Indicators: What to Include in School Ratings

Indicators: What to Include in School Ratings

In designing their school accountability systems, state leaders will have to wrestle with a number of big questions, including: What goals do we want to set for our schools? How will we determine which schools are doing well, and which are struggling? How will states and districts support struggling schools?

The very first question states will have to answer, however, is, “What information will we use to measure school performance?”

The purpose of these fact sheets is to provide advocates with some of the information they need to help make sure their state leaders select meaningful measures (called indicators) of how well schools are serving all students, especially low-income students, students of color, English learners, and students with disabilities.

  • This overview fact sheet lays out key requirements related to accountability indicators in the Every Student Succeeds Act, and identifies important things to keep in mind when deciding which measures should be part of a school’s rating.
  • Each indicator-specific fact sheet summarizes what research and data tell us about an indicator, highlights the benefits and risks associated with rating schools based on that indicator, and raises some questions to ask if your state is considering including the indicator in school ratings.
  • The indicator traffic light table uses the colors green, yellow, and red to suggest the level of confidence — or conversely, caution — that advocates should have when thinking about whether to include each measure in school ratings, in a needs assessment that follows the rating, and in public reporting, respectively.

What does ESSA require when it comes to indicators?

The Every Student Succeeds Act requires states to continue to measure certain things that are already part of most states’ accountability systems — for example, proficiency rates on state assessments and graduation rates. But the law also requires some new indicators.

Specifically, the law requires states to rate schools each year based on the following measures, all of which (save progress toward English-language proficiency) have to be disaggregated by student group:

  • Academic achievement: A measure of how schools’ proficiency rates in reading/language arts and math for all students and each student group compares with state-set goals. For high schools, states can also include student growth as part of this indicator.
  • Another academic indicator:
    • For high schools, a measure of how graduation rates for all students and each student group compare with State-set goals.
    • For elementary and middle schools, this measure may be individual student growth or another statewide, valid, and reliable indicator of student learning.
  • English-language proficiency: A measure of the progress that a school’s English learners are making toward English proficiency.
  • Additional indicator of school quality or academic success: Another valid, reliable, and statewide indicator of school quality, which may include measures of postsecondary readiness or student engagement. (Note: While this is described by some as the “non-academic” measure, the law clearly permits measures of academic success.)

Requirements for all indicators:
Any indicator a state uses has to “meaningfully differentiate” between schools. That is to say, all schools can’t show roughly the same results on the measure.

And importantly, a school’s rating has to be based on how the school is doing on all of these indicators for each group of students that it serves.

What are the minimum parameters an indicator should meet to be included in school ratings?

There are lots and lots of ideas out there right now about measures that states could include in school ratings. The questions below represent the minimum criteria that indicators should meet to be considered for inclusion in school ratings. If the answer to any question is “no,” that measure could be part of a needs assessment or a school report card but should not be part of a school rating.

  • Is the indicator focused on students? A school’s rating should reflect how schools are serving their students. While measures of staff satisfaction or teacher absenteeism may be important for a school’s needs assessment, for example, they should not affect a school’s rating.
  • Can the indicator be measured by student group? To help ensure that school ratings reflect how schools are doing for all groups of students, ESSA requires all indicators to be disaggregated by student group. Things that cannot be measured by student group — for example, the number of advanced classes offered at a school, or the percent of teachers who have a major or minor in the field they are teaching in — can be part of a school’s needs assessment and/or report card but cannot be part of its rating.
  • Is the indicator aligned with readiness for post high school success? A key goal of accountability is to push schools to focus on preparing all students for success beyond high school, particularly readiness for college and/or a meaningful career. All states should be able to show that any indicator included in school ratings is aligned with that ultimate goal.
  • Does the indicator differentiate between schools? This means that an indicator can’t make all schools look the same. Average daily attendance, for example, may not allow sufficient differentiation between schools since most schools report very high average daily attendance rates. Chronic absenteeism rates, on the other hand, differ much more from school to school and are more likely to meet this requirement.
  • Can the indicator hold the weight of accountability? When an indicator gets included in a school’s rating, schools have a strong incentive to look good on that measure. That’s generally a positive thing: The rating should incentivize schools to improve key student outcomes. But some measures can be “improved” without making meaningful changes for students — by, for example, misreporting results. While no measure is immune to gaming, some are more susceptible to it than others. To ensure that ratings reflect how well schools are serving their students, they should be based on indicators that are difficult to game.

What are some things to keep in mind when thinking about indicators?

  1. School ratings can do some things well, but they can’t do everything. Here’s what ratings do well:
    • Communicate expectations for preparing all groups of students for college and/or a meaningful career;
    • Communicate to educators, parents, and the public how schools are performing against these expectations, both overall and for each group of students; and
    • Signal that improvement is needed whenever outcomes for any group consistently don’t meet expectations.

    But on its own, a rating can’t tell educators what they need to do to improve student outcomes. Nor can it, on its own, tell parents everything they need to know about a school.

    Keeping this in mind is important when selecting measures for the rating system — otherwise, it’s tempting to include everything we could possibly measure about a school in a school’s rating.

  2. When we include too many measures in a school rating, there are too many targets, and nothing matters enough to really focus school improvement efforts.
    The opportunity to broaden school ratings beyond test scores and graduation rates can be exciting. Because ratings drive attention and resources, it’s tempting to include everything that could possibly be important about a school in the rating calculation. But there are real risks to this approach: With every addition, there is the probability of less attention to something else — and the very real possibility that schools will get “A” or 5Dstar ratings based on greater parent satisfaction or students’ “growth mindset,” even though student learning is low and declining. Instead of trying to include as many indicators as possible in a school’s rating, advocates will want to press for a limited number of high-quality measures that provide critical information about how schools are serving their students and direct schools to focus on what matters most.
  3. Some measures might be better reported on the school report card, but not included in a school rating.
    Parents, community members, and the public need to know a lot of things about schools. Excluding some measures from school ratings doesn’t mean that they aren’t important or shouldn’t be attended to. It just means that they’re not appropriate for school ratings. Some information — teacher quality or the incidence of school violence, for example — might be important to parents, but either can’t meet the disaggregation test or would likely lead to less honest reporting if included in school ratings because of the pressure to make the school look good. Such measures should be included in the needs assessment or on school report cards but should not factor into a school’s rating.1
  4. Adding indicators costs money — sometimes, lots of it.
    As we think about which information to include in school ratings and on school report cards, we need to keep in mind the costs associated with collecting that data and ensuring its accuracy. High-quality student, parent, and teacher surveys, for example, can be extremely expensive and may be more feasible as a tool for a subset of struggling schools. In making decisions about what to collect, advocates and state leaders will need to think hard about what matters most.
    Additionally, some important indicators may not be ready for inclusion in school ratings just yet. States should be open to adding high-quality measures in future iterations of their accountability systems.

1 The school improvement process should always begin with a needs assessment, which is a look at a broader range of data that helps to identify school-based causes of underperformance.