh achievement by 10 percentage points or more than growth.

Expanding Definitions of School Quality

The goal of including an ESSA provision on school quality and student success measures (SQSS) was to provide states with the flexibility to develop a broader definition of school quality that aligns with community values.

• Ideally, schools should be incentivized to develop school services that support a more expansive definition of student success. EdTrust suggests states use a small number (no more than three) of student success indicators for each grade span that measure the quality of opportunities provided to students.

Including a few key SQSS measures helps ensure that schools prioritize specific areas and that changes in performance on these measures are connected to meaningful changes in a school’s rating. Having too many indicators might keep schools from focusing their improvement efforts and limit the potential impact that improvement in any one area can have on a school’s final rating.

Overall, most states have limited the number of SQSS measures in alignment with EdTrust’s recommendation: Just 14 have more than three indicators at any one school level, and only three of these states have more than three indicators across all school levels. In all 14 of these states, these indicators are concentrated at the high school level, which likely reflects states’ emphasis on measuring college and career readiness — a topic we’ll explore in more detail below.

Common Indicators of School Quality & Student Success

The types of indicators states have chosen fall into three categories: (1) student engagement, (2) college and career readiness, and (3) achievement and access in subjects other than English language arts (ELA) and math.

1. Student Engagement

Measuring student engagement can incentivize schools to focus on building a welcoming, supportive school climate, which research shows is a prerequisite for allowing students to fully engage with classroom content and develop their academic knowledge and skills.^{15 16 17 18} These measures can also identify students who need intervention, prompting schools to provide targeted supports to reengage those students, particularly if schools also track progress throughout the year. ¹⁹

Consistent with our previous analysis of state plans, chronic absenteeism measures are by far the most common: 35 states include them in their accountability systems. Chronic absenteeism is considered a strong proxy for student engagement and school climate, and data on absenteeism is readily available. ²⁰ While most of these states rate schools on the percentage of students who miss less than 10% of school (which is the common threshold for chronic absenteeism²¹), a handful of states take slightly more innovative approaches. For instance, Montana sets a higher bar of 5% of missed school days. Indiana also sets a higher bar (4% of days missed) and gives schools credit for every student whose attendance rose by 3 percentage points from the previous year. This could incentivize schools to support students with many absences.

A smaller number of states use other measures to try to capture student engagement. For example, eight states use a ninth grade on track indicator to determine whether students are earning enough credits to graduate high school on time and encourage high schools to ensure a smooth transition from middle school, which is a crucial predictor of student engagement and persistence. ^{22 23} Nevada and Arkansas go a step further by also tracking credit accumulation into 10^th and 11^th grades, respectively.

Five states use school climate surveys to measure whether schools are creating a supportive, welcoming learning environment for students; and Montana plans to add a similar indicator but has yet to determine how school climate will be measured. Though understanding school climate is important for supporting student needs, incorporating the results of these surveys into school ratings — which four of five states do — could create perverse incentives for schools to influence the results, which would produce inaccurate and unhelpful data. ²⁴ To avoid this, Illinois rates schools on survey participation rather than survey results. This encourages widespread completion of the survey, which school leaders and educators can then use to improve practice.

• Measuring College and Career Readiness 

Gaining an understanding of the extent to which students are prepared for postsecondary opportunities remains a key and important consideration of states’ accountability systems: 38 states have at least one measure of college and career readiness as a part of their accountability system. Given the limited requirements for high school assessment and variations in district graduation requirements, having additional measures of the ways in which high schools are preparing students for success is a welcome outcome of the flexibility provided by SQSS indicators.

Based on our examination of states’ use of college and career readiness (CCR) indicators, EdTrust would urge state leaders to consider using both college indicators (like Advance Placement exams and dual credit courses) and career indicators (like CTE concentrations or industry certifications), as well as indicators that focus on student success rather than access to opportunities.

We found that of the 38 states with CCR indicators, more than half (21) include both college and career indicators, and encouragingly, nearly all focus on success in their chosen indicators. States were most likely to focus on passing scores on AP/IB tests (24), dual enrollment course completion (18), qualifying scores on the SAT/ACT (16), industry certifications (16) and CTE course/concentration completion (10). An additional 12 states also include measures of military readiness, typically success on the Armed Services Vocational Aptitude Battery (ASVAB) assessment.

EdTrust believes that high schools should set students up for success after high school. Therefore, we also looked at the extent to which states included measures of student success beyond high school. Just four states — Connecticut, Georgia, Iowa, and West Virginia — do, and all four use a measure of enrollment in postsecondary education as a part of their college and career indicator. State leaders may be under political pressure from school leaders to omit such measures. And many school leaders report that they do not even have access to data about their students’ postsecondary applications, postsecondary enrollments, or FAFSA completions, and feel that the most significant barriers preventing students from successfully transitioning to postsecondary opportunities are beyond their control.²⁵

As highlighted in an All4Ed series on states’ CCR indicators, Georgia stands out as having the most meaningful measure of postsecondary success; it looks specifically at the proportion of students that enroll in non-remedial college coursework. This is important because students who enroll in remedial courses are less likely to graduate and must pay for courses that don’t count toward degree requirements. ²⁶ As states continue to develop and refine their CCR indicators — many states have made updates or clarifications to these indicators in recent years — additional research should be conducted to better understand how these indicators are connected to student success and how they are impacting school practices.

3. Achievement & Access in Other Subjects

One widespread criticism of NCLB was that the focus on proficiency rates in ELA and math led schools to narrow their curriculum and limit instructional time in other subjects.^{27 28 29} Using ESSA’s flexibility, a subset of states has tried to address this by including measures that rate schools in other subjects. For example, 14 states include a measure of achievement in science. This is perhaps unsurprising, since states are required by ESSA to assess science achievement at least once in grades 3-5, 6-9, and 10-12. Seven of these states also include a measure of achievement in history/social studies.

A smaller group of states is looking beyond core academic classes to incentivize schools to increase student access to enrichment courses, which is linked to higher attendance and student engagement.^{30 31 32}  Four states use indicators connected to physical education and arts classes. Georgia includes indicators for both, as well as for world languages, and rates schools on passage rates rather than enrollment rates, which can help ensure that students have access to high-quality opportunities. Michigan and Illinois use enrichment indicators that measure student-to-staff ratios, in addition to enrollment. While this could incentivize schools to ensure student access, schools may not be able to control their staff ratios if they are set at the district level. As other states contemplate including indicators for enrichment classes, they should carefully consider whether this will incentivize schools to provide high-quality opportunities and whether school leaders have the ability to control their own outcomes.

Bright Spots in the States

Chronic absenteeism measures: Most states rate schools on the percentage of students missing less than 10% of school, but a handful use more innovative approaches. Montana sets a higher bar of 5% of missed school days. Indiana also sets a higher bar (4% of days missed) and gives schools credit for every student whose attendance rose by 3 percentage points from the previous year, potentially giving schools more incentive to support students with large numbers of absences.

College & career readiness measures: Georgia has a particularly meaningful measure of postsecondary success, which looks specifically at the proportion of students who enroll in non-remedial college coursework.

School climate measures: Some states use school climate surveys to measure whether schools are creating a supportive, welcoming learning environment for students, but Illinois rates schools on survey participation rates instead of survey results to encourage widespread completion of the survey, which school leaders and educators can use to influence practice.

N-Size 

Setting the n-size, or minimum number of students required for a student group to be included in school rating calculations or the identification of schools, is a key way for states to ensure that all students are seen by accountability systems. To protect student privacy and ensure that an individual student’s results aren’t identifiable, schools cannot and should not make the n-size too small in accountability systems. A very small n-size can also lead to large swings in performance from year to year, as the performance of any one student would have an outsize impact on a group’s overall performance. Yet, setting a needlessly high n-size could let many schools escape accountability for the performance of all students. A Brookings Institution analysis notes that an n size of 30, the threshold set by 10 states, would exclude nearly 40% of Black elementary-grade students nationwide from subgroup accountability. As a result, schools that are underserving certain groups of students might be overlooked for identification and support because they don’t have enough students from those subgroups to meet n-size thresholds.

• EdTrust recommends states use an n-size of 10 students to balance student privacy against the need to ensure as many students as possible are seen — but only 13 states meet this threshold.

A plurality of states (17) uses an n-size of 20, while 13 states use even higher n-sizes, substantially limiting their accountability systems’ ability to hold schools accountable for the performance of all students.

Student Groups in School Ratings

School, district, and state leaders — as well as parents and communities — use school ratings to understand whether schools are meeting students’ needs. While ESSA does not require states to incorporate the performance of individual student groups into a school’s rating? — just the average of all students’ performance — doing so would help ensure that disaggregated group performance on each metric — beyond assessment results — is publicly reported.

• EdTrust believes school ratings provide an opportunity for states to center equity by incorporating student group performance into how states rate schools on each metric in their accountability system.

Unfortunately, states have largely failed to ensure that school ratings reflect how well schools are serving all groups of students. Just 11 states incorporate the performance of student groups into their overall school performance rating.

Colorado, New Jersey, and Washington, DC, stand out for having the clearest and most substantial incorporation of student group performance into their accountability system. Both Colorado and New Jersey weigh the performance of student groups, which counts for one-third and one-half of the total indicator score, respectively. Washington, DC, dedicates 70% of each indicator score to student group performance, but assigns different weights by group. The performance of students from low-income backgrounds is weighted more heavily than the all students group (40% vs. 30%). DC also puts substantial weight on the performance of students with disabilities (10%) and multilingual learners (5%), with the remaining 15% distributed equally across each race/ethnicity. These three state systems help ensure that schools are accountable for improving the outcomes of all students by ensuring that the performance of student groups is not only used to identify schools for improvement but also counts substantially toward a school’s overall score.

Supergroups

Like the states highlighted above, Connecticut factors student group performance into each indicator score. However, instead of including all student groups, the state’s accountability system features a “high needs” group, which averages the performance of students from low-income backgrounds, multilingual learners, and students with disabilities. While Connecticut should be commended for not using the average of all students to rate schools, as the vast majority of states do, the use of this “supergroup” is problematic because it deliberately excludes student racial and ethnic groups and masks how students of color are performing in a state with some of the widest racial opportunity gaps in the country.³³ Texas rates schools using a similar high needs group that also includes highly mobile students; it also consolidates the two lowest performing racial/ethnic groups from the previous year into a single category. This latter supergroup not only fails to include students from all racial and ethnic groups but may create a moving target for improvement if the groups included change from year to year.

By contrast, five other states use student supergroups to increase the visibility of all students in their accountability systems. In these cases, groups of under-represented students are aggregated and included in school ratings or identification only when a group does not meet the state’s minimum n-size. Yet, in all these cases, the state is using a n-size larger than 10, so reducing the n-size could be better way to include more students in their accountability system. Additionally, while the point of using supergroups is to increase the visibility of all groups of students, this kind of consolidation may wrongly suggest that all student groups have similar needs, rather than unique needs that require different supports.

Bright Spots in the States

Including student groups in school ratings: Colorado, New Jersey, and Washington, DC, clearly incorporate student group performance into their accountability systems. Both Colorado and New Jersey weigh the performance of student groups, which counts for one-third and one-half of the total indicator score, respectively. Washington, DC, dedicates 70% of each indicator score to student group performance, but assigns different weights by group. The performance of students from low-income backgrounds is weighted more heavily than the all students group (40% vs. 30%). DC also puts substantial weight on the performance of students with disabilities (10%) and multilingual learners (5%), with the remaining 15% distributed equally across each race/ethnicity.

This school identification process under ESSA matters for three key reasons

• Identification unlocks additional state and federal resources. States are required to set-aside 7% of their Title I funds for school improvement activities at identified schools. While these funds are typically insufficient to stand up large-scale programs or interventions, (in the 2019-20 school year, identified schools received an average of $105,000), ³⁴ they can be used to build the capacity needed to thoughtfully identify needs and plan for improvements or potentially set-up pilot programs. Additional resources can also take the form of support from state or regional staff.
• Identification allows SEAs to prioritize distribution of their limited funds and capacity to the schools with the most need. It’s, therefore, important that states choose identification criteria that balance the following considerations: identifying as many schools as possible that have high needs (to ensure access to supports) and limiting the number of schools that will be identified, so that finite state capacity and funds aren’t stretched so thin that identified schools don’t get adequate support.
• Being identified under ESSA sends a clear signal to school leaders, staff, and communities that a school must make substantial changes to better meet the needs of all students. Identified schools must work with their districts and communities to complete required school improvement activities, and identification may also prompt community and advocacy groups to push schools to address what they see as pressing needs.

What’s more, the failure to identify a low-performing school may limit access to additional resources and targeted interventions for students in those schools, potentially exacerbating educational disparities and hindering efforts to address specific challenges these students may face and impeding their academic progress and overall educational outcomes.

Despite the important equity implications of TSI and ATSI identification, states have substantial flexibility in defining their criteria for TSI identification. States can determine what qualifies as a “consistently underperforming” student group, which triggers TSI identification.

• EdTrust recommends states choose TSI identification criteria that can act as an “early warning” system that identifies schools that are at risk of becoming ATSI schools, which are schools with extremely low student group performance — i.e., they have groups that are performing at or below the level of all students in CSI schools (the lowest 5% of schools in the state). States must consider student group performance in a way that upholds high standards for all students and focuses solely on the performance of individual student groups.

Half of states define consistently underperforming student groups in a way that aligns with EdTrust recommendations. Some states, including Hawaii and South Carolina, use a straightforward approach: They define consistently underperforming groups as those with performance scores in the bottom 10% in the state, double that of ATSI. Promisingly, our analysis revealed that, since 2018, seven states have improved their definitions of consistently underperforming student groups to better distinguish between TSI and ATSI schools.

Other states with TSI definitions that align with our recommendations use student group performance on a subset of indicators. For example, Pennsylvania identifies schools with one or more student groups that have low achievement, low growth, and below-average performance on any SQSS indicator. And Colorado identifies schools in which a student group earns the lowest rating across three indicators. Maine does the same, but looks only at a single indicator, which while it captures more schools than ATSI, could potentially identify more schools than the state can adequately support.

Some states, however, have used federal flexibility to set TSI definitions that are likely to capture fewer schools than ATSI, which is counter to the spirit of the law. For instance, Arkansas identifies schools with groups performing in the bottom 1% of the state, and Arizona identifies schools with groups performing two standard deviations below the state average, which in the 2022-23 school year resulted in the identification of just 19 schools — less than 1% of schools in the state.

When we looked at how states are considering student groups, we were encouraged to find that two-thirds of states appropriately factor in group performance in ways that hold all students to high standards and focus only on student group performance. However, 15 states define consistently underperforming student groups in ways that lower expectations. These states set their thresholds based on group performance that is lower than the average for that group, rather than the average for all students. This effectively means that they are setting different bars for different student groups. Some states do this by explicitly comparing performance within that group. Other states, like Georgia, consider student groups’ progress toward the state’s long-term goals but set different targets for different student groups. Ultimately, ESSA flexibility in the school identification process has led many states to define consistently underperforming student groups in ways that fail to identify — and support — many schools that need to address disproportionate student outcomes.

Bright Spots in the States

Identifying Schools for Improvement: Hawaii and South Carolina, defining consistently underperforming groups as those with performance scores in the bottom 10% in the state, double the threshold of ATSI. Another promising identification practice is focusing on student group performance on individual indicators: Pennsylvania identifies schools with one or more student groups that have low achievement, low growth, and below average performance on any SQSS indicator. And Colorado identifies schools where a student group earns the lowest rating across three indicators.

Accessibility and Clarity of State Report Cards

When we examined states’ online school report cards, EdTrust hoped to see several key features:

• All accountability information should be available on a single webpage, with a clear summative rating at the top of the page
• Ratings should be accompanied by a clear explanation of how the grade was assigned, including information on the number of points earned out of the total points possible, as well as how the rating was calculated
• All data should be disaggregated, and the site should allow comparisons between schools — including by school type, size, and demographic
• There should be clear connections between school ratings and states’ school improvement systems of support.

• Summative ratings are important for states because they provide a quick snapshot of how schools are performing, which allows users to clearly understand how individual schools are serving students. But it’s important for states to publish disaggregated data because it shows where schools, districts, and states should target their resources, highlights potential opportunity gaps, and provides a deeper understanding of which student groups are being underserved by schools in the state.

Our investigation found that states have a lot of room for improvement when it comes to meeting these standards of transparency and accessibility. While all but two states (Hawaii and New Hampshire) had easy-to-find school reporting websites, just 11 states met all our criteria.

Half of states provide summative ratings of school performance on their online state report card to give stakeholders a clear overview of a school’s performance, but state summative ratings take a few different forms. Just eight states assign schools an A-F letter grade, a practice that has been criticized for unfairly stigmatizing schools. ³⁵ Fifteen states give schools a numerical score on a 1-100 scale, the most common practice. Four states use a 5-star rating system, while five states use descriptive performance bands (e.g., Exceeds Expectations, Meets Expectations, Meets Few Expectations, etc.). Six states combine two of these approaches. Some states, like Kentucky, also indicate whether a school’s summative performance rating represents an improvement, decline, or steady performance from that of previous years.

While we looked for the presence of any summative rating, states may want to use a hybrid approach that provides numerical and descriptive ratings of school performance, so stakeholders can have a clearer understanding. It’s paramount, though, to ensure that school leaders, educators, families, and communities understand and are bought into the rating system. This means state leaders should consult with people from around the state — rather than choose a summative rating approach on their own.

Encouragingly, 30 states allow comparisons across schools on their online report cards; however only 10 of these — among them, Maryland and Nebraska — let stakeholders compare schools with similar characteristics or include disaggregated data in comparisons — as Connecticut and North Carolina do. Both types of comparisons are important because they let school, district, and state leaders consider why schools with similar characteristics might have different performance ratings overall or among certain student groups. Considering such questions in the context of the services and opportunities schools are providing to students can help form the basis for improvement planning — while comparisons of dissimilar schools may be less valuable, since performance differences could be better explained by differing school populations or levels of available resources.

Asset-Framing

• School rating websites should use asset-based language to highlight the responsibility of the education system toward students, provide opportunities for schools to describe how they meet student needs in ways that may not be captured by data, recognize the impact of finite and inequitable resources on outcomes, and highlight school strengths without masking areas for growth.

Asset-based framing highlights structural barriers and patterns in the education system that contribute to low achievement among students. But asset-based framing is not required by ESSA and the push to use it is recent. It’s, therefore, unsurprising that just four states — Idaho, Kansas, North Dakota, and Oklahoma — include asset-based language in their online school report cards. North Dakota’s data, for instance, is preceded by a two-paragraph introduction that includes a statement that says, “North Dakota public education is rooted firmly in the belief that all schools should be engaged in a cycle of continuous improvement. … While no one report can tell the full story of North Dakota schools, the dashboard features information on important measures that inform stakeholders how to better support students….” Such contextual information can help build trust and understanding about the use and importance of school accountability data.

Bright Spots in the States

Using Asset-based language: Idaho, Kansas, North Dakota, and Oklahoma use asset-based language in their online school report cards. North Dakota’s introduction says, “North Dakota public education is rooted firmly in the belief that all schools should be engaged in a cycle of continuous improvement. … While no one report can tell the full story of North Dakota schools, the dashboard features information on important measures that inform stakeholders how to better support students … .” Such contextual information helps build trust and understanding about the use and importance of school accountability data.

Enabling school comparisons in online school report cards: Maryland and Nebraska let stakeholders compare schools with similar characteristics, while Connecticut and North Carolina include disaggregated data in comparisons. Both comparisons are important because they let school, district, and state leaders consider why schools with similar characteristics might have different performance ratings overall or among certain student groups.

Exit Criteria

Once schools are identified as CSI, TSI, or ATSI, leaders must work with their staff and communities, along with district and state support staff, to meaningfully improve their performance. Under ESSA, states must determine how much improvement schools must make to no longer be identified as needing improvement.

• States should set ambitious but attainable exit criteria that prioritize sustainable and meaningful improvement. This means schools must move substantially past the identified threshold, provide evidence of meaningful participation in school improvement activities, and show clear improvements in performance to set the conditions for sustainable, long-term improvement.

Unfortunately, not one state meets all our criteria, and just four states set exit criteria that require schools to have performance rates that exceed the cutoff point for classification as chronically low performing. Florida and Louisiana require schools to have an overall C rating to exit identification. CSI schools in Nebraska must outscore the bottom 10% of schools across multiple indicators (double the threshold for identification) and show progress across all indicators. Similarly, Tennessee requires schools to outscore the bottom 10% of schools over two years or the bottom 15% in a single year — though schools are exempt from this requirement if they demonstrate very high academic growth over two years.

While most states don’t set a high-performance bar, many have additional requirements. Thirty states require schools to show substantial growth across some or all indicators in their school rating system to exit identification status. Of these, 11 also require schools to participate in school improvement activities and develop school improvement plans.

Pennsylvania has particularly strong connections to the school improvement process in their exit criteria. In addition to requiring growth in achievement and SQSS measures, Pennsylvania’s exit criteria ask schools to create an improvement plan that explains how they will sustain growth in the absence of supports for identified schools and demonstrate sustained participation in the SEAs technical assistance opportunities for identified schools.

More Rigorous Interventions

Strong actions such as state takeovers have mixed evidence of success and when done need to be undertaken in collaboration with communities. ³⁶ However, having these state actions as options can help motivate school and district leaders to make strategic, systematic changes to policy and practice to raise performance. It can also help state leaders clearly outline how they will support or intervene when changes aren’t made. Additionally, it’s important to require next steps to hold districts and states accountable when they let schools remain in a state of stagnant performance.

• Clear next steps with a focus on improved student outcomes are crucial to bringing about substantial changes in the practice or control of school decision-making and ensuring that all students can attend a high-quality school.

States are also supposed to outline a progression of escalating interventions for schools that don’t meet exit criteria over a period set by the state. Unfortunately, the federal government only required states to provide minimal information about their plans for schools that fail to make adequate progress. As a result, most state ESSA plans lack specifics and include generic language that leaves the public with little to no understanding about the consequences or actions that will be taken when schools consistently underperform.

In their plans, states focused on two main categories of more rigorous interventions: 1) increased monitoring paired with additional supports and technical assistance; and 2) changes to a school’s governance or structure.

Nearly three-quarters of states indicated that schools that don’t meet exit criteria will receive additional supports and be subject to increased monitoring. While SEA supports and monitoring are important, such supports should ideally be implemented soon after schools are identified, rather than reserved as more rigorous interventions. Missouri, for instance, states that schools needing more rigorous interventions will be required to meet with school improvement specialists monthly, participate in principal coaching, and create 30-, 60-, and 90-day action plans that will be monitored by state staff.

In contrast, just 17 states indicate that schools that fail to meet exit criteria could be subject to more substantial changes to policy or school oversight, and of these, 11 list potential actions the state could take without committing to any specific intervention — or even intervention at all. Colorado, for example, lists potential actions, including management of the school by an external partner, conversion to a charter school, or school closure, without committing to using any of these strategies at identified schools that don’t make enough progress.

In fact, just six states detailed clearer, more specific consequences in their ESSA plans. Connecticut and Tennessee stand out for outlining comprehensive strategies for escalating supports and interventions for schools that don’t progress. Tennessee has specific tiers of support and interventions based on school performance and improvement, while Connecticut details progressively higher requirements, monitoring, and support for each year a school doesn’t meet the exit criteria, with the final tier including options for changes in school governance or closure. Yet, while these states have clearly defined state supports and interventions, they also have timelines that extend over 10 years, which allows for prolonged periods of low performance.

In contrast, Florida and Rhode Island have specific and immediate consequences for schools that fail to exit identification. In Florida, schools must engage in a two-year turnaround process, which can include school control moving to a charter or other external operator, while Rhode Island requires schools to undergo a total school-redesign process using one of five models, in lieu of the more piecemeal improvement efforts schools may have engaged in since being identified.

These more immediate consequences might create a greater sense of urgency to make changes to school practices but should be paired with the types of supports outlined by Connecticut and Tennessee.

School Improvement Funds

Being identified for improvement also unlocks access to federal school improvement funds. States are required to allocate 7% of their total Title I funding to identified schools. States have lots of flexibility about how these funds are allocated and can attach requirements for obtaining or using these resources. States can allocate these funds via a formula, a competitive process, or a combination of both.

• Ideally, states should take a combined approach to allocating school improvement funds. Providing all identified schools with small formula-based grants would let schools dedicate staff capacity specifically to school improvement planning and give schools that have the most need for support — and might otherwise lack the capacity to craft a well-developed grant application — access to improvement funding. Through competitive grants, states could guide or require identified schools to thoughtfully consider local data and choose evidence-based interventions and strategies.

Our investigation of SEA websites found that 37 states published information on how they are using their school improvement funds, while 27 published information on which schools are ultimately receiving these funds. Of the 37 with published information, approximately half use competitive grants (14 states) and the other half use formula grants (16 states). Just two states —Connecticut and Delaware — use our recommended combined approach. Additionally, two of the states using formula grants — California and Minnesota — have application processes that allow them to ensure strong use of funds without a competitive process, another potentially effective approach to allocating school improvement grants.

Indiana has an interesting approach to support schools that have limited capacity to develop school improvement grant applications: Schools that fail to meet the state’s competitive grant requirements receive guidance and support on strengthening their application and are eligible to reapply the following year. Texas is committed to building the capacity of school leaders in identified schools: Grant recipients are required to participate in leadership development and coaching that aligns with their planned improvement areas and utilization of improvement funds.

Michigan, meanwhile, has opted for a unique, and potentially misguided, approach to using its improvement funds. In lieu of allocating school improvement funds to identified schools, the state has opted to use these funds to staff SEA offices that focus on school improvement: such as the Office of P-20 Data and Information Management, Office of Safe and Supportive Schools, and Office of Education Improvement and Innovation. While it is important to build the capacity of SEA offices that support schools, these funds were specifically allocated to support identified schools via direct funding. Further investigation is needed to determine whether this unusual approach is moving the needle on performance in the state’s lowest-performing schools.

Resource Allocation Reviews

In addition to Title I set-aside funds for school improvement, ESSA contains another key provision to support school improvement efforts in identified schools: resource allocation reviews (RAR). Under this provision, states must “periodically” review how districts with a “significant” number of schools identified as CSI, TSI, or ATSI are allocating resources across schools. This process was intended to help districts consider how to redistribute funding and other resources in more equitable ways and ensure that schools and students with the most needs have access to adequate resources. However, the lack of clarity in the words “periodically” and “significant” has allowed states to effectively weaken this requirement. The Department of Education issued a Dear Colleague letter in July 2023 that contained clear, strong guidance on how states should conduct their resource allocation reviews, but this guidance is not mandatory.³⁷

• States should align the definition of a “significant” number of identified schools that is required to trigger a LEA resource allocation review with the department’s recommendation, which includes setting a percentage and a number threshold to ensure that small and large districts are identified for support. Ideally, states would set thresholds that strike a balance between identifying more districts than they can support and identifying so few that districts aren’t getting the support they need to better allocate resources.

Additionally, while ESSA doesn’t specify the types of resources that should be included in the reviews, the department’s guidance wisely suggests that states look beyond funding amounts and consider equitable access to other types of resources, such as experienced educators, rigorous coursework, and high-quality early learning opportunities. These considerations align with the Alliance for Resource Equity framework ³⁸  — designed by EdTrust and Education Resource Strategies — which some states, including Michigan, highlight in their state guidance documents on this process. ³⁹

Unfortunately, likely due to the lack of strong federal requirements, publicly available information on if and how states are leveraging resource allocation reviews is extremely limited. Just 14 states provide a clear threshold for identifying districts that are eligible for a review, and 15 states indicate that reviews will consider more than just funding, either within their ESSA plan or on their SEA website. Only four states — Michigan, Nebraska, Ohio, and Rhode Island — do both. It’s, therefore, imperative that advocates and the department push states to do more to ensure strong implementation of resource allocation reviews and transparency in the process.

When it comes to identifying districts for review, only Florida includes both a percentage and numerical threshold: 10% of schools or two identified schools. Five other states have a numerical threshold, but four of these states say that reviews will be triggered in districts with any identified schools. This approach would likely result in the identification of more districts than the SEA could effectively manage.

Seven states determine if a district contains enough identified schools to trigger an RAR using percentage thresholds that range from 10% to 50% of schools identified. States that have a threshold of 10% could identify more districts than the SEA could manage. Conversely, those using a threshold of 50% are likely missing many districts that may need help reallocating resource to support a large number of identified schools. Just two states, Nebraska and Ohio — which are, notably, two of the four states with the most comprehensive information on the process overall — strike a good balance with thresholds of 20% and 30%, respectively.

Bright Spots in the States

Outlining consequences & next steps for schools that fail to improve: In Florida, schools that fail to exit identification must engage in a two-year turnaround process, which can include school control moving to a charter or other external operator. In Rhode Island, those schools must undergo a total school-redesign, in lieu of more piecemeal improvement efforts schools may have been engaging in since being identified.

Connecticut and Tennessee outline comprehensive strategies for escalating supports and interventions for schools that don’t progress. Tennessee has specific tiers of support and interventions based on school performance and improvement, while Connecticut details progressively higher requirements, monitoring, and support for each year a school fails to meet the exit criteria, with the final tier including options for changes in school governance or closure.

Pennsylvania asks schools to create an improvement plan explaining how they will sustain growth in the absence of supports for identified schools and demonstrate sustained participation in the SEA’s technical assistance opportunities for identified schools.

Helping schools apply for improvement grants: In Indiana, schools that fail to meet the state’s competitive grant requirements receive guidance and support on strengthening their school improvement grant application and are eligible to reapply the following year. In Texas, school leaders in identified schools that receive improvement grants are required to participate in leadership development and coaching that aligns with their planned improvement areas and utilization of improvement funds.