Is Accountability Holding Back High School Transformation?

March 31, 2018

By Rebecca Holmes and Elliott Asp  |  March 31, 2018

Imagine you are the principal of a high school with limited resources for tutoring. You have a group of kids more than three years below grade level, struggling to access content in all areas because of the gaps in their literacy skills. You have another group of kids who are doing pretty well in school, who have mastered many of the grade level standards, and with a little push, might score “proficient” on the state test. Who gets the tutors? Now, layer a public, high-stakes accountability system on top of that question. Would you make the same choice?

The power of school accountability systems to convey priorities and compel behavior is enormous. Because CEI is positioned at the hand-off from policy to practice, we often see unintended consequences and educators struggling with trade-offs like the one above. What we are eager to see and what we hope to encourage through this EdPaper series on accountability is a community-wide conversation about the mismatches in what we most want schools to do and what our current system has encouraged. Perhaps nowhere is this mismatch more glaring than in how the accountability system judges the quality of a high school.

Numerous efforts over the last 25 years have focused on reimagining the American high school (with very little wide-spread success), but today’s increasingly rapid pace of change in our economy and society have amplified the urgency driving us to reconsider what we want and need high schools to do for their students. There is truth to the narrative that our schools were designed for another era and that our high schools were originally designed for drop-out rates not unlike what we currently see. Camille Farrington’s Failing at School tells this history well and urges us to examine the design elements at play in our schools. When we decided that High School Redesign was one of CEI’s next “big bets,” we put an intentional focus on the “re.” Roughly 90% of young people in Colorado attend a high school that’s been around for at least a few decades. As Professor Jack Schneider, author of Beyond Test Scores, reminds us, in reimagining high school, it may be a false premise that you should “start from scratch.” “This idea might hold water if schools hadn’t been steadily evolving over the past four centuries. Yet schools are organic systems, always shifting and changing. They aren’t like clunky old computers that we might replace with sleek tablets. They are like natural environments that are always adapting in response to internal and external forces.” Getting the forces of accountability right, then, is critical to imagining high schools that are designed to reliably and equitably produce the student outcomes young people need to be successful in today’s society.

As the 16 high schools in our current High School Redesign learning network progress, we’ll be paying close attention to several implications for accountability:

  1. Academic growth matters. Some advocates encourage re-weighting the Colorado accountability system to more deeply value grade-level proficiency. What we hear time and again from school leaders, particularly those in secondary settings, is that valuing growth matters as much for the 9th-grader who reads at a 6th-grade level as it does for 9th-grader already ready for college-level content. How we construct growth ratings may still need tweaking, and it’s also worth asking why we hold high schools more accountable for performance in a performance-growth ratio than we do elementary and middle schools, and what behaviors that may be incentivizing.
  2. Expand our definition and measurement of postsecondary and workforce readiness (PWR). Colorado’s set of postsecondary and workforce readiness (PWR) indicators should include leading measures. The current lagging measures (graduation rates, drop-out rates, and mean SAT scores) are certainly important for public accountability. Reporting across subgroups of students on these indicators is critical for helping schools and communities face the realities of their racialized achievement gaps and the low performance of other groups who have been traditionally underserved by our current system (e.g., students with disabilities, English language learners and students living in poverty). However, holding schools accountable for these long-term outputs appears to do very little to drive continuous improvement on important leading indicators of school and student success. Moreover, we’ve too often seen schools employ short-term practices to improve these rates rather than a systemic approach to school improvement or redesign, particularly in service of closing “achievement gaps.” “Achievement gaps” are most often symptoms of systemic bias and opportunity to learn (OTL) gaps. Therefore, including OTL measures that are tied to the student outcomes we value could push schools and districts to systematically examine which kids take which classes (or pathways) and which teachers teach which kids. Finally, the graduation rate measure in Colorado, the only state in the U.S. without a single statewide set of graduation requirements, has long been a pain point for system leaders. Districts with more strict course requirements have been compared on this measure alongside their neighbors who have required fewer courses for graduation. The graduation guidelines that go into effect in 2021 raise new opportunities and new challenges for this metric. All that said, it is somewhat unique that the current Colorado system allows schools to earn framework points for the best of a five, six, and even seven-year graduation rate.  This is a feature we strongly recommend maintaining, as it recognizes that students can progress toward a common bar at a more individualized pace.
  3. Create a more meaningful balance between comparability and improvement. We can trade off some comparability in exchange for more robust, relevant, and actionable data. Our current accountability system does make it possible to provide a rating to each of the 500 Colorado high schools and sort them on a handful of data points. To what end? The lack of research across those schools and ratings itself is a signal that the system was never designed to drive a learning agenda or school improvement, particularly for the vast majority of schools that are not on the state clock.  More pragmatically, most Coloradoans live in a place where at most four or five of those high schools are legitimate options for their students. High school principals tell us repeatedly that the current accountability system fails to capture the broader vision of their school and the schools they most want to learn from are those up to similar work. We can imagine a system in which high schools are still accountable for core academic and success outcomes but beyond that select in to one of four or five sets of measures that best capture the results they are most working toward. For a subset of schools these may include Advanced Placement access and pass rates and persistence beyond the first year of a two or four-year college.  For another subset the measures may be focused on industry credentials, concurrent enrollment credit attainment, and the completion of high quality internship experiences. Others, particularly struggling schools, may choose additional youth engagement outcomes as leading indicators of systemic improvement, measured through survey tools like those included in 5Essentials.
  4. Students and parents are giving us a message when they opt-out of state testing. Widescale testing opt-out, especially at the high school level, deserves further investigation. School leaders tell us that what families are most protesting are assessments that didn’t feel relevant and an accountability regime their community did not value. What this can teach us about performance assessments and community driven measures is significant. Our current work on next generation assessment of learning has the CEI team studying the Somerville Framework and the Massachusetts Consortium for Innovative Education Assessment. The principles they outline for a more dynamic picture of student learning and school performance are compelling. While those principles may seem a distant stretch from the current Colorado system, piloting this approach to accountability in high schools seems like a plausible starting point.


The issues with growth, indicators of PWR, balancing comparability with improvement and testing opt-out are symptomatic of an accountability system built primarily for state purposes that is awkward and insufficient for local, community, and school improvement purposes. We think it is time to have a state-wide discussion about how our accountability system can encourage high schools to dig deeply into systemic precursors of PWR and identify root causes of the failure of some students to meet this bar. Until we do this in a way that allows and invites meaningful participation in redesigning the current system – from students, teachers, families, and school and district leaders – we will continue to see families opting out of the process and many schools paying little attention to the results. Our system will remain largely unhelpful in encouraging or informing continuous improvement.

Since January, CEI has engaged the education community in an ongoing dialogue about Colorado’s accountability system. We believe there is an exciting opportunity at hand to revisit our priorities and examine the limitations of our current system with an innovator’s mindset. Each month, we have published an EdPaper that takes a deeper look at the subject of school accountability. We invite you to be part of the conversation and share your thoughts with us via email at

JANUARY EdPaper: The Nature of Accountability is Ready for Change
FEBRUARY EdPaper: S-CAP (Student-Centered Accountability Project) and its Implications for Local Measures
MARCH EdPaper: High School Accountability
APRIL EdPaper: Accountability in a Competency-Based System
JUNE EdPaper: To Whom Are Schools Most Accountable?

We invite you to be part of the dialogue. Email your thoughts to join the conversation!