The

Iceberg
Problem
How Assessment and Accountability
Policies Cause Learning Gaps in Math
to Persist Below the Surface . . . and
What to Do About It

2019
Contents
ACKNOWLEDGMENTS

THE ICEBERG PROBLEM

EXECUTIVE SUMMARY 1

INTRODUCTION 6

KEY INSIGHT #1: Math is cumulative – unfinished 11

learning from prior years makes it harder for students
to master more advanced concepts

KEY INSIGHT #2: Current educational policies favor 22

grade-level instruction, and that may be hindering
students longer term success

KEY INSIGHT #3: Balancing pre-grade level, on grade- 35

level, and post-grade level skills to each student’s
needs can better support their longer-term success

RECOMMENDATIONS 43

CONCLUSION 49

APPENDIX I: METHODS 52

APPENDIX II: FOCUS GROUPS 55

APPENDIX III: EVOLUTION OF POLICY LANDSCAPE 57

ENDNOTES 63


Acknowledgments
We are grateful to the many policy makers, educators, researchers, and education-sector
leaders who contributed to our thinking on this project. Their willingness to set aside time
for robust and thoughtful discussions on policy, research, and practice substantially
enriched this work. We’re also particularly grateful to the teachers and school leaders who
voluntarily participated in focus groups, both live and virtually.

We are also grateful to Barr Foundation, Carnegie Corporation of New York, and the Nellie
Mae Education Foundation for their generous support of this publication and their
thought partnership on these issues. Many people within New Classrooms contributed to
the publication, especially Joel Rose, Chris Rush, Jennifer Stillman, Charles Voltz, and
Michael Watson. Team members from New Classrooms’ Board of Directors, External Wing,
Leadership Team, and Program Wing served as vital reviewers and editors throughout the
process. New Classrooms contracted with Bellwether Education Partners to provide
research and advisory support and conduct the focus groups.

None of this would be possible without our partner schools, teachers, and students, from
whom New Classrooms Innovation Partners learns every day.
ACKNOWLEDGMENTS

The following interviewees, partners, funders, advisors, reviewers, and external supports
helped make this publication a reality, but all views, content, and recommendations
expressed here, including any errors of fact or omissions, are those of New Classrooms
alone.

Sara Allan
Deputy Director for Education, Bill & Melinda Gates Foundation

Tony Alpert
Executive Director, Smarter Balanced Assessment Consortium

Jay Altman
Co-Founder, FirstLine Schools

Kristen Amundson
Former President and CEO, National School Boards Association

Karim Ani
Founder, Mathalicious

Robert M. Avossa
Ed.D., Senior Vice Presidentat LRP Media Group; Former Superintendent in Fulton County,
GA and Palm Beach, FL

John Bailey
Advisor to Walton Family Foundation and Visiting Fellow at AEI

Charlie Barone
Chief Policy Officer, Democrats for Education Reform

Justin Barra
Director, Education at Chan Zuckerberg Initiative

Dr. Damian W. Betebenner

Senior Associate, National Center for the Improvement of Educational Assessment

Jim Blew
Assistant Secretary, U.S. Department of Education

Jo Boaler
Professor of Mathematics Education, Stanford Graduate School of Education

Noah Bookman
Executive Director, CORE Data Collaborative

Jason Botel
Former Assistant Secretary, U.S. Department of Education

Andrea Castañeda
Chief Innovation Officer, Tulsa Public Schools

Chris Cerf
Former Commissioner of Education, State of New Jersey

Stacey Childress
CEO, NewSchools Venture Fund

Cliff Chuang
Senior Associate Commissioner, Massachusetts Department of Elementary and Secondary
Education

Peter Coe, Chief Academic Officer, UnboundEd

ACKNOWLEDGMENTS

Peter Coe
Chief Academic Officer, UnboundEd

Ted Coe
Director of Mathematics, Achieve

Michael Cohen
President, Achieve

Larry Cuban
Professor Emeritus of Education, Stanford University

Richard Culatta
Chief Executive Officer, ISTE

Linda Darling-Hammond
President and Chief Executive Officer, Learning Policy Institute

Dale Erquiaga
National President and Chief Executive Officer, Communities in Schools; Former
Education Commissioner, Nevada

Deborah Gist
Superintendent, Tulsa Public Schools

Joe Gleberman
Managing Director, The Pritzker Organization

Laura Hamilton
Senior Behavioral Scientist and Distinguished Chair in Learning and Assessment, the
RAND Corporation

Leah Hamilton
Director of Education, The Barr Foundation

Donna Harris-Aikens
Senior Director, Education Policy & Practice, National Education Association

Neil Heffernan
Professor of Computer Science, Worcester Polytechnic Institute and Co-Founder
ASSISTments

Andrew Hodge
Director of Math Innovation Zones, Texas Education Agency

Michael Horn
Co-Founder, Clayton Christensen Institute

Dr. Gisèle Huff

Executive Director, Jacqueline Hume Foundation

Abby Javurek
Senior Director, NWEA

Shavar Jeffries
President, Democrats for Education Reform

Lindsay Jones
President and CEO, National Center for Learning

Disabilities

David Keeling, Founding Partner, EdNavigator

ACKNOWLEDGMENTS

David Keeling
Founding Partner, EdNavigator

Angela Kennedy-Toon
Managing Partner, Education Elements

Sal Khan
Founder, Khan Academy

Anthony Kim
Founder and CEO, Education Elements

Kenneth Klau
Executive Office of Education, Commonwealth of Massachusetts

Joel Klein
Former Chancellor, New York City Department of Education

Wendy Kopp
CEO and Co-Founder, Teach for All

Lewis Leiboh
Senior Program Officer, U.S. Education, Bill & Melinda Gates Foundation

David Levin
Co-Founder, KIPP

Art Levine
Senior Fellow and President Emeritus, Woodrow Wilson Foundation

Michael Levine
Chief Knowledge Officer, Sesame Workshop

Michael Magee
Chief Executive Officer, Chiefs for Change

Carmel Martin
Former Assistant Secretary, U.S. Department of Education

Dan Meyer
Chief Academic Officer, Desmos

Chris Minnich
Chief Executive Officer, NWEA

Pedro Noguera
Distinguished Professor of Education at the Graduate School of Education and
Information Studies, UCLA

Lillian Pace
Vice President of Policy and Advocacy, KnowledgeWorks

John Pane
Senior Scientist, RAND Corporation

Todd Penner
Portfolio Director, Michael and Susan Dell Foundation

Mike Petrilli
President, Thomas P. Fordham Foundation

Carrie Heath Phillips, Senior Program Director of Student Transitions, Council of Chief
School
ACKNOWLEDGMENTS

Carrie Heath Phillips

Senior Program Director of Student Transitions, Council of Chief State School Officers

Stephen L. Pruitt, Ph.D.

President, Southern Regional Education Board (SREB)

Macke Raymond
Director, Center for Research on Education Outcomes (CREDO) at Stanford University

Jonah Rockoff
Professor of Economics, Columbia Business School

Jason Schweid
Executive Director, Research, Design, and Strategy, UnboundEd

Mora Segal
Chief Executive Officer, The Achievement Network

Joy Silvern
Independent Consultant; Former Deputy Chief of Staff, U.S. Department of Education

Joshua Starr
Chief Executive Officer, PDK International

Diane Tavenner
Co-Founder and Chief Executive Officer, Summit Public Schools

Saskia Thompson
Program Director, New Designs for Schools and Systems, the Carnegie Corporation of New
York

Thomas Toch
Director, FutureEd

Philip Uri Treisman

Professor of Mathematics, University of Texas at Austin

Marla Ucelli-Kashyap
Assistant to the President for Educational Issues, American Federation of Teachers

Daniel Weisberg
Chief Executive Officer, TNTP

Joanne Weiss
President, Weiss Associates; Former Chief of Staff to U.S. Secretary of Education Arne
Duncan

Richard J. Wenning
Executive Director, Be Foundation

Anne Wicks
Anne Kimball Johnson Director of the Education Reform Initiative, George W. Bush
Institute
The
Iceberg
Problem
The Iceberg Problem refers to
the observation that only a very
small amount of information is
available or visible about a
situation or phenomenon,
whereas the more
comprehensive information or
bulk of data remains hidden
from view.

It gets its name from the fact

that only about one-tenth of an
iceberg’s mass is seen outside
while about nine-tenths of it is
unseen, deep underneath the
water’s surface.
EXECUTIVE SUMMARY 1

Executive Summary

Ms. Rodriguez has many high hopes for her sixth-grade math students. She hopes they will
find joy in learning about the beauty and complexity of mathematical concepts and make
connections to the world around them. She hopes they perform well on the end-of-year
test so they are set up to succeed in the seventh- and eighth-grade courses designed to
prepare them for high school. She hopes they will excel in high school math, enroll and
succeed in college, and perhaps pursue a degree in science, technology, engineering, or
math. She hopes that these degrees will open up opportunities to pursue rewarding and
lucrative careers.

When the school year began, Ms. Rodriguez’s students arrived from six different
elementary schools. Since she didn’t have access to their grades or incoming state test
scores when the school year began, she was not sure what to expect. In the first few
weeks, she realized that of her class of 30 sixth-grade students, maybe five were keeping
up with grade-level work. She is now frustrated but not surprised that some of her lessons
don’t seem to stick. She tries her best to help her students understand what the sixth-
grade work is asking for, but some just seem lost. She wishes she had the time to work
with each of them one-on-one, to break down any misunderstandings and figure out what
they may have missed in the past.
 2

One day, sensing that many of her students were struggling with Operations on Decimals
because they hadn’t quite mastered Decimal Place Value in the fifth grade, she taught her
students a lesson on Decimal Place Value that she thought might help. (Decimal Place Value was
not included in the sixth-grade curriculum that her district adopted, so she found a lesson online
that she thought might work). That day her principal also happened to come in for a classroom
observation. In her post-observation conference, her principal told her to adhere to the grade-
level curriculum since that is what would be covered on the statewide summative test and would
thus serve as the basis for the school and district evaluation. There was little time to cover much
beyond that.

Ms. Rodriguez has high expectations for all of her students and believes that all of them are
capable of being ready for the rigors of high school math. But she does not see how they will
ever get there if she is unable to properly address her students’ unfinished learning from
elementary school. She is beginning to wonder if an exclusive focus on grade-level material is
truly what is best for each of her students.

This is just one example of the varied

challenges educators experience every day in
schools across the country—their hopes for
their students are high, but the tools teachers
have and the rules they are told to follow
often do not yield the results students need.

In our experience, the fastest way to

accelerate student learning is to provide
opportunities where students are challenged
at the appropriate level for their existing skills
and knowledge—not too easy, not too
difficult. A student might not be able to conquer a brand-new topic on their own, but with the right
supports,
conquer athey can learn
brand-new andon
topic retain
theirsomething
own, but new that was previously out of reach. This insight,
known
with theacross educational
right supports, and
they psychological
can learn and literature as the “zone of proximal development,”
undergirds many new
retain something widely used
that wascurricular
previouslyand instructional strategies.
out
of reach. This insight, known across
But policies and
frompsychological
district, state,literature
and federal
educational as educational authorities signal to them to focus their
instruction
the “zone ofonproximal
grade-level standards each year regardless of their students’ zones of proximal
development,”
development. Grade-based accountability systems are understood as necessary safeguards against
undergirds many widely used curricular and
inequity, but strategies.
they may also limit the potential for more effective, student-centric instructional
instructional
approaches that can better achieve college and career readiness for each student.
But policies from district, state, and federal
educational authorities signal to them to
focus their instruction on grade-level
EXECUTIVE SUMMARY 3

In developing this paper, we have drawn advocates, and researchers, including those
upon seven years of experience operating a with perspectives that differ from our own.
program called Teach to One: Math in We analyzed publicly available data and our
partnership with hundreds of teachers across own internal data on student progress. We
15 states, serving more than 40,000 students. examined results from focus groups with
Our work has enabled us to operate in middle school math teachers in three cities, in
schools governed by public school districts, schools both within and outside of our
charter school boards, and independent partner network, to hear directly about
entities in urban, suburban, and rural settings. teachers’ instructional strategies when
We have worked with students who are students come in with unfinished learning
behind grade-level expectations and with from prior years as well as teachers’
students who are ahead; with students who experiences with curriculum, assessment, and
qualify for special services; with English accountability.
learners; and with students from across a
variety of racial and ethnic groups. Working directly with districts and schools
across the country to address this challenge
Our perspective is further informed by a has given our organization a firsthand
concerted research and development effort perspective on the challenges faced by
we conducted that is focused on how best to educators to improve these outcomes. In
accelerate students through middle grade some communities, there are particular
math standards. As part of that effort, we challenges in recruiting, developing, and
have meticulously investigated the standards retaining high-quality math teachers, many of
and underlying concepts reflected at each whom might have more attractive
grade level, explored and tested the employment opportunities in other sectors. In
mathematical relationships among those other communities, ongoing leadership
concepts, and reviewed tens of thousands of transitions at the school or district level can
lessons that relate to those concepts. We lead to continual shifts in organizational
also analyzed the results of over 100,000 direction. Poverty-related issues such as
summative and formative assessments, trauma, violence, and nutrition are all, of
administered over six million assessments of course, highly relevant to student academic
our own, and partnered with universities and performance. So too are the expectations
research firms in order to advance our that adults have for students.
collective understanding of how students
learn math. While these and other factors undoubtedly
contribute to the challenges of preparing
This paper is not only based on the more students for high school math, we
experiences of our day-to-day work; it draws believe there is another consideration at play
upon existing research, policies, and that has gone relatively unnoticed by policy
literature. We conducted extensive makers: the underlying policy landscape itself
interviews with policy leaders, math experts, and its ultimate impact on teacher practice.
EXECUTIVE SUMMARY 4

We believe there is another consideration that has gone

unnoticed: the underlying policy landscape itself.

Based on our experience and publicly available research and data about middle school
math, we argue:

1. Math is cumulative. Unfinished learning from prior years makes it harder for
students to master more advanced concepts.

2. Policies incentivize an exclusive focus on grade-level instruction. Current

education policies signal to educators to focus their instruction on annual
grade-level standards regardless of individual student needs.

3. This approach is hindering college and career readiness. An instructional

focus on grade-level instruction keeps students from addressing the
unfinished learning from prior school years that is required to master more
advanced concepts.

To be clear, this is not a call to reverse the principles of standards, accountability, rigor,
transparency, and equity that undergird the Every Student Succeeds Act (ESSA). They
are essential elements for building a school system worthy of the students they serve. Our
education system gained significantly from the development of these systems, and they
are substantial accomplishments.

But these accomplishments cannot be the end. Even under the most optimistic of
circumstances, it would take decades for our schools to ultimately achieve the vision of
every child succeeding. If our nation is to ever have an educational system that can enable
all students to unlock their full potential, we will need new ideas and approaches to get
there.

This is a call to federal, state, and local leaders to create the space within ESSA for more
innovative approaches to learning and measurement that allow for students to take
different paths to the same outcome of college and career readiness. While ESSA
provides states with far more flexibility than was permitted under No Child Left Behind
(NCLB), the primary growth measures used for purposes of accountability are limited by
the fact they are confined to the narrow band of each grade’s standards and assessments.
So long as that single path defines the benchmark of success, it is unlikely that
approaches to learning that accelerate students from their unique performance levels can
be successful.
 EXECUTIVE SUMMARY 5

To accelerate math achievement, opportunity, and equity, this paper urges federal,
state, and local education leaders to:

! Measure learning growth through the use of assessments that cover standards from
across multiple grade levels.
! Modify accountability systems in order to incentivize instructional practices that
best support each student’s ability to accelerate to grade level and beyond.
! Launch Math Innovation Zones.
! Make available high-quality instructional supports and strategies that account for
unfinished learning from prior school years.
! Advance a future vision for assessment and accountability that incorporates more
precise measures of student learning growth.

Our goals for this paper are to push

foward the conversation about This paper is NOT arguing:
assessment, accountability, innovation,
and student learning, and to find ways
1. that high expectations and academic rigor
to resolve growing tensions between are unimportant. They are essential.
grade-based accountability systems and
2. that our nation’s system of schooling is free
more personalized approaches to from systemic biases. These biases are
instruction. The resolution must pervasive, and our educational policies must
prioritize high, rigorous standards and target overcoming them.
protect against systemic bias so that 3. that standards-based reform and related
students from every community can accountabilities should be eliminated. They
benefit from all the opportunities that are key building blocks to future progress.
come with a college- and career-ready 4. that proficiency doesn’t matter. It does.
education. We welcome perspectives However, for some, achieving college and
from others in the field, including those career readiness in the long term requires
building key foundational skills in order to
who may disagree with our assessment
get there.
of the problem and potential solutions.
And we recognize that there are real 5. that students should not learn any grade-
level content. Grade-level exposure matters,
tensions here, as there are in almost but an exclusive focus on grade-level
every policy-related educational issue. material can keep some students from filling
critical pre-grade gaps and others from
Nonetheless, it is time to honestly accelerating beyond grade-level
confront the challenges facing too many expectations.
schools and work together toward an 6. that the recommendations in this paper
educational system where every student apply to anything other than middle grade
math. That’s just what we know best.
can reach his or her full potential.
 INTRODUCTION 6

Introduction
Jobs requiring science, technology, positions.3 National and international
engineering, and math (STEM) skills are measures show mediocre performance and
among the fastest growing and highest minimal systemic growth for American
paying in the country—average wages in students in math. These results are
STEM fields are double that of non-STEM exacerbated for students of color and low-
occupations.1 income students, for whom large, persistent
achievement gaps indicate systemic failures
By 2022, the Bureau of Labor Statistics and unacceptable inequities.
predicts that there will be more than one
million open STEM jobs added to the US
economy.2 These jobs are likely to be the
engines of economic growth and opportunity National and international
for decades to come, a vital American
bulwark against offshoring, and they require measures show mediocre
a strong understanding of math starting as performance and minimal
early as possible.
systemic growth for
But in too many schools, students regularly
miss out on learning the mathematical skills
American students in
and concepts required to attain these math.
INTRODUCTION 7

Struggles with math are particularly acute in

Figure 1
middle school grades, where millions of
students enter with unfinished learning from
elementary school that can lead to their
% At or Above Proficient
falling further behind as material becomes
more complex and learning gaps accumulate.4

Struggles with math are

particularly acute in
middle school grades.

Once students fall behind in math, most

schools are ill-equipped to help them rebuild
what they missed and simultaneously keep
up with grade-level standards. As a result,
unfinished learning accumulates such that by
the time they enter high school, many are
unprepared for mathematical coursework
that will keep them on track for high school
graduation and postsecondary success.

On the most recent administration of the

National Assessment of Educational Progress
(NAEP), only 33 percent of American
students in eighth grade met the proficiency
level or above in math, with historically
disadvantaged student groups performing at
less than half the rate of white students.5

Standards-based reform and accountability

strategies arose in recognition of these
unacceptable inequities in education
achievement and opportunity. Federal
legislation adopted in 2001 under NCLB and
amended in 2015 under the ESSA requires
INTRODUCTION 8

each state to administer annual math and reading tests aligned with grade-level standards
for all students annually in grades three through eight and at least once in high school.
The cumulative impact of recent decades of assessment, standards, and accountability
reforms has yielded progress in several areas, including

• far greater transparency into achievement gaps between student subgroups;

• increased clarity for teachers on what students need in order to stay on a

college and career-ready trajectory;

• improved consistency in the setting of high expectations for every student

based on college and career readiness; and

• more objective information for families on whether students are reaching key
educational milestones.

More important, NAEP scores in math improved in the early years following the adoption
of NCLB.6 But over the last decade, NAEP scores in math have been relatively flat while
equity gaps have remained wide and fairly static.7 ACT math scores recently hit a twenty-
year low8 while SAT scores reflect a similar decline.9 Policy makers can fairly debate the
myriad factors that go into student performance trends and the overall impact of the law
itself, but few could credibly argue that our nation’s current approach to teaching
students math is systematically succeeding.

This paper examines accountability and policy challenges affecting instruction through
the lens of middle school math (usually, grades six through eight, and often including
algebra). Middle grade math is an important focal point because middle schools are
responsible for educating students who may have already fallen substantially behind, and
educators have just a few years to lift all their students to a high school level. In some
districts, middle school performance helps determine high school choices for students,
further amplifying the importance of a high-quality middle school experience for all
students.

As we have engaged in this work, what has become clearer to us is an acute tension
between an instructional program that is best for each student to ensure they are ready
for college and career and an underlying policy context rooted in grade-level
expectations. The mathematical skills required for students to engage with grade-level
material in middle school and high school are built upon a deep, conceptual
understanding from previous years. Yet while many students arrive at middle school
without these foundational skills, state and federal policy systems incentivize teaching to
INTRODUCTION 9

grade-level expectations in order to curtail It is important to note that the findings and
low expectations and inequitable outcomes. recommendations in this paper are exclusive
We do not see strong evidence in the field or to the grade span and domain we are most
in the research that in math, strict adherence familiar with: middle school math instruction.
to grade-level content in order to accelerate The cumulative nature of math, combined
learning works for all students. with unfinished learning that many students
bring with them from elementary school,
create a unique challenge for our nation’s
middle school math teachers. We encourage
Policy can set the stage others with experience in other grade spans
or domains to explore how these findings and
for even greater recommendations apply in other areas, if
transformation. at all.

Figure 2
Moreover, we have seen in our work that
Demographic Profile of Students
individualized instruction and high Participating in Teach to One: Math
expectations can go hand in hand, and that if 2018–19
we are able to identify and address
unfinished learning from prior years, students
can advance more quickly and successfully Students %
toward college- and career-ready goals.
White 14%
Although our current education policy
arrangements evolved in response to
Black 28%
profound educational inequities, they require
refinement so that the resulting incentives
Hispanic 49%
and measures do not create new barriers to
student success. While policy alone will not
Asian/Pacific Islander 2%
solve all challenges in the classroom, policy
can set the stage for even greater
American Indian/Alaskan Native 4%
transformation. Our hope is for a new
perspective on accountability that preserves
Two or More Races 3%
rigor, transparency, and equity, while also
creating the space for new approaches to
learning that have the potential to achieve Free/Reduced Lunch 76%

better results for all students.

INTRODUCTION 10

New Classrooms’ Teach to One: Math

New Classrooms’ Teach to One: Math is a personalized learning model that

schools adopt as a core or supplemental mathematics program for students in
grades 5–11. Teach to One: Math leverages the power and potential of multiple
educators working together to give each student a targeted, individualized
learning experience that enables them to learn with different groups of students
and in a variety of instructional modalities, including teacher-led instruction,
collaborative learning, and independent study.

Each year, the program generates a personalized curriculum that includes a set
of mathematical skills and concepts connecting where students are to high
academic standards. Students experience that curriculum through a
combination of teacher-led, collaborative, and independent learning modalities
designed to both build mathematical fluency and habits for lifelong success.
Each day, a scheduling algorithm leverages up-to-date student performance
data and other research in order to generate a recommended daily schedule for
each student. This allows students to progress at their own pace and to take
more ownership over their own learning.

More information about Teach to One: Math can be found at newclassrooms.org.

KEY INSIGHT #1 11

Key Insight #1:

Math is cumulative—unfinished learning from prior years makes it
harder for students to master more advanced concepts.

Mathematical skills build upon one another over time.

Mathematical concepts and skills build upon one another as students advance through
middle school. The instruction that students receive reflects a coherent body of
knowledge made up of interconnected concepts and designed around coherent
progressions from grade to grade so that students can build new understanding onto
foundations built in previous years.10

Through our work, we have continually leveraged academic research on how students
effectively progress through the K–12 mathematics landscape in order to map specific
mathematical concepts and skills to the college- and career-ready grade-level standards.11
In doing so, we see that each grade’s set of skills require students to have knowledge of
prerequisite skills from prior years.
KEY INSIGHT #1 12

For example, as reflected in Figure 3, when a Figure 4

seventh grader learns about Operations with

Rational Numbers, the instruction builds upon
predecessors from prior grades, including
Understanding Integers and Rational
Numbers (sixth grade), Operations with
Decimals (fifth and sixth grade), and
Operations with Fractions (fifth grade).

Figure 3

Similarly, as reflected in Figure 4, Applying

Percents (also a seventh-grade skill) builds
upon predecessor skills from prior grades
such as Understanding Percents (sixth grade),
Ratios and Ratio Reasoning (sixth grade), and
Multiplying Decimals (fifth grade).

A link to the Major Concept Map that undergirds Teach to One: Math can be found in the endnotes.12
This map reflects just one way that college and career standards can be converted to more distinct
and interrelated mathematical concepts and skills, and after extensive review and analysis, has
been validated by Professor Neil Heffernan at Worcester Polytechnic Institute.
KEY INSIGHT #1 13

Many students enter middle school with significant learning gaps from their
time in elementary school.

Middle school educators must confront the reality that students come into their
classrooms with a broad range of math skills and knowledge and often with substantial
unfinished learning from prior school years. Figure 5 below shows fourth-grade NAEP
math results for the past three test administrations. In 2017, one in five fourth graders fell
into the lowest tier of math performance, well below grade level.

Figure 5
NAEP Fourth Grade Math Achievement
2013–17

NAEP does not administer tests in fifth or sixth grades, which would allow for nationally
representative estimates of math performance when students typically enter middle
school course work, but fourth grade performance is a useful reference point for the large
number of students nationally who enter middle school in need of significant academic
supports.

There are many reasons why large numbers of students fall behind in math by the time
they enter middle school, through no fault of their own. Research on elementary school
math instruction found that different schools and individual teachers vary widely in how
they spend time teaching different areas of math content.13 Significant numbers of
elementary school educators (and educators in other grades) struggle with math
themselves and thus report difficulties teaching math to their students.14 While these data
suggest the need for deeper analysis and reform in elementary school math that falls
outside the scope of this paper, middle schools must nonetheless serve the students who
arrive at their doors, regardless of their starting point.
KEY INSIGHT #1 14

Achieving grade-level proficiency requires both filling pre-grade gaps and

mastering grade-level material.

In traditional school models, students spend most of their time working as a group on the
same skills, and success is measured based on annual state tests. Figure 6 demonstrates
just how problematic that approach can be for students who fall far below grade level.

The chart reflects another way of presenting data from the Teach to One: Math Major
Concept Map. The darker boxes at the top of each column represent math skills that are
included on the state-mandated summative test for each grade level. In the sixth grade,
the summative test can cover approximately 44 skills, including, for example, how to
evaluate numerical expressions. But students won’t know how to evaluate numerical
expressions if they never learned how to multiply and divide large numbers in prior
grades, as reflected by the underlying yellow boxes. Mastering the 44 sixth-grade skills,
for example, requires knowing 29 out of the 37 fifth-grade skills as well as 34 skills from
before fifth grade.

Figure 6

Challenges arise when students don’t fully master prerequisite skills in previous school
years. This unfinished learning makes it difficult for students to learn more complex
KEY INSIGHT #1 15

concepts in subsequent years. It is not a simple

Unfinished learning
matter of catching up from the previous year
either; in some cases, critical prerequisite skills makes it difficult for
may have been introduced several years prior
students to learn more
and are revisited in increasingly complex
applications over time. When students do not complex concepts in
master the simpler applications, these learning
subsequent years.
gaps can accumulate if students’ future
classroom experiences fail to recognize this
fundamental challenge.

Many middle school math teachers have too little time in a single year to both
cover grade-level material and address students’ unfinished learning from
prior years.

The comprehensiveness of the math standards themselves generally require teachers to

dedicate the full school year to fully cover all of the grade-level topics. For example, the
44 skills and concepts reflected in the sixth grade include topics such as Ratio Reasoning,
Understanding Percents, and Understanding and Solving Simple Equations—each of which
can take anywhere from three to four days to adequately cover (for a total of 132 to 176
school days). Over a 180-day school year, teachers may have 160 to 170 of those days for
regular classroom instruction when one considers schoolwide events, testing, teacher
absences, snow days, and other matters that can disrupt regular instruction. Even in a
class with mostly on-grade students, teachers can struggle to cover all of grade-level
topics during that time period.

But when a student starts the school year with unfinished learning from prior years, the
challenge of both covering grade-level material and strategically addressing unfinished
learning can become even more daunting. For example, eighth graders are expected to
learn about Multi-Step Equations during the course of the school year, though some
students begin the school year not having mastered critical predecessor skills such as
Simple Equations, Operations on Rational Numbers, or Adding and Subtracting Algebraic
Expressions. Each of these topics could readily take three to four days to sufficiently
cover—and they are only the predecessors for one grade-level skill. Without sufficient
time to both adequately address unfinished learning and cover grade-level material, more
learning gaps can readily accumulate.

The ground some students must cover in a single year in order to attain
grade-level proficiency is substantial and improbable.
KEY INSIGHT #1 16

For students who enter a school year having demonstrated proficiency in prior years,
achieving proficiency in the subsequent year’s curriculum is both appropriate and
reasonable. But for students who come into middle school multiple years behind, the
teachers’ challenge of both addressing unfinished learning while also successfully teaching
all grade-level content is profound.

To illustrate the magnitude of this challenge, we reviewed the NWEA MAP assessment
scores of incoming sixth-grade students served through Teach to One: Math. The MAP
assessment is an adaptive assessment that measures student performance in ways
agnostic to students’ enrolled grade levels, but includes benchmarks for learning growth
and performance at each grade level. MAP uses a measure called the RIT scale, which
spans across grades to allow for more straightforward growth comparisons.15

For the purposes of this report, we divided student scores into four categories based on
NWEA’s norms and linking studies with statewide assessments aligned to college- and
career-ready standards.16 We define the achievement categories as students who began
their sixth-grade school year:

• four or more years below grade-level standards;

• two and three years below grade-level standards;

• one year below grade-level standards, at grade-level standards, and one year
above grade-level standards; and

• two or more years above grade-level standards.

Among the population of students who have participated

in Teach to One: Math over the past four years in grades
six through eight, at the beginning of the sixth grade
school year, 9 percent of students began the school year
four or more years below, 56 percent began the school
year two or three years below, 33 percent of students
began either one year below, at grade level, or one year
above, and only 2 percent begin the school year two or
more years above grade level.17 More details on these
categories and calculations are available in Appendix I.

How much does a student who is four or more years

below grade level standards need to progress during
their middle school years to be ready for high school? To
 KEY INSIGHT #1 17

calculate this, we used the concordance tables provided by NWEA that link students’
eighth-grade spring RIT score on MAP to the corresponding PARCC readiness levels
(similar studies are available for other state assessments).18 Because the students we
serve all take the MAP when they enter the sixth grade, we are able to calculate the three-
year RIT gains they must achieve in order to reach proficiency on the eighth-grade
PARCC.

Figure 7 shows that students who began sixth grade four or more years below grade level
would have to grow by 63 RIT points over three years in order to reach a level where they
are 80 percent likely to pass a rigorous state test such as PARCC at the end of eighth
grade. Students who were two or three years behind must grow 40 RIT points over three
years in order to meet the same bar. As the average student typically gains 13 RIT points
during the same three-year period, achieving these benchmarks would require growing
at a rate three to five times the national average.19

Figure 7

20

*Based on NWEA, “Linking Studies,” nwea.org/resource/

type/linking-studies/, accessed November 1, 2018.  Growth
required for 50% chance of proficiency are comparable:
 Way Below: 63; Below: 40; On: 22; Above: 3.

**See methodology included in “Three-Year Growth at

Schools Using Teach to One: Math” in Appendix I.

*Growth Required

National average 3-Year RIT Growth

4 or More Years 2 or 3 Years Below 1 Year Below, 2 or More Years

Below At Grade Level, Above
1 Year Above

**Incoming Performance Level Relative to Grade Level Standards

While the population of students in our partner schools is not representative of the nation
as a whole (76 percent of students qualify for a free or reduced lunch versus 52 percent
nationally21), the need for our nation to develop viable pathways that enable students to
catch up and achieve college and career readiness is no less profound. If policy and
instructional systems are built around the assumption that proficiency can always be
reached in a single year and few students will need to substantially revisit and rebuild
prior year’s content, gaps in high school readiness will persist for many years to come.
KEY INSIGHT #1 18

The profound challenge and improbability of students catching up in middle school math
was further reinforced in a recent policy brief published by the Institute for Education
Policy at Johns Hopkins University. Using publicly available state assessment data,
researchers analyzed sixth- and eighth-grade cohort data from 1,651 schools across six
states and the District of Columbia and found less than 1 percent of schools were able to
improve their proficiency levels by at least 30 percentage points.22 The study also found
that less than 1 percent of schools were able to both consistently improve their
proficiency levels (by any amount) between sixth and eighth grade while also reducing
the number of students in the lowest performing category.23

Teachers feel the tension between grade-level instruction and individual

student needs.

“It’s hard for a student who has shown progress and is growing to not be able to see that

growth on the test that matters most. It affects our students who are consistently scoring

the lowest and struggling the most.” —Middle school math teacher

In our experience, the fastest way to

With the right
accelerate student learning is to provide
opportunities where students are challenged supports, students can
at the appropriate level for their existing skills
learn something new
and knowledge—not too easy, not too
difficult. A student might not be able to that was previously out
conquer a brand-new topic on their own, but
of reach.
with the right supports, they can learn and
retain something new that was previously out
of reach. This insight, known across educational and psychological literature as the “zone
of proximal development,” undergirds many widely used curricular and instructional
strategies.24 Examples of strategies that can bring a new skill into students’ ideal learning
zone include “scaffolding” different forms of assistance around a concept until students
can eventually complete the problem on their own25 or “spiraling” curriculum so that
teachers revisit and deepen fundamental concepts frequently.26

But what happens when that ideal zone for learning falls outside the menu of topics in a
students’ grade-assigned level? Or when the scaffolding that would be required for
students to access grade-level material is both so vast and nuanced to each student’s
unique circumstances that these strategies are effectively impractical? How do teachers
approach this challenge, and how (if at all) do district, state, and federal policies support
them?
KEY INSIGHT #1 19

“Students’ feelings about math are tied to their success in math. As soon as you start

giving them opportunites to succeed, and they see the growth, their feelings change.”

—Middle School Math Teacher

As part of this project, New Classrooms contracted with Bellwether Education Partners to
conduct focus groups with teachers in and out of the New Classrooms network to hear
directly from educators about their experiences with standards, assessment, and
accountability systems, and to learn how these systems might play out in different school
contexts.27 In each of the focus groups, we heard consistently that teachers shared a
strong belief in the need for students to reach high expectations, but many feel caught
between an expectation to focus only on grade-level materials and the kinds of skills and
knowledge they think their students may need. One teacher expressed the tension this
way: “The curriculum we were given says the kids should already know everything up to
their grade, and they don’t. I was even told a couple of times when I first started that I was
teaching below-grade work, and I should be doing on-grade work. But the students
weren’t ready for that yet.”

Through the course of this research, one barrier to effective personalized learning
approaches that surfaced is access to high-quality materials from across multiple grade
levels and supports in service of accessing grade-level content. Many participating
teachers did not feel fully confident teaching skills outside of the grade-level standards in
which they were most familiar. Teachers generally receive little training in how to
diagnose and address unfinished learning. Focus group participants pointed to two
primary strategies in response to large gaps in student knowledge: stick exclusively to on-
grade content and hope for the best, or spend substantial time independently searching
for off-grade instructional resources that may or may not be of high quality.

Teachers were rarely fully satisfied with available resources on how to help students who
are far behind. Because many states and educational advocates define exclusive
adherence to grade-level standards as a primary determinant of curriculum quality,
teachers who want to provide instruction on pre-grade skills often fend for themselves to
find materials for their classroom or devise their own strategies that may not be effective.

Failing to adequately address unfinished learning in middle school has long-

term consequences

Longitudinal studies of individual students over time can show more precisely how
students who fall behind are likely to stay behind. For example, Figure 8 shows some of
the findings from a 2012 study conducted by ACT.28 Researchers tracked results from
KEY INSIGHT #1 20

tens of thousands of students, one group from fourth grade to eighth grade, and another
from eighth grade to twelfth grade. In both groups, the vast majority of students who
started out behind in math stayed behind. Chances of catching up were even worse in the
upper grades: a student who was “far off track” in eighth grade math had only a 3 percent
chance of reaching college readiness by the end of high school.29 This same study also
found that students in the “far off track” group were much more likely to attend high-
poverty schools.

One reason why it may be difficult for students who come to high school off track to
catch up to their peers may have to do with the way in which high school coursework is
aligned to graduation requirements. The courses that students must pass as a condition to
high school graduation generally require predecessor knowledge from middle school.
However, high schools may not have these courses available since they generally would
not qualify for high school credits. As a result, students have little choice but to take high
KEY INSIGHT #1 21

school courses they are not yet ready for—often multiple times. A 2014 transcript study by
Phil Daro for the SERP Institute found that in one district, students commonly repeated
courses such as Algebra I or Algebra II two or three times and that only 5 percent of
students took a presumably normal sequence of courses from Algebra I through Calculus.30

As students struggle to pass credit-bearing high school courses, they are often allowed to
participate in online credit recovery programs. The quality of these programs varies widely,
and researchers have begun to question whether their popularity is playing a key role in
the fact that national high school graduation rates have increased while objective
indicators of student performance remain relatively unmoved.31
KEY INSIGHT #2 22

Key Insight #2
Current educational policies favoring grade-level instruction are
hindering many students’ longer-term success.

Education policies are oriented around annual grade-level expectations.

At the core of today’s federal and state educational policies for K–12 schools is a system oriented
around annual expectations at each grade level and a set of standards, assessments, and
accountabilities designed to drive instructional behavior toward meeting those annual
expectations. Appendix III provides a brief overview of how assessment and accountability policies
have evolved since A Nation at Risk was published 1983.

These policies aim to advance several important principles, including

• providing every student with access to rigorous grade-level instruction that

will prepare them for the future;

• closing achievement gaps and combating systemic bias against historically

marginalized student groups by providing equitable learning opportunities
and setting clear and common expectations for success;

• holding adults in school systems accountable for ambitious, measurable

learning outcomes rather than process-oriented inputs;
KEY INSIGHT #2 23

• giving families annual information they need to understand their student’s

progress, make educational decisions for their student, and advocate for
change; and

• giving policy makers and system leaders information they need on an annual
basis to evaluate school success and address areas for improvement.

The grade-level standards that lie at the heart of these policies are anchored in college
and career readiness. ESSA requires states to adopt “challenging state academic
standards” that apply equally to all public school students in the state in math, English
language arts, science, and any other subjects the state designates.32 States should
demonstrate the rigor of their standards by aligning with entrance requirements for
credit-bearing postsecondary coursework.

The shift to more rigorous

college- and career-ready
standards has been one of
the biggest education policy
developments of the past
decade. Well-designed
college- and career-ready Annual
standards show a logical Grade-Level
progression of skills and
Expectations
knowledge from one grade
to the next, which, in an ideal
world, all children would
follow to the letter. They are
also intended to guide the
instruction that students
receive each year.

But in math, when students miss key steps along the way in this progression or learn at a
pace that is faster or slower than the state standards anticipate, the standards alone do
not provide guidance to teachers on where to focus instruction. They signal to a seventh-
grade teacher, for example, that all seventh-grade students should be taught seventh-
grade content—whether they happen to be performing two years behind grade level or
two years ahead.

The grade-level expectations embedded in policy reflect a single path to college and
career readiness for all students. They leave little room for other instructional paths to
that same goal that are more attuned to each student’s incoming performance level.
KEY INSIGHT #2 24

Summative assessments are not State assessments

structured to measure the full range of
learning growth. are unlikely to detect
Under ESSA, all students in grades three
many of the learning
through eight must take a statewide gains that students
summative assessment aligned to their
enrolled grade level. All sixth graders take
might make from
the sixth-grade test, all seventh graders take pre-grade skills.
the seventh-grade test, and so forth. ESSA
allows for only a narrow set of exceptions to
these guidelines.33 to reduce test time and length, most
summative state assessments are almost
The tests must assess student performance
exclusively focused on grade-level content.
against the standards for their grade level,
and only on-grade measures can fulfill ESSA’s As a result, state assessments are unlikely to
requirements for accountability purposes (as detect many of the learning gains that
discussed in the next section of this paper). students might make from pre-grade skills.
This means there must be enough The gains, for example, made by a seventh-
assessment items aligned to grade-level grade student entering the school year three
standards so that each student’s years behind her peers and finishing the
performance can be measured relative to school year one year behind would likely not
grade-level expectations. The assessments be fully reflected on the seventh-grade
themselves, which generally comprise 30–50 assessment, since the test itself would likely
questions of varying complexity, must be not ask her to demonstrate knowledge on
valid, reliable, and administered statewide. topics beyond her assigned grade level.

ESSA includes provisions that permit states The same holds true for advanced students:
to adopt assessments that also “measure Learning gains made by a seventh-grade
academic proficiency and growth using items student who happens to progress into eighth
above or below the student’s grade level.”34 grade are also not likely to be detected as
However, recently issued guidance from the growth, since only seventh-grade items will
Department of Education specifies that if appear on her state summative assessment.
states design tests to include additional
measures of off-grade performance, they Moreover, while many of the skills that
must still measure and score students’ on- appear on each grade-level test build upon
grade performance accurately, and any off- knowledge from the previous year, entire sets
grade measures would not apply to primary of new skills are also introduced. For
academic indicators under ESSA.35 As a example, a test focused on seventh-grade
practical matter, due to the broad set of standards would generally include topics
standards at each grade level and pressures related to Probability that have little to do
with what is required to succeed in the eighth
grade. Similarly, there are a number of
eighth-grade skills such as those related to
KEY INSIGHT #2 25

with what is required to succeed in the eighth grade. Similarly, there are a number of
eighth-grade skills, such as those related to Geometric Transformations and Scientific
Notation, that do not necessarily build on anything covered in the seventh grade. While
the standards are written to be cumulative, not all math skills within a given grade are
necessarily applicable to more advanced concepts in subsequent grades, as Figure 10
demonstrates.

Responsible technical decisions in test design can ensure that proficiency designations
from year to year follow a logical pattern, link well to the tests before and after, and
reflect standards, but they cannot change the fact that many of the underlying tested
skills are different year to year. This reality can lead to misinterpretation about the
learning progress that students actually make from one year to the next.

Figure 10 reflects how an annual focus on grade-level proficiency can cause skill gaps to
grow in ways that make it harder for some students to achieve college and career
readiness. When a sixth-grade student is taught sixth-grade material, some of those skills
will be learned and some will go “unlearned” for a variety of reasons (e.g., lack of
predecessor knowledge, uneven teacher quality, student absences). The next year, as the
focus of accountability shifts to the seventh-grade assessment, the unlearned skills from
sixth grade remain unaddressed, even though those very skills may be essential to
mastering seventh-grade content. By eighth grade, even more learning gaps accumulate
so that by the time student enters high school, he is simply unprepared for more
advanced mathematical topics.
 26
Figure 10
How Learning Gaps Accumulate Over Time
6th-Grade School Year 7th-Grade School Year 8th-Grade School Year
4
Performance Level

6th-Grade 7th-Grade 8th-Grade

Skills Learned Skills Learned Skills Learned
Cumulative Skills Not Learned

6th-Grade
Skills Not Learned
7th-Grade
Skills Not Learned
8th-Grade
Skills Not Learned

6th-Grade
Skills Not Learned

7th-Grade
Skills Not Learned

Modest declines in proficiency levels can

mask a dramatic accumulation of
unfinished learning.
6th-Grade
Skills Not Learned

We call this phenomenon the Iceberg Problem

because, like an iceberg, only a very small amount of information (the tip) is
visible while the more comprehensive information remains hidden from view.
KEY INSIGHT #2 27

In K–12 education, while the gaze of policy makers is focused on how students are
performing relative to grade-level assessments, learning gaps continue to accumulate
below the surface, making longer-term success harder to achieve. There is an educational
path for each student to unlock their full potential, but it requires seeing students more as
individuals than as a homogeneous group enrolled in a particular grade level.

Accountability systems are limited by the constraints of the underlying

assessments.

Federal law requires states to use statewide assessments as the basis for accountability
systems and to set goals for increasing the share of students who meet state standards in
reading and mathematics, accelerating progress of underperforming subgroups, and
improving graduation rates.36 States must also identify their lowest-performing schools
for varying levels of support and intervention.37 Many states use these systems to assign
ratings (such as A to F school grades) to all schools as a means of communicating
performance publicly rather than just for identifying the lowest performers, but a school
rating is not required under ESSA.

States have flexibility in the specifics of their identification system for low-performing
schools but must include certain types of indicators:

Three academic indicators:

• Academic performance in ELA and math based on state assessment results at

each grade level in grades three through eight and at least once in high school.

• Student growth or another valid and reliable indicator that allows for meaningful
differentiation in student performance.

• High school graduation rates based on the four-year adjusted cohort rate.

Two additional types of indicators:

• English language proficiency, which applies only to English learners.

• At least one additional indicator of school quality or student success, such as

chronic absenteeism or school climate (also referred to as “the fifth indicator”).38
KEY INSIGHT #2 28

States determine how each indicator factors

These incentives
in the determination of which schools require
intervention and for school ratings where further reinforce an
states elect to assign them, but the three
instructional focus on
academic indicators must each carry
“substantial weight,” and together carry grade-level material
“much greater weight” than the indicator(s)
regardless of students’
of school quality or student success.39
unique starting points,
The stakes can be high.
which make it even
Under ESSA, schools identified as low harder for students to
performing under these accountability
systems may be subject to intensive
ultimately achieve
interventions in certain states, such as state college and career
takeover or sanctions.
readiness.
Beyond ESSA school accountability systems,
states, districts,
states, districts, and
and schools
schools use
use assessment
assessment metrics as a key component in many other
metrics as aand
evaluation keydecision-making
component in many otherFor example, 39 states use a measure of
activities.
evaluation
student and decision-making
growth activities.
on summative tests as partForof their teacher evaluation systems.40 Many
example,
district 39school
and states administrators
use a measure believe
of studenttheir positions are dependent upon success on
growthsummative
annual on summative tests as part
assessments. And offor
their
many charter schools, meeting test-based goals
teacher
for evaluation
proficiency and systems.40 Manycan
student growth district
mean the difference between closure and
and schoolexistence.
continued administrators believe their
positions are dependent upon success on
In thesesummative
annual cases, the assessments.
required academic performance and growth measures are based on
And for
grade-level
many charterassessments. Thesetest-based
schools, meeting incentives further reinforce an instructional focus on
grade-level material regardless
goals for proficiency of growth
and student students’
canunique starting points, which make it even
harder fordifference
mean the students to ultimately
between achieve
closure andcollege and career readiness.
continued existence.

Growth metrics used for accountability do not provide a full picture of

In these cases, the required academic
student learning gains.
performance and growth measures are based
on grade-level
Under assessments.
ESSA, growth metrics These incentives
using state tests have taken on greater importance. State
further reinforce an instructional focus on
ESSA plans have included a variety of approaches to capturing growth. Some states use
grade-level material
“criterion-based” regardless
growth ofto
metrics students’
measure the degree to which students are closer to
unique starting points, which make
meeting grade-level expectations thanit even
they were the previous year. Others use
harder for students
“normative” to ultimately
approaches, such as achieve
Student Growth Percentiles, to measure how students’
college and career readiness.
test scores compare to those of students with similar past performance. Others use a
KEY INSIGHT #2 29

hybrid of the two. Readers can find links to several detailed analyses of these approaches
in this paper’s endnotes.41

The merits, strengths, and weaknesses of all these metrics have been widely debated at
both the state and national level. But these debates rarely acknowledge what all growth
measures described above cannot do as long as they are based on traditional grade-level
tests: They cannot precisely measure learning and knowledge gains that the tests do not
assess—namely, content above or below grade level. This means that “growth” measures
currently possible with state assessments only detect a narrow slice of potential learning.

Figure 11 shows how the current accountability system can give educators and parents the
misimpression that a student is growing. In this example, an individual student scored at
Level One on the sixth grade, as many of the grade-level skills he or she was taught over
the school year went unlearned.

Those skill gaps then carried over into seventh grade, where instruction was focused on
seventh-grade material (some of which built off of sixth-grade material, and some of
which built off of content from earlier grades).

At the end of the year, the student may have learned enough to score a Level Two on the
state test, but much of his or her unfinished learning has only compounded along the way.
Even though it might appear for accountability purposes that the student is growing, the
accumulated gaps will prevent the student from successfully learning more advanced
concepts in subsequent years.

6th-Grade School Year 7th-Grade School Year

4
Performance

Figure 11 3
Level

The Misimpression of
Learning Growth 2

1
6th-Grade 7th-Grade
Skills Learned Skills Learned

7th-Grade
Cumulative Skills

Skills Not Learned

Not Learned

6th-Grade
Skills Not Learned

6th-Grade
Skills Not Learned
KEY INSIGHT #2 30

Often, we find that school leaders and policy makers alike do not fully understand these
limitations, and speak about state growth metrics as if they are synonymous with
student learning progress. In fact, the “growth” that districts and states generally report
on could more accurately be described as “changes in relative performance,” because
the content of each grade’s assessment can be quite different.

Similarly, state and district leaders may be under the misimpression that scale scores from
summative assessments that linked to one another (known as vertical scaling) can be
used to precisely measure student learning growth. While such comparisons are strongly
supported when tests are comparable to one another, comparing grade-level summative
assessments—which have varied content from one year to the next—is far weaker and
inappropriate for use in higher-stakes contexts.42 Recent guidance published by the
National Education Policy Center was even more explicit about the misuse of vertically
aligned instruments to measure student growth in high-stakes contexts.43 According to
the study’s author, Madhabi Chatterji, Director of the Assessment and Evaluation Research
Initiative at Teachers College, “growth, measured using IRT-based scale score metrics
should be avoided altogether in accountability contexts, as these scales are too limited to
allow inferences about student learning from grade to grade.”44

States want to value growth in accountability systems, but their options

are limited.

Under ESSA, 47 states incorporated one or more of the growth measures described
above as accountability measures in elementary and middle school.45 There are several
often-cited policy rationales behind emphasizing growth measures in an accountability
system.46

• Provide a fairer opportunity for schools whose students enter with relatively low
performance to demonstrate success

• Encourage schools to support students across the achievement spectrum

• Reward schools moving student achievement in the right direction

• Identify schools where students exhibit flat or declining performance for extra
support or corrective action

According to a summary of state ESSA plans by the Data Quality Campaign, only two
states (California and Kansas) do not plan to include a growth measure in their ratings for
grade 3–8 schools.47 An analysis by the Fordham Institute found that 18 states weighted
growth metrics for more than half of their school ratings system, and 24 states count
growth at between 33 and 47 percent of a school’s overall rating.48
KEY INSIGHT #2 31

The uses of growth measures go beyond ESSA accountability. Any metric used in accountability
must also be reported to the public in state report cards,49 and several states include growth
indicators on student-level test reports that go to parents.50 Regardless of which approach to
measuring growth is used, the fact that the underlying assessments are nearly exclusively focused
on grade-level items dramatically limits the kinds of learning growth that can be detected and
rewarded in statewide accountability systems. These limitations may hold states back from
encouraging other valuable kinds of growth and progress.

Federal policy limits the ability of computer-adaptive tests to measure learning growth
across grades for accountability purposes.

Theoretically, the most comprehensive way intent of the assessment was to measure
to measure growth would be to test every “maximally accurate” results for each student
student on every skill, every year, which is by sampling content above and below grade
obviously not practically feasible or a good level.51 However, the design of the algorithm
use of students’ time. But with computer used to select items for the assessment limits
adaptive test (CAT) designs, which ESSA the test’s ability to measure gains on skills
explicitly allows, more sophisticated growth included in earlier and later grade-level
measures that cross multiple grades may be content standards because federal law
achievable. CAT tests hone in quickly on a requires states to measure the full breadth of
students’ skill levels by adjusting the items grade-level standards for every student.
students see according to their prior answers. Smarter Balanced therefore prioritizes grade-
If a student excels on the initial questions, level content by requiring that the first two-
they will receive more difficult questions in thirds of the assessment (approx. 22
order to test the upper bounds of their questions) assess grade-level material using
knowledge. If a student gets a question items written specifically to measure grade-
wrong, they might get an easier question level content. The pool of items may then
next. The selection of items for each student expand to include easier or harder items that
is governed by a test algorithm. As a result, are also aligned to grade-level content but
CATs can pinpoint students’ performance were initially written for a different grade
more quickly and precisely, especially for level. The purpose of the expanded pool of
students at the low or high extremes of the items is to add precision to the measure of
performance spectrum, and potentially grade-level content knowledge and skills, not
support more nuanced performance to provide information on above- or below-
measures. grade performance or overall learning
growth.52
The most widely used example of a
summative CAT test currently used at the Another CAT, the MAP assessment from
state level is Smarter Balanced, a test created NWEA, is an adaptive assessment that
by a multistate test consortium. An original measures student performance agnostic
KEY INSIGHT #2 32

to grade levels. Because it does not currently

As testing algorithms
measure proficiency against grade-level
expectations, states may not use the MAP to become more
meet ESSA accountability requirements.
prevalent, it is
NWEA has, however, developed studies that
link MAP performance with scores on a important that
variety of state summative tests.53 We
[computer adaptive
explore the implications of those studies in
further detail in Figure 7. assessments] are
Computer-adaptive assessments carry great developed with a keen
potential to shape the future of student eye on equity so they
assessment. But as testing algorithms
become more prevalent, it is important they do not reinforce
are developed with a keen eye on equity so preexisting biases.
that they do not reinforce preexisting biases.

There are barriers to innovative assessment pilots yielding more comprehensive

measures of growth.

Recognizing the possibility for innovative states and emerging technologies to enhance
assessments, ESSA includes an innovative assessment demonstration authority, commonly
called innovative assessment waivers. These waivers allow up to seven states to design and
pilot different types of summative tests that could roll out statewide after several years.

However, the requirements of this program limit states’ ability to address the challenges
around learning growth and grade levels described above in several ways:

• Innovative assessments must still align with grade-level standards.54 Test items can
extend beyond grade-level standards (just as the current assessments can), but
innovative assessments face the same practical limitations of test time and length.

• Participating states must still generate valid, reliable, and comparable results
including annual summative determinations for all students and subgroups of
students.55 Not only would the comparability provision suggest a need to focus on
grade-level content, but guidance issued by the Department of Education through
the Assessment Peer Review Process further reinforces the federal requirement to
report against grade-level proficiency.56

• ESSA requires that the innovation be brought statewide within three years and
ultimately meet the same peer review requirements as traditional assessments.57
This does not give states much time to design and iterate on innovative
approaches to assessing student learning.
KEY INSIGHT #2 33

• ESSA requires that the innovation be brought statewide within three years and
ultimately meet the same peer review requirements as traditional assessments.57
This does not give states much time to design and iterate on innovative
approaches to assessing student learning.

• ESSA only allows up to seven states to participate. Innovation for the remaining
states would have to wait until the next renewal of the Elementary and Secondary
Education Act.

Georgia, Louisiana, New Hampshire, and North Carolina have been approved to
participate in the innovative assessment pilot, though most are still focused on grade-level
assessments.58 Louisiana’s plan is focused on a combined test for English and social
studies, New Hampshire’s plan expands on the use of performance assessments,59 and
North Carolina’s plan calls for the use of multiple formative assessments, all of which
60
would be aligned
would beto aligned
grade-level standards.standards.60
to grade-level

Currently, the most promising approach to

mitigate the Iceberg Problem comes from
The instructional focus
Georgia’s plan. As part of its plan, these policies
participating districts will administer three
formative assessments over the school year incentivize may be at
with items from multiple grade levels. The
state will then aggregate those items that
odds with what may
relate to grade level standards to determine be truly best for each
a state summative score.61 If successfully
executed, Georgia’s approved plan holds the student given their
most promise for integrating comprehensive
unfinished learning
measures of learning growth while still
generating grade level performance. from prior years.

The instructional implications of grade-based assessments and accountability

The federal policies that undergird statewide assessment and accountability systems send
an unmistakable signal to middle grade math teachers: focus your instruction on the
grade-level standards.

Without clear and common expectations and grade-based reporting every year,
achievement gaps and systemic inequities might flourish in hiding. And the most
disadvantaged students are the most likely to suffer under these conditions—as happened
frequently in the era before standards-based accountability, when disbelief that
KEY INSIGHT #2 34

disadvantaged students could reach high levels of achievement was an accepted norm
among many educators and political leaders. Today’s teachers have more clarity into the
level of rigor required for students to be on a college and career trajectory so they can
better maintain high expectations.

But at the same time, we must acknowledge a real cost to a policy orientation focused on
grade-level expectations—that the instructional focus these policies incentivize may be at
odds with what may be truly best for each student given his or her unfinished learning
from prior years. What may have been intended by some policy makers as the equivalent
of an educational “dipstick” to gauge how students are performing may instead be driving
an instructional experience that can cause some students to fall further behind.

The unfortunate truth is that millions of students, including the vast majority of students
from historically disadvantaged communities, are coming to middle school with unfinished
learning from elementary school.62 This places an immense burden on the middle grade
math teacher to not only cover grade-level material, but to also diagnose and fill each
student’s unique pre-grade gaps that relate to grade-level content all within a single
school year—a tall order for even the most talented of teachers. This challenge becomes
even more daunting as students progress through middle school and learning gaps
continue to accumulate.

Students arriving into middle school multiple years behind grade-level standards need a
viable instructional bridge that enables them to catch up and move ahead. This requires a
strategic mix of pre- and on-grade skills, often for more than a single school year, given
the unfinished learning that has accumulated over time. Today’s assessment and
accountability policies oriented around annual grade-level proficiency make it far harder
to take this kind of instructional approach.
KEY INSIGHT #3 35

Key Insight #3
Balancing pre-grade level, on-grade level, and post-grade
level skills to each student’s needs can better support their
long-term success.

Evidence is emerging on the relationship between student performance and

the level of instructional content.

While ESSA requires an accountability system that focuses on grade-level mastery,

districts and schools are free to develop and apply their own parallel accountability
systems so long as they continue to meet state and federal requirements.

Most districts generally measure growth in the ways reflected in their state’s ESSA plan.
But some focus on different growth measures based on NWEA MAP, partly because it
includes items from across multiple grade spans. These schools still take the state test
and, for purposes of state accountability, their growth is measured in ways aligned with
federal policy. But under their own district- or school-based accountability system, they
are able to focus on comprehensive learning growth.

Over the last six years, our program has operated in districts and schools with different
philosophies around teaching students grade-level material. In schools where
KEY INSIGHT #3 36

accountability systems imposed at the district or school level are based on growth
measures that cross multiple grades, we can tailor a personalized curriculum for each
student that includes a mix of pre-grade, on-grade, and post-grade material, depending
on their unique starting points. For many students who enter sixth grade multiple years
behind, this means spending meaningful time addressing unfinished learning in service of
grade-level material. This can also mean not necessarily exposing students to all grade-
level standards in a single year, given the time it takes to fill multiple years of unfinished
learning.

By contrast, in partner schools judged primarily on annual state assessment growth

measures, school leaders have often asked us to weight students’ personalized curricula
more heavily toward grade-level content for the year. This can mean leaving important
pre-grade gaps unaddressed.

The impact of accountability policy on student performance in the context of Teach to

One: Math was explored by MarGrady Research in a 2019 study funded by the Bill &
Melinda Gates Foundation. The study included several correlational analyses that
compared the MAP growth of students participating in Teach to One: Math over three
years to national norms.

First, the study found that, overall, students served through Teach to One: Math over their
three years in middle school grew 20 percentile points (from the 15th percentile to the
35th percentile).63

Figure 12

Note: Includes only students who were 6th-graders in 2015–16, 7th graders in 2016–17, 8th-graders in 2017–18, enrolled in the same school
for all three years, and had both a fall 2015 and a spring 2018 MAP score. Students are not required to have a test score in every period to
be included.
KEY INSIGHT #3 37

Second, the study found that schools that operated within accountability systems that
valued learning growth (as reflected on the MAP) grew 38 percentile points over the
three-year period while those focused largely on state proficiency grew 7 points.64

Figure 13

Note: Figure shows percentile gain for consistently-enrolled students. The MAP Growth Aligned category includes
schools 4, 5 and 6. The State Growth & Performance category includes schools 1, 2, 3 and 7. The State Proficiency
Focused category includes schools 8–14.

And third, the study also included unique analysis that juxtaposed the learning growth
that students made on MAP to the academic level reflected in their daily lessons. To do
so, MarGrady included a concept called the “content gap,” which it defined as the
difference between the incoming performance level of students and the average level of
instructional content they received over the school year. For example, a sixth-grade
student who enters the school year on a fourth-grade level and receives content that
averages at a sixth-grade level would have a content gap of two for that year.
KEY INSIGHT #3 38

Figure 14
Relationship Between Content Gap and School-Level
Percentile Gains (3-Year)

In Figure 14, each bubble reflects a school (and the size of bubble corresponds to the size
of the school). The x-axis reflects the average three-year content gap for each school
(effectively recalculating the annual content gap each year and then averaging the results
over three years), and the y-axis reflects the three-year gains students made on MAP. The
study found suggestive evidence that schools with a smaller content gap—those where
the math content presented better matched students’ tested grade level from the
beginning of the year—tended to see stronger gains.65

As with any school-based program, there are a number of factors—including curriculum

implementation, teacher quality, and staffing, and factors external to the school—that can
influence outcomes. The MarGrady study was not structured to establish causality. We
expect future studies can control for many of these variables in order to more
conclusively determine the impact that underlying accountability systems may be having
on student acceleration over the longer term.

Additionally, we do not yet have sufficient data to determine the impact that these
different progression strategies have on eighth-grade state test performance. Our partner
schools to date have asked that we focus student progression strategies either on annual
measures of growth (as measured by MAP) or a blend of learning growth and annual
KEY INSIGHT #3 39

grade-level skills, and have sometimes shifted this focus in different school years and
within different student cohorts. This has prevented third-party studies from drawing
generalizable conclusions about the impact of Teach to One: Math on state tests.66

Comprehensively testing the impact of this approach on eighth-grade state tests would
require comparing the impact of students who receive a prioritized mix of pre- and on-
grade skills over three years—all in service of eighth-grade performance—to students who
receive grade-level content for each of three years, controlling for factors including those
listed above. Future studies (both for our work and elsewhere) may wish to focus here.

There is little evidence to suggest that for students who are far below grade
level, focusing instruction exclusively on grade-level content is effective.

While evidence of the value of balancing

pre- and on-grade skills is still emerging,
it should be viewed against the evidence
base for what policy currently
incentivizes—providing all students with
grade-level content.

Many studies have found that too many

students in the past were subjected to
repetitive, ineffective instruction that did
little to prepare them for college and a
Schoolswhere
career.67 Schools wherelow
lowexpectations and ineffective instruction reigned
expectations and ineffective
disproportionately were thoseinstruction
that served low-income students, students of color, and
reigned
other disproportionately
historically were students.
disadvantaged those Students spent years in remedial math courses
served
that got low-income
them students,
no closer to graduating. This unacceptable state of affairs has yielded
students
policy ofcurricular
and color, andremedies
other historically
that focus on grade-level instruction and assessment as key
disadvantaged
drivers students.
for ensuring Studentsrigor.
instructional spent
years in remedial math courses that got
At theno
them same time,
closer we find littleThis
to graduating. evidence in the research to support the notion that in
middle grade math,
unacceptable grade-level
state of affairs hascontent
yieldedfor students who have already fallen far behind
works
policy without carefulremedies
and curricular strategies to focus
that build up key missing skills. This is a difficult question to
study,
on in part because
grade-level curricula
instruction and vary between districts and schools, and curricula on paper
do not indicate
assessment howdrivers
as key teachers
for spend their time and energy under current policies. Even in
ensuring
schools with rigor.
instructional a grade-level curriculum focus, the best teachers may still be doing a lot to
spiral in prerequisite skills that students need.
At the same time, we find little evidence
One
in theway that researchers
research to support have explored the effects of giving students content far outside
the notion
that in middle grade math, grade-level
content for students who have already
fallen far behind works without careful
KEY INSIGHT #3 40

their current skills is by looking at students

taking algebra in eighth grade instead of high
school. A policy push in the early 2000s
placed many eighth-grade students in algebra
who would have otherwise taken a pre-algebra
eighth-grade math course. One study found
that students enrolled in advanced eighth-
grade algebra with low incoming math skills
performed about seven grade levels below
their peers on NAEP and struggled with
questions testing elementary-level
understanding.68 Another study found that low-achieving students pushed into algebra did less well
in subsequent math courses through high school, especially in geometry.69 This is possibly because
students never had a chance to build up prerequisite skills that would have helped them later on.

The above gives us reason to believe that in math, low-performing students pushed into grade-
level content without appropriate support and attention to prerequisite skills may not be better off
in the long run.

Some districts are taking the lead in comprehensively measuring learning growth.

As previously discussed, a number of districts and stakeholders, and identify schools in

and schools incorporate instruments that need of intervention. Growth data are also
measure learning growth across multiple grade used to inform instructional planning in
levels into their accountability systems, even schools and are used in some grades as a
though their states continue to measure growth factor in decisions about grade promotion
through changes in statewide summative and admissions to selective high schools.71 In
performance. First, in Chicago Public Schools addition, CPS uses accountability ratings to
(CPS), the district has created a local school offer struggling schools more standards-
accountability system where three out of four based professional development, coaching,
academic success metrics are growth-based and support, showing that a growth focus
and the primary assessment to measure growth and a commitment to rigorous standards can
is the MAP.70 This system, enacted in 2014, is complement one another.72
separate from the state’s ESSA tests and
accountability systems, but it fulfills CPS’s More recently, CPS announced that

responsibilities under Illinois state law to graduates attaining college- and career-

identify and support struggling schools. CPS’s ready credentials grew from 31 percent in

ratings are used to track school performance 2014 to 47 percent in 2018.73 Average math

over time, share data publicly with families

KEY INSIGHT #3 41

test scores, course grades, and passing rates If policymakers want

have improved over the past four years
among middle school students, especially in to create the space
schools receiving the most support from the
for schools to better
district to understand and implement
rigorous math standards.74 Performance in meet the unique
grades three through eight math improved
needs of their
from 45 percent of students meeting national
norms at their grade level in 2013 to 57 individual students,
percent of students in 2018.
they must create
Another interesting example of district-level new policy
accountability and instructional uses of data
is underway in Tulsa, Oklahoma.75 Tulsa
frameworks that
Public Schools developed data dashboards enable and
that allow educators to view, sort, and
analyze data, including multiple years of
encourage these
academic and non-academic measures. practices.
Dashboards like these could form the
stepping-stones to setting and understanding
multiyear growth goals. Like CPS, Tulsa uses MAP in addition to state tests. The district
MAP growth
uses in addition toprimary
as its state tests. The district
internal success metric and regularly reviews performance
uses growth
measures likeasproficiency
its primaryto
internal success
check that students are progressing to college and career
metric andTulsa
readiness. regularly reviews
considers its performance
own growth metrics as the most useful measure of student
measures
and like
system proficiency to
performance. Thecheck that
district supports school principals to set their own rigorous
students
and are progressing
attainable to college
goals for MAP growthand
at the school level. The district is also planning a
career readiness.
large-scale Tulsa considers
high school its own that aims to align assessments, accountability,
redesign process
growth
and metrics
internal as theaway
systems mostfrom
useful measure
grade- and of
time-based models for student achievement.76
student and system performance. The district
Districts
supportssuch as principals
school Chicago Public
to setSchools and Tulsa Public Schools are outliers when it
their own
comes
rigoroustoand
measuring learning
attainable goals growth
for MAPacross
growth multiple grades. In general, districts are more
likely
at theto accede
school to math
level. measures
The district and tests adopted by their states rather than develop
is also
parallel
planningsystems. If policy
a large-scale makers
high schoolwant to create the space for schools to better meet the
redesign
unique
processneeds of their
that aims individual
to align students, they must create new policy frameworks that
assessments,
enable and encourage
accountability, thesesystems
and internal practices.
away
from grade- and time-based models for
A focusachievement.
student on true learning
76 growth can also motivate teachers and students to
reach high expectations.
Districts such as Chicago Public Schools and
Tulsa Public
Many Schools
teachers are outliers
see students when it when they are taught or assessed on skills that
demotivated
comes to measuring learning growth across
multiple grades. In general, districts are more
likely to accede to math measures and tests
KEY INSIGHT #3 42

are far beyond their zone of proximal development. As one teacher said, “We want
students to feel confident and competent, and to stay excited. If the only test they have is
a grade-level test that they bomb every time, then what kind of motivation do they have
to keep trying?” On the other hand, when students have opportunities to see learning
growth and achieve small victories along the way, math seems more engaging and
accessible. “[Students’] feelings about math are tied to their success and achievement in
math. As soon as you start giving them opportunities to succeed, and then they see the
growth, their feelings change.” Teachers with whom we spoke were passionate about the
importance of students understanding where they stood relative to long-term goals and
having opportunities to meet high expectations. More nuanced growth measures can
motivate more students and teachers to successfully meet college- and career-ready
expectations.

Accountability models that run parallel to ESSA and emphasize more comprehensive
growth metrics that incorporate students’ starting points can allow students and teachers
to track continuous progress and see rewards for positive change. In a recent national
survey, 74 percent of teachers said that measures of student academic growth should be
a part of their school evaluations, and 64 percent said student growth measures should be
a part of teacher evaluations.77

Teachers told us that growth metrics help students see their progress and serve as a
motivator. When schools and teachers are judged on certain goals, students are likely to
set similar goals for themselves, and they may be more motivated by goals emphasizing
ambitious but attainable levels of growth over a seemingly arbitrary or faraway
proficiency bar.78 In schools and districts that balanced pre-grade and on-grade
instruction and emphasized learning growth, teachers felt that students were set up to
achieve success.
R E CO M M E N DAT I O N S 43

Recommendations
We believe deeply in the impetus behind the We must candidly
shift to standards-based accountability—a
commitment to equity, rigor, and acknowledge the
transparency conceptualized to ensure that trade-offs and costs a
all students have access to the full range of
opportunities that achieving college and policy orientation
career readiness affords. Our education focused on grade-
system gained significantly from the
development of these systems. We now have level expectations
strong rigor and alignment in academic creates.
standards in most states. We now have
access to significantly more information about schools’ performance, and we are shining a
abouton
light schools’
the gravity
performance,
and persistence
and weof
are
gaps in learning among traditionally underserved
shining a light
students. These onare
thesubstantial
gravity andaccomplishments.
persistence
of gaps in learning among traditionally
But at the same
underserved time, we
students. must
These candidly
are acknowledge the trade-offs and costs a policy
substantial
orientation focused on grade-level expectations creates. For far too many students, these
accomplishments.
costs are substantial given their unfinished learning from prior years. This is especially
But at the same time, we must candidly
acknowledge the trade-offs and costs a
policy orientation focused on grade-level
expectations creates. For far too many
R E CO M M E N DAT I O N S 44

problematic for middle grade mathematics given the cumulative nature of math itself, the
fact that students are entering the middle grade years with unfinished learning from
elementary school, and that high schools are generally ill-equipped to catch students up.

Rigor, equity, clarity, and transparency remain foundational principles for high-quality
accountability systems. These systems also should encourage instructional practices that
meet students’ needs and effectively advance them to college and career readiness. The
current, standards-centric accountability system—under which middle grade math
instruction falls—is oriented around measuring performance annually. While this can
create the superficial appearance of equity, this approach creates instructional incentives
that prioritize covering grade-level material over meeting students in their zone of
proximal development. For students with unfinished learning from prior years, this can
actually keep them from ultimately achieving college and career readiness.

In the current political climate, there appears to be little appetite at the federal level to
revisit the core tenets of ESSA anytime soon. As a result, states and districts that want to
explore new pathways for assessment and accountability may find their aspirations are
limited by a) current federal policy that requires an accountability system based on the
statewide measurement of performance against grade-level expectations and b) practical
limitations in the frequency and length of the assessments themselves.

Nonetheless, policy makers can still take action in response to the challenges we describe.

Recommendation #1: Measure Comprehensive Learning Growth

Statewide assessments are almost exclusively focused on grade-level standards and thus
any learning gains made on pre- or post-grade skills are unlikely to be reflected on these
instruments. While future assessments may be able to effectively measure both learning
growth and grade-level proficiency, state and districts looking to better capture learning
growth now may wish to complement current statewide assessments with the adaptive
assessments that incorporate standards from multiple grade levels. The MAP assessment
(NWEA), i-Ready (Curriculum Associates), and STAR Math (Renaissance Learning) are
three products schools use to achieve these goals. Some of these instruments have
studies that help users to better understand each student’s current performance relative
to national norms and to predict performance on statewide assessments.

Recommendation #2: Modify Accountability Systems

ESSA places many restrictions (both explicit and practical) on the ability of state
accountability systems to incorporate measures of comprehensive learning growth and
R E CO M M E N DAT I O N S 45

incentivize more strategic approaches to catching students up. Nonetheless, several

strategies that states may want to consider when addressing the challenges outlined
above include:

• Incorporate multiyear growth metrics in school ratings systems. While any

measure in a state accountability system must be calculated annually, that need
not preclude multiyear measures. States may choose to compare changes in
proficiency rates over multiple school years as part of their growth calculations
within their accountability systems. For example, schools could be rewarded for
the degree to which student proficiency levels changed between the start of sixth
grade and the end of eighth grade, thereby encouraging schools to adopt a long-
term view of performance to strategically catch students up and move them
ahead.

• Weight key transition points more heavily. Some states, such as Arizona, place
extra weight on critical benchmarks, such as third-grade reading. A state could, for
example, weigh the performance of fifth- and/or eighth-grade students on their
respective summative assessments more heavily than other middle grade levels.
This approach would emphasize the importance of high school math readiness and
also allow schools to strategically take a longer-term view.

• Explore the use of adaptive assessments that span multiple grades into measures
of school quality and student success. ESSA restricts academic measures to
grade-level tests, but states could explore options under ESSA’s “fifth indicator” to
use adaptive assessments in order to capture learning growth beyond a single
grade level. The fifth indicator is broadly defined by ESSA as a measure of school
quality and student success and is a required piece of school accountability
determinations. Many states have chosen chronic absenteeism or other non-
academic indicators for this category, but nontraditional academic measures are
an additional option. For example, in states such as Nebraska, where all students
take adaptive tests that span multiple grades, changes in school growth
percentiles on the MAP could be a component of a fifth indicator score.

• Create supplemental growth indicators. Several states dissatisfied with the

structures of NLCB and ESSA have created their own parallel or supplementary
systems governed only by state law. These systems cannot overrule what is
required under federal law, but they can provide additional information to help
inform administrators, teachers, and parents about other aspects of school
performance. If these systems were targeted at innovative or alternative schools
R E CO M M E N DAT I O N S 46

• not served well under ESSA, they could play a valuable role in defining and
encouraging a different vision of success. Some states have already begun to
incorporate supplemental indicators in other domains (e.g., social-emotional
learning and English Learners).

States may be able to implement many of these recommendations without federal

approval. Other recommendations may need to be incorporated into future statewide
ESSA plans (which can be revised at any time) and approved at the federal level before
they can be fully implemented.79

Districts have more flexibility to serve as laboratories for accountability innovation, with
the backstop of state accountability systems providing an additional layer of assurance
and transparency. They can take many of the steps described above for states even
further, given their governance responsibilities. The cases of Chicago and Tulsa show what
is possible when districts offer their own visions for accountability and align their support
systems toward those goals. Districts such as Lindsay Unified in California have taken this
even further, building competency-based learning progressions that include standards
from across multiple grade levels into each student’s instructional experience.80

Recommendation #3: Launch Math Innovation Zones

The State of Texas launched a statewide effort to incubate high-quality blended learning
programs aligned with state standards to “dramatically impact the life trajectory of
students with a focus on eighth-grade algebra.”81 Participating districts must use an
independent and state-approved growth metric three times a year (fall, winter, spring) as
one component of Math Innovation Zones is to measure the effectiveness of the
personalized learning program. These results on learning growth are monitored by the
state on a quarterly basis and districts who have been selected for this program are
eligible to receive a designation on the state accountability system.82 Other states could
introduce similar, math-specific, innovation initiatives.

Recommendation #4: Make Available High-Quality Instructional Supports

and Strategies That Account for Unfinished Learning

Policy changes around accountability systems are necessary but not sufficient to improve
outcomes for students. Change happens at the classroom level, when educators have the
supports, incentives, and tools they need to guide and accelerate student learning. This
paper’s purpose is not to prescribe one curricular approach or learning model. There are
support systems and resources that could improve teaching and learning on a broader
scale that states and districts could offer right away:
R E CO M M E N DAT I O N S 47

• Implement strategies to build skills in support of grade-level standards. One way

to help reconcile the tension between grade-level expectations and students’
starting points is by ensuring that students are always either accessing grade-level
content or advancing toward it by focusing on the unfinished learning required to
meet grade-level standards. This would mean that pre-grade learning is not
aimless or based on low expectations—it should enable acceleration toward
meeting a particular grade-level skill. This strategy can help students access and
master grade-level content even if they begin with significant gaps.

• Find more time. The rigor and depth reflected in today’s college- and career-ready
standards generally require a full school year of instruction. Any time spent helping
students address unfinished learning is less time available to focus on grade-level
material. If districts and schools expect all students to successfully learn grade-
level material, they must find and make effective use of additional instructional
time (e.g., double blocks, after-school programs, summer school) and ensure
teachers understand and implement strategies that promote efficient learning to
successfully accomplish both objectives.

• Promote job-embedded professional learning. Teachers will benefit from training

in the content standards from multiple grade levels. Such trainings can focus on
the subject matter itself, the relevant learning progressions, and common
misconceptions. Simply offering this form of professional development will signal
to teachers a deeper awareness of the inherent challenges they face.

• Provide high-quality instructional materials anchored in grade level but inclusive

of pre-grade predecessors. High-quality instructional materials are critical to
effective instruction, and, first and foremost, schools and districts should invest in
quality and consistency across grades. However, even districts that have
purchased high-quality, grade-aligned material can end up leaving teachers on
their own to search for high-quality, off-grade content. A more systematized way
to source and leverage instructional material that includes content from multiple
grade levels (either through models such as Teach to One: Math or through other
innovative approaches) will both drive quality and save teachers time.

• Ensure time for collaboration. Teachers will benefit from planning time with those
who teach students enrolled in other grade levels, where they can explore in more
detail strategies and approaches for teaching content that may not be explicitly
included in grade-level standards but are necessary predecessor skills for student
mastery of grade-level content.
R E CO M M E N DAT I O N S 48

Recommendation #5: Advance a Future Vision for Assessment and

Accountability

Given the varied incoming mathematical skill levels that students begin the year with and
the limited instructional time teachers have with students, one can imagine a future
assessment and accountability system that is both aligned with the foundational principles
of rigor, equity, clarity, and transparency and also with what is best instructionally for
each student. Such a system could include

• Adaptive Assessments. Assessments that span multiple grades and can adapt to
student responses in order to give credit for learning gains that are outside of
students’ enrolled grade level.

• Competency-based Assessments. School success could be determined based on

the ability for students to reliably demonstrate mastery of key academic skills and
concepts throughout the school year, as opposed to a single statewide summative
assessment.

• Curriculum-embedded Assessments. Innovative learning models and next-

generation curriculum offerings can ultimately embed reliable assessments that,
with appropriate controls and comparability capabilities, could one day replace
annual summative assessment with far greater levels of precision.

No doubt other possible hallmarks of what a future assessment and accountability system
might look like will emerge, and healthy debates will be had along the way about what is
best. It will be incumbent upon policy makers, innovators, advocates, academicians,
researchers, parents, and others to begin to design what these solutions might look like
and to ensure there is space within the current policy landscape for them to evolve.
CONCLUSION 49

Conclusion
Our aspiration in publishing this paper is to Policy makers cannot
catalyze robust conversations and debates
about how accountability systems can simply ignore the
support strong instructional practices and
fact that math
advance equity for the most disadvantaged
student groups and where our current learning is
systems fall short of their goals. This is not
cumulative.
just one conversation or a problem with a
single answer—these issues require discussion at the local level, in schools and districts,
discussion
across states,
at the
andlocal
nationally
level, in
among
schools
educators,
and policy makers, advocates, and instructional
districts,organizations.
support across states, There
and nationally
are important
amongissues and legitimate competing points of
educators,
view on thepolicy
best paths
makers,
forward
advocates,
for instruction,
and accountability, and equity. Ignoring these
instructional
issues out of support
fear of disrupting
organizations.
the status
Therequo
are is unacceptable.
important issues and legitimate competing
Policy
points makers
of view cannot simply
on the best ignore
paths the fact
forward for that math learning is cumulative—the skills
and knowledge
instruction, gained in one
accountability, andyear provide an essential foundation for accessing and
equity.
mastering skillsissues
Ignoring these in multiple
out ofgrade levels
fear of down the line. When students do not fully master
disrupting
foundational skills,
the status quo this unfinished learning accumulates over time, making it increasingly
is unacceptable.
challenging for the student to catch up. The instructional incentives and pressure to
deliver exclusively grade-level content created by the predominant assessment and
CONCLUSION 50

accountability structures is fundamentally at transparency these systems were designed

odds with the needs we see in the high to protect.
proportions of students who enter middle
school multiple grade levels behind. Those who disagree with our position may
argue that effective teaching practices can
This is not about having low expectations of enable students to learn both grade-level
students; it is a fact inherent in the skills and the applicable pre-grade learning
sequential nature of mathematics itself and gaps. While this may be true in other subjects
the need to honestly confront the unfinished or in lower grade spans, we see little
learning gaps that have accumulated evidence in middle grade math to suggest
throughout the elementary school years in that such an expectation should serve as the
service of students’ longer-term success. basis for national policy. A teacher delivering
a lesson on Quadratic Equations cannot
In our work with partner schools, we have simply provide a side lesson on Exponents
found that with the right set of supports and for those students who happen to have that
structures, students who enter middle school learning gap. Each topic requires dedicated
below grade-level standards can catch up time and attention in order for them to be
and move ahead. For some, the path to properly covered. Simply expecting all
getting there requires addressing unfinished teachers to achieve what few (if any)
learning that may take more than a single teachers can reasonably accomplish is not a
year. But, in the end, readiness for high recipe for systematic success.
school is achievable.
We similarly reject the notion that teaching
However, today’s assessment and students pre-grade content is somehow
accountability policies fixated solely on deemed not rigorous. From an instructional
grade-level content make that journey more perspective, teaching pre-grade skills does
difficult. Policies push teachers to focus on not equate to teaching procedural
grade-level material to the exclusion of approaches only. In fact, access to grade-
individual growth, which may be causing level content often requires students to apply
some of the most disadvantaged students to deep, conceptual understanding of
fall even further behind. The policies may also foundational skills that require rigorous
be preventing advanced learners from instructional practices to develop.
progressing to skills beyond their assigned
grade level, even when they have the ability From a content perspective, equating grade-
to do so.83 Even the greater emphasis on level expectations with academic rigor also
growth metrics emerging under ESSA cannot ignores the fact that students have different
fully address the challenge because those starting points. One eighth grader may not
metrics are still based on grade-level be able to successfully access a lesson in
assessments. The resulting blind spot in Systems of Equations because they lack the
accountability threatens the equity and predecessor knowledge, while another may
transparency these systems were designed
to protect.

Those who may disagree with our position

CONCLUSION 51

find that skill not rigorous enough based on what they already know. Equating grade-level
instruction with “rigorous instruction” mistakenly assumes all students within a grade level
have the same starting point.

We are calling on the field—practitioners, policy makers, philanthropists, and other

stakeholders—to come together to honestly and productively grapple with the tensions
in our current system, to understand where policies may create barriers to our most
ambitious goals for students, and to create a better educational system for all students.

We hope to bring our perspective together with leaders and practitioners of all levels,
both those who may agree with us and those who may not, for a productive dialogue
focused on our shared goal of enabling all students to achieve college and career
readiness and success. We also hope our perspective helps to spur the development of
more innovative learning models that seek to effectively balance this tension.

We are also encouraging states and districts to create the space for developing what one
day might be a new accountability system that preserves high expectations, transparency,
and equity while also incorporating more robust measures of learning growth that
incentivize more effective instructional practices. Our recommendations include a set of
near-term recommendations for like-minded states and districts to consider within the
current framework of ESSA in order to advance these efforts.

We fully recognize that policy is not the sole source of the problem and that policy
changes do not provide a complete solution. There is work to be done to ensure that
schools and teachers have the instructional resources, skills, and knowledge to provide
the nuanced and effective math instruction students need to rapidly advance and excel.
We believe this is an area ripe for both additional research and instructional innovation.
APPENDIX I:METHODS 52

Appendix I: Methods
In this paper, New Classrooms sought to understand whether and how standards,
assessments, and accountability systems ensure that students develop the necessary skills
to demonstrate learning in math, enable teachers and schools to provide high-quality
math instruction that properly serves students at all skill levels, and enable emerging
efforts in personalized and mastery-based learning.

To answer these questions, New Classrooms drew on internal expertise and external
sources and used a variety of quantitative and qualitative research methods:

Literature review:

• History, evolution, and impact of standards-based accountability policies

• Common features of state policies on math standards, assessment, and

accountability systems

• Landscape of standards and assessments in states

• District- and state-level accountability systems

• Secondary mathematics curriculum, instructional methods, and student learning

progressions

• Assessment methods, design, and metrics, including measures of performance and

growth

Data analysis:

• Reviewed publicly available NAEP, state assessment, and longitudinal data on

student math outcomes and progress toward college and career readiness,
particularly at key transition points like eighth grade

• Analysis of New Classroom’s NWEA MAP and state assessment data from Teach to
One classrooms across multiple years and sites

Expert interviews:

• Internal interviews with New Classrooms leaders on their experiences with partner
schools and districts, and emerging findings from Teach to One sites
APPENDIX I:METHODS 53

• More than 20 external interviews with experts on standards, assessment,

curriculum, and math instruction to understand various perspectives and
experiences on the impact of standards-based reform and the role it should play in
math instruction, assessment, and accountability

! These included policy experts and leaders at the federal level and in various
states, leading researchers in assessments and accountability, curriculum
developers, competency-based learning experts and advocates, including
many people whose views and approaches on math instruction and policy
differ from those of New Classrooms

Teacher focus groups:

• Convened six focus groups in multiple cities consisting of middle school math
teachers to gather perspectives on their own experiences with math instruction
and how the current system of standards and assessments influences their day-to-
day work

• Supplemented with interviews of principals and/or instructional leaders at those

school sites

• More detail on teacher focus groups below

Data analysis for Figures 12, 13, and 14 were based on four years of NWEA MAP results
among Teach to One: Math students, from 2014–2015 through 2017–2018. Only students
with a fall and spring score in each year were included. New Classrooms used the NWEA
MAP to PARCC linking study to create performance categories.84 Among the summative
test linking studies available from NWEA, we selected PARCC based on its evidence of
college and career rigor as an appropriate benchmark.85 Not all students in this data set
live in states where PARCC is administered. The process to calculate performance relative
to assigned grade level is as follows:

• The PARCC linking study establishes a MAP RIT score that predicts a 30 percent
chance of proficiency on each grade’s end-of-year exam. Those scores were used
as the cut point for “effective grade-level bands” in the MAP vertical scale (e.g., the
grade-level band for sixth-grade math based on this study was 224 to 233).

• Each student’s Fall RIT score was assigned to one of these effective grade-level
bands.

• We calculated the difference between each student’s effective grade performance

and their assigned grade. For instance, a student assigned to seventh grade with
APPENDIX I:METHODS 54

• performance in the fifth-grade band would have a difference of -2. A student

assigned to fifth grade with performance in the sixth-grade band would have a
difference of +1.

• These differences determined the achievement bands relative to grade level:

• four or more years below grade-level standards;

• two and three years below grade-level standards;

• one year below grade-level standards, at grade-level standards, and one

year above grade-level standards; and

• two or more years above grade-level standards

We compared this relative grade level to a 80 percent probability of achieving PARCC

proficiency in Figure 7. This variable is calculated using a Spring MAP to PARCC Linking
Study. There is a minimum Spring RIT score identified for each grade level for a 80
percent chance of proficiency on that grade’s PARCC. We subtract the student’s Fall RIT
score from this RIT score to get the growth needed over the course of the school year to
reach that RIT score. We divide that amount of growth by the fall-to-spring national
average growth for that grade, which results in the “growth needed” variable.
APPENDIX II:FOCUS GROUPS 55

Appendix II: Focus Groups

New Classrooms contracted with Bellwether Education Partners to conduct five focus
groups with middle school math teachers from five schools in different parts of the
country: three located in New York City, one in the Atlanta area, and one in Chicago.
Focus groups aimed to incorporate educator voices and insights into the paper to
support, challenge, or supplement other research findings. Schools were recruited from
New Classrooms’ and Bellwether’s network and partners. School leaders agreed to
participate in the project and then extended invitations to their teachers. While these
sessions were conducted rigorously, they do not provide a representative sample of
educators.

Of the participating schools, two were traditional public schools and three were public
charter schools. Two participated in New Classrooms’ Teach to One: Math model; three
used other models of math instruction. All schools served students who were
predominantly from low-income households, but other student demographic trends
varied. Among the Teach to One focus groups, most teachers also had experience
working in more traditional math teaching environments.

Focus groups were conducted in groups of five to ten teachers, without administrators
present. Focus group facilitators used a common, preplanned protocol for each focus
group, but the content of each conversation varied somewhat on the school context and
participants’ areas of interest, and facilitators asked follow-up questions as necessary. The
focus groups did not ask for teachers to reach consensus. The goal of the focus groups
was to understand teachers’ experiences with various models of math instruction and
explore the impact of assessment and accountability systems in their work.

The protocol for the focus groups included three areas of inquiry: instructional
approaches, assessment and accountability impact, and suggestions for improvement.
Example questions below do not fully represent the scope of the focus group protocol.

Instructional approaches

• How do teachers measure students’ knowledge and skills at the beginning of the
school year and track their progress?

• How do teachers address students’ variation in knowledge and skills in their

instructional approach?

• How do teachers balance instruction centered on grade-level standards with the

APPENDIX II:FOCUS GROUPS 56

• need to differentiate instruction based on students’ individual needs?

Assessment and accountability impact

• How do teachers use student assessments, and what do they use them for?

• What impact, if any, do accountability structures linked to student assessment

results have on teachers’ instructional approaches?

• How do assessments affect how teachers think about their own effectiveness and
their students’ attitude toward math instruction?

Suggestions for improvement

• How do teachers think their schools could improve the way students’ math
knowledge and progress are measured?

• What policy changes do teachers think would best support student learning across
various skill levels?

All focus group participants received a consent form notifying them that participation was
voluntary and informing them of the purpose and content of the discussion. They were
notified that any remarks would not be reported back to their schools or linked with their
names or schools in this publication. Teachers received a small gift in exchange for
participation. New Classrooms ensured that findings resulting from the focus groups
removed any potential personally identifying information from quotes included herein.
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 57

Appendix III: Evolution of National Policy

Figure 15 — Policy Timeline

Timeline of Standards-Based Reform Milestones

April 1983 The National Commission on Excellence in Education publishes “A Nation at Risk: The
Imperative for Educational Reform,” a report warning of a “rising tide of mediocrity” in
America’s educational system.

October 1994 President Bill Clinton signs the Improving America’s Schools Act (IASA) into law, requiring
every state to adopt academic standards for students in reading and math and regularly
administer aligned assessments in certain grade spans.

January 2002 President George W. Bush signs the No Child Left Behind Act (NCLB) into law, requiring states
to test students in reading and math annually in grades 3–8 and once in high school, and tying
those test results to specific consequences for schools and districts.

December 2008 CCSSO, NGA, and Achieve publish “Benchmarking for Success: Ensuring US Students Receive a
World-Class Education,” a report encouraging states to adopt common, internationally
benchmarked standards in ELA and math to ensure that students have the knowledge and skills
to be globally competitive.

May 2009 CCSSO and NGA begin developing the Common Core State Standards.

July 2009 Obama administration announces that states can compete for $4.35 billion in Race to the Top
grants to pursue policies like adopting college- and career-ready standards, recruiting and
retaining effective teachers and principals, and implementing school turnaround efforts.

June 2010 CCSSO and NGA release the final version of the Common Core State Standards.

September 2010 Obama administration awards roughly $330 million in Race to the Top funding to the
Partnership for Assessment of Readiness for College and Careers (PARCC) and the Smarter
Balanced Assessment Consortium (SBAC) to develop assessments aligned to the Common
Core State Standards.

September 2011 Obama administration announces that it will grant states waivers from the most burdensome
provisions of NCLB in exchange for pursuing certain policies on standards, accountability, and
evaluation systems for teachers and school leaders.

continued
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 58

December 2015 President Barack Obama signs the Every Student Succeeds Act (ESSA) into law, granting
states greater flexibility in how they assess students, design accountability systems, and
identify and support schools in need of improvement.

May 2017 16 states and the District of Columbia submit ESSA plans.

September 2017 Remaining 34 states submit ESSA plans.

For decades, standards-based reform and accountability have helped advance equity,
transparency, and rigor in America’s educational system. In 1983, the National Commission
on Excellence in Education, formed by then–Secretary of Education Terrell Bell, published
its seminal report, “A Nation at Risk: The Imperative for Educational Reform.” The report
warned that “the educational foundations of our society are presently being eroded by a
rising tide of mediocrity that threatens our very future as a nation and a people.”86 This
report marked the beginning of a bipartisan movement to improve students’ academic
outcomes via state, federal, and local education policies. Throughout its history, proponents
of standards-based reform have emphasized accountability for adults at various levels of
education, high academic standards for what skills students should attain by graduation,
and particular attention to the needs of low-income students, students of color, and
students in persistently low-performing schools.

The first major federal law in this new era of educational reform was the 1994 Improving
America’s Schools Act (IASA),87 which reauthorized the Elementary and Secondary
Education Act (ESEA)—first enacted in 1965.88 Among other things, this legislation required
that every state adopt academic standards for students in reading and mathematics, and
regularly administer assessments aligned to those standards “at some time” in grades three
through five, grades six through nine, and grades ten through twelve.89 However, federal
enforcement was weak, and by January 2001, only eleven states were in compliance with
IASA’s assessment provisions.90

The federal role in accountability was expanded further in 2001, when Congress passed the
No Child Left Behind Act (NCLB) with strong bipartisan support under President George W.
Bush. This legislation put in place the building blocks for modern accountability systems by
requiring that states adopt “challenging academic content standards,” test students with
standards-aligned assessments annually in grades three through eight and once in high
school, and link low proficiency rates on those assessments to specific consequences for
schools and districts—including state intervention.91
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 59

NCLB also served an important role in shifting the focus of education policy toward
equity. Accountability consequences and interventions were based on not just overall
student performance but also the performance of particular subgroups, including
economically disadvantaged students, students from major racial and ethnic groups,
students with disabilities, and students with limited English proficiency. The law also
required states to report assessment results in every tested grade and subject, and
disaggregate results by these subgroups.92 For the first time, schools and districts faced
meaningful consequences for failing to properly serve all student groups, and families,
advocates, and educators had access to consistent data showing the extent and
persistence of achievement gaps.

While NCLB helped advance reform in some ways, over time it became clear that its goals
were increasingly unrealistic and potentially counterproductive for schools. For example,
NCLB required that all students reach the proficient level on state assessments by 2014
(which no state succeeded in meeting).93 The law’s limited requirements on standards also
meant that states established vastly different expectations for their students, leading to a
patchwork of high- and low-quality standards across the country.94

Due to the inconsistent quality and content of academic standards under NCLB, states
began to mobilize themselves in response to the challenge of low-quality and inconsistent
academic standards. In 2008, the National Governors Association (NGA), the Council of
Chief State School Officers (CCSSO), and Achieve released “Benchmarking for Success:
Ensuring US Students Receive a World-Class Education,” a report that recommended
states “upgrade state standards by adopting a common core of internationally
benchmarked standards in math and language arts for grades K–12 to ensure that
students are equipped with the necessary knowledge and skills to be globally
competitive.”95 Low-quality standards resulted in countless high school graduates who
were not qualified or prepared to advance to postsecondary opportunities. They
especially hurt students in low-performing schools, who were more likely to lack access to
course content that could support them in college readiness. To address this challenge,
state leaders developed a common set of standards that would eventually become known
as the Common Core State Standards (CCSS).

The following year, development of CCSS began, incorporating input from state leaders,
educators, nonprofits, and content experts, as well as feedback from the general public.
Multiple organizations, including ACT and the College Board, released their own versions
of college- and career-readiness standards. These influenced the development of CCSS, as
did content area standards from groups like the National Assessment Governing Board
and the National Council of Teachers of Mathematics, as well as input from states
considered to have high-quality standards, including Massachusetts and California.
Information from international bodies, such as the Trends International Mathematics and
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 60

Science Study (TIMSS) and the Programme for International Student Assessment (PISA),
also helped ensure that CCSS set a sufficiently high bar. Additionally, a validation
committee, whose members were appointed by a group of governors and chief state
school officers, was created to review CCSS’s evidence base and development process.96

In 2009, while CCSS were being developed, the Obama administration announced that
states could compete for $4.35 billion in Race to the Top grants to pursue policies like
adopting college- and career-ready standards, recruiting and retaining effective teachers
and principals, and implementing school turnaround efforts.97 When NGA and CCSSO
released the final draft of CCSS in 2010,98 the Race to the Top program played an
important role in encouraging adoption of the standards. At their peak, CCSS had been
adopted by 46 states.99

However, political backlash from both the right and left caused many states to rethink
their adoption of CCSS, with some either revising the standards or replacing them with
state-developed standards. Conservative opponents of CCSS, primarily associated with
the rise of the Tea Party, objected to the federal government’s involvement in promoting
the standards and shared tests through Race to the Top. Teachers’ unions have also
expressed opposition, though this is less focused on the standards themselves than on
implementation efforts. They cited a lack of adequate time, training, financial and
instructional resources, and other support from states and districts to properly implement
CCSS in classrooms, and on the use of test results for teacher evaluations.100

Both the impact of CCSS and the resulting backlash can be seen in ESSA. Similar to NCLB,
ESSA requires states to adopt “challenging academic content standards” and “aligned
academic achievement standards” that include at least three level of achievement.101 In
addition, the law requires states to demonstrate that these standards are “aligned with
entrance requirements for credit-bearing coursework in the system of public higher
education in the state and relevant state career and technical education standards.”102
However, in response to the policies of the Obama administration, ESSA also includes a
provision prohibiting the US Secretary of Education from mandating, directing,
controlling, coercing, or exercising any direction or supervision over states’ standards.103

Today, 33 states have maintained their adoption of CCSS (though Minnesota only
adopted the ELA standards), and 13 states have adopted standards generally similar to
Common Core, while only four states (Alaska, Nebraska, Texas, and Virginia) never
adopted Common Core.104

According to a recent report from the Thomas B. Fordham Institute,105 most states not
using CCSS have opted for weaker math standards. For example, some states lack fully
coherent middle school progressions that make the appropriate connections between
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 61

interrelated standards and topics. CCSS has still encouraged some positive trends among
those states. The report also identified four “positive trends” in states’ math standards,
which the authors attribute partly to the continued influence of CCSS. The trends are:

1. A stronger focus on arithmetic in grades K–5;

2. More coherent treatment of proportionality and linearity in middle school,

including rates and ratios, slope, and linear relationships and functions;

3. An appropriate balance between conceptual understanding, procedural fluency,

and application; and

4. Better organization and teacher supports, including focused introductions for

individual grade levels and courses, mathematically coherent organizational
approaches that highlight the connections between standards, and helpful ancillary
materials.

Along with the shift to common standards, states also began moving toward common
assessments. In 2010, again under the Race to the Top program, the Obama
administration awarded grants to two consortia of states—the Partnership for Assessment
of Readiness for College and Careers (PARCC) and the Smarter Balanced Assessment
Consortium (SBAC)—to develop assessments aligned to CCSS. By 2011, 45 states had
joined one or both consortia.106

However, the same political backlash that plagued CCSS also caused many states to leave
the testing consortia, instead opting to partner with different assessment vendors or
create their own assessments. This pushback focused acutely on the perception of “high
stakes” tests out of the individual states’ control tied to consequences for students,
teachers, and schools. As of spring 2018, only 18 states were still administering
assessments affiliated with these consortia in some way, and several have announced
plans to create their own tests in the next few years.

As states were developing new standards and assessments—aided by the federal Race to
the Top program—states began advocating strongly for relief from some of the key
requirements of NCLB, particularly consequences for schools not meeting Adequate
Yearly Progress (AYP) goals.107 Congress failed to enact new legislation, and in 2011 the
Obama administration began issuing waivers from the law. These waivers granted states
flexibility from some of NCLB’s requirements—including AYP and the 2014 proficiency
goal—in exchange for implementing certain policies, such as the adoption of college- and
career-ready standards and teacher evaluation systems based in part on student
achievement.108 Forty-three states, the District of Columbia, and Puerto Rico had their
waiver requests approved.109
APPENDIX III:EVOLUTION OF N AT I O N A L POLICY 62

In 2015, Congress finally reauthorized ESEA, and also passed the Every Student Succeeds
Act (ESSA) with broad bipartisan support.110 The law maintains key provisions of NCLB
while granting states significantly more flexibility and authority around testing,
assessments, and accountability.

• Testing: ESSA preserves many of NCLB’s testing provisions, including testing

requirements for reading, math, and science, as well as the disaggregated
reporting of test results among student subgroups. However, ESSA also clarifies
that states may use computer adaptive tests (CAT) and allows tests to “measure
academic proficiency and growth using items above or below the student’s grade
level,”111 in addition to academic performance at grade level.

• Assessments: Under ESSA, states are now allowed to administer multiple interim
tests that add up to a final score rather than using one summative test,112 and
states may apply for innovative assessment waivers for additional flexibility.113

• Accountability: Like NCLB, ESSA requires that state assessment results serve as
components in school accountability systems. These systems must set goals for
increasing the share of students who meet state standards in reading and
mathematics, accelerating progress of underperforming subgroups, and improving
graduation rates.114 ESSA requires states to identify their lowest-performing
schools for varying levels of support and intervention115 but provides them
flexibility in the specifics of their identification system. While the law requires that
states use certain types of indicators, it largely leaves it up to states to decide how
each indicator is weighted.116

Sixteen states and the District of Columbia submitted their plans for accountability under
ESSA to the US Department of Education (ED) for review in May 2017, while the remaining
34 states submitted their plans in September of the same year. The new state
accountability plans took effect starting in the 2017–2018 school year.117
ENDNOTES 63

1 Stella Fayer, Alan Lacey, and Audrey Watson, “STEM Occupations: Past, Present, and Future,” US Bureau of Labor Statistics,
2017, bls.gov/spotlight/2017/science-technology-engineering-and-mathematics-stem-occupations-past-present-and-future/pdf/
science-technology-engineering-and-mathematics-stem-occupations-past-present-and-future.pdf; Anthony P. Carnevale, Ban
Cheah, and Andrew R. Hanson, “The Economic Value of College Majors,” Georgetown University Center on Education and the
Workforce, 2015, cew-7632.kxcdn.com/wp-content/uploads/The-Economic-Value-of-College-Majors-Full-Report-web-
FINAL.pdf.

2 Emily Richards and Dave Terkanian, “Occupational Employment Projections to 2022,” Monthly Labor Review, US Bureau of
Labor Statistics, December 2013, doi.org/10.21916/mlr.2013.41.

3 Throughout this paper, we use the word skills to include procedural, conceptual, and applied knowledge as a part of a
student’s overall mathematical understanding.

4 While the term gap might suggest a binary determination of student proficiency, gaps more likely reflect a partial
understanding of a particular skill or concept.

5 National Center for Education Statistics, “The Nations Report Card: Mathematics,” National Assessment of Educational
Progress, 2017, nationsreportcard.gov/math_2017/#?grade=4.

6 Michael Hansen et al., The 2018 Brown Center Report on American Education: How Well are American Students Learning? The
Brookings Institution, June 2018, brookings.edu/research/2018-brown-center-report-on-american-education-trends-in-naep-
math-reading-and-civics-scores/.

7 Ibid.

8 Catherine Gewertz, “Math Scores Slide to a 20-Year Low on ACT,” Education Week, October 17, 2018, edweek.org/ew/articles/
2018/10/17/math-scores-slide-to-a-20-year-low.html?r=1611432497&mkey=8026D2C4-0AD1-11E9-A985-EA9FC819EBCD.

9 Nick Anderson, “College Admission Test Scores Raise Warning Signs about Math Achievement,” The Washington Post, October
25, 2018, washingtonpost.com/local/education/college-admission-test-scores-raise-warning-signs-about-math-achievement/
2018/10/24/ab37ba0a-d7a5-11e8-83a2-d1c3da28d6b6_story.html?utm_term=.51a16adcad10.

10 Common Core Standards Initiative, “Key Shifts in Mathematics,” corestandards.org/other-resources/key-shifts-in-

mathematics/. Accessed February 7, 2019.

11 Resources include: “The Coherence Map,” achievethecore.org/coherence-map/. “Principles and Standards for School
Mathematics, nctm.org/Standards-and-Positions/Principles-and-Standards/. Progression Documents for the Common Core Math
Standards ime.math.arizona.edu/progressions/.

12 New Classrooms, “Teach to One: Math Major Concepts Map,” newclassrooms.org/wp-content/uploads/2016/11/

NewClassrooms_TeachtoOneMath_Concept-Map.pdf?pdf=conceptmap.

13 William H. Schmidt and Curtis C. McKnight, Inequality for All: The Challenge of Unequal Opportunity in American Schools (New
York: Teachers College Press, 2012).

14 Ibid.

15 Yeow Meng Thum and Carl H. Hauser, “2015 MAP Norms for Student and School Achievement Status and Growth,” NWEA,
November 6, 2015, solutions.nwea.org/research/2015-map-norms-for-student-and-school-achievement-status-and-growth.

16 NWEA, “Linking Studies,” nwea.org/resource/type/linking-studies/. Accessed November 1, 2018.

17 Based on four years of data across Teach to One sites (2014–2018). N = 9,037.

18 NWEA, “Linking Studies,” nwea.org/resource/type/linking-studies/. Accessed November 1, 2018.

19 J. Margolis, “Three-Year MAP Growth at Schools Using Teach to One: Math,” MarGrady Research, 2019, margrady.com/tto/.
Accessed March 5, 2019. Includes summer learning losses between sixth and seventh grades and seventh and eighth grades.

20 See Methods in Appendix I.

21 National Center for Education Statistics, “Digest of Education Statistics,” nces.ed.gov/programs/digest/d17/tables/

dt17_204.10.asp. Accessed April 11, 2019.

22 Alanna Bjorklund-Young and Jay Plasman, “Reducing the Achievement Gap: An Empirical Analysis of Middle School Math
Performance in Six States and Washington, D.C.,” The Institute for Educational Policy at Johns Hopkins University, April 2019,
edpolicy.education.jhu.edu/wp-content/uploads/2019/04/Achievement-Gap-Policy-Brief.pdf. Accessed April 17, 2019.

23 Ibid.

24 The term was coined in the 1930s by psychologist Lev Vygotsky; see more in Seth Chaiklin, “The Zone of Proximal
Development in Vygotsky’s Analysis of Learning and Instruction,” in Vygotsky’s Educational Theory and Practice in Cultural
Context, ed. Alex Kozulin et al. (Cambridge, UK: Cambridge University Press, 2003), people.ucsc.edu/~gwells/Files/
Courses_Folder/documents/chaiklin.zpd.pdf.

Jennifer Hammond and Pauline Gibbons, “What Is Scaffolding?” Teachers’ Voices 8 (2005), ameprc.mq.edu.au/docs/
research_reports/teachers_voices/Teachers_voices_8.pdf#page=15.
ENDNOTES 64

25 Jennifer Hammond and Pauline Gibbons, “What Is Scaffolding?” Teachers’ Voices 8 (2005), ameprc.mq.edu.au/docs/
research_reports/teachers_voices/Teachers_voices_8.pdf#page=15.

26 R. M. Harden and N. Stamper, “What Is a Spiral Curriculum?” Medical Teacher 21, no. 2 (1999), faculty.med.virginia.edu/
facultyaffairs/files/2016/04/2010-3-23.pdf.

27 See Appendix I.

28 Chrys Dougherty and Steve Fleming, “Getting Students on Track to College and Career Readiness: How Many Catch Up from
Far Behind?” ACT, November 2012, eric.ed.gov/?id=ED542022.

29 Ibid.

30 Jason Zimba, National Council of Teachers of Mathematics 2017 Research Conference, April 7, 2017, San Antonio, Texas.

31 Amber M. Northern and Michael J. Petrilli, “Credit Recovery: Good Intentions, Poor Execution,” The Fordham Institute,
edexcellence.net/articles/credit-recovery-good-intentions-poor-execution. Accessed February 7, 2019.

32 114th Congress, “S.1177—Every Student Succeeds Act,” Section 1111(b).

33 ESSA text 1111(b)(2)(J); ESSA permits states to exempt eighth-grade students taking advanced math courses from the
statewide math assessment used for to eighth-grade students. Students in such courses can instead take the corresponding end-
of-course assessment, so long as their results on the end-of-course assessment are included in accountability determinations for
that year and they take another more advanced mathematics assessment in high school, and that score is included in
accountability determinations for students’ high schools.

34 Exceptions for eighth-grade students in algebra: 1111(b)(2)(C) of the Elementary and Secondary Education Act of 1965 (ESEA),
as amended by the Every Student Succeeds Act (ESSA).

35 “A State’s Guide to the U.S. Department of Education’s Assessment Peer Review Process,” US Department of Education, 2018,
p. 25, 2.ed.gov/admins/lead/account/saa/assessmentpeerreview.pdf.

36 “What’s in the Every Student Succeeds Act? Accountability,” Education Trust, January 13, 2016, edtrust.org/resource/whats-
in-the-every-student-succeeds-act-accountability/.

37 Susan Lyons, Juan D’Brot, and Erika Landl, “State Systems of Identification and Support under ESSA: A Focus on Designing
and Revising Systems of School Identification,” CCSSO, November 2017, pp. 4–12, ccsso.org/sites/default/files/2017-12/
State%20Systems%20of%20ID%20and%20Support%20-%20Designing%20and%20Revising%20Systems_0.pdf.

38 Council of Chief State School Officers, “Considerations for Including Growth in ESSA State Accountability Systems,” January
2017, p. 4, ccsso.org/sites/default/files/2017-10/CCSSOGrowthInESSAAccountabilitySystems1242017.pdf.

39 (c)(4)(C)(ii) of Section 1111 of ESSA, congress.gov/bill/114th-congress/senate-bill/1177/text.

40 Elizabeth Ross et al., “2017 State Teacher Policy Yearbook,” NCTQ, 2017, p. 79, nctq.org/dmsView/
NCTQ_2017_State_Teacher_Policy_Yearbook.

41 For more information on growth measures: Juan D’Brot, “Considerations for Including Growth in ESSA State Accountability
Systems,” CCSSO, January 2017, nciea.org/sites/default/files/pubs-tmp/CCSSO_Growth_Resource.pdf; “Students Can’t Wait:
Individual Student Growth,” Education Trust, edtrust.org/students-cant-wait/individual-student-growth/. Accessed October 20,
2018. “Progress Tables (Value Tables): Another Measure of Student Growth,” Virginia Department of Education, January 21, 2015,
doe.virginia.gov/boe/committees_standing/accountability/2015/meeting_materials/jan-21_measure_of_student_growth.pdf;
Katherine E. Castellano and Andrew D. Ho, “A Practitioner’s Guide to Growth Models,” Council of Chief State School Officers,
February 2013, scholar.harvard.edu/files/andrewho/files/a_pracitioners_guide_to_growth_models.pdf?m=1364611983; “Value-
Added Modeling 101: Using Student Test Scores to Help Measure Teaching Effectiveness,” RAND Corporation, 2012, rand.org/
pubs/corporate_pubs/CP693z4-2012-09.html; “Growth Data: It Matters and It’s Complicated,” Data Quality Campaign, January
2019, dataqualitycampaign.org/resource/growth-data-it-matters-and-its-complicated/.

42 Richard J. Patz, “Vertical Scaling in Standards-Based Educational Assessment and Accountability Systems,” Council of Chief
State School Officers, 2007, pdfs.semanticscholar.org/c1b8/e97bb12142ac842e88da4f825e72ddf9dbc7.pdf.

43 Madhabi Chatterji, PhD, “A Consumer’s Guide to Testing Under Every Student Succeeds Act (ESSA): What Can the Common
Core and Other ESSA Assessments Tell Us?” Prepared for the National Education Policy Center, February 21, 2019,
nepc.colorado.edu/publication/rd-assessment-guide.

44 M. Chatterji, personal communication, March 7, 2019.

45 Julie Woods, “50 State Comparison: States’ School Accountability Systems,” Education Commission of the States, May 31,
2018, ecs.org/50-state-comparison-states-school-accountability-systems/.

Matt Barnum, “The Growth vs. Proficiency Debate and Why Al Franken Raised a Boring but Critical Issue,” The 74 Million,
January 18, 2017,the74million.org/article/barnum-the-growth-vs-proficiency-debate-and-why-al-franken-raised-a-boring-but-
critical-issue/.

Data Quality Campaign, “Growth Data, It Matters and Its Complicated,” dataqualitycampaign.org/resource/growth-data-it-
ENDNOTES 65

46 Matt Barnum, “The Growth vs. Proficiency Debate and Why Al Franken Raised a Boring but Critical Issue,” The 74 Million,
January 18, 2017, the74million.org/article/barnum-the-growth-vs-proficiency-debate-and-why-al-franken-raised-a-boring-but-
critical-issue/.

47 Data Quality Campaign, “Growth Data, It Matters and Its Complicated,” dataqualitycampaign.org/resource/growth-data-it-
matters-and-its-complicated/. Accessed on January 23, 2019.

48 Brandon L. Wright and Michael J. Petrilli, “Rating the Ratings: An Analysis of the 51 ESSA Accountability Plans,” Thomas B.
Fordham Institute, November 2017, edex.s3-us-west-2.amazonaws.com/publication/pdfs/11.15%20-
%20Rating%20the%20Ratings%20-%20An%20Analysis%20of%20the%2051%20ESSA%20Accountability%20Plans.pdf.

49 “Show Me the Data 2017,” Data Quality Campaign, 2pido73em67o3eytaq1cp8au-wpengine.netdna-ssl.com/wp-content/

uploads/2018/01/DQC-Show-Me-the-Data_final.pdf. Accessed October 22, 2018.

50 “Growth Fact Sheet,” Colorado Department of Education, August 2018, cde.state.co.us/accountability/growth-fact-sheet-for-

parents-2018.

51 Smarter Balanced Assessment Consortium, “Race to the Top Assessment Program Application,” US Department of Education,
2010.

52 T. Alpert, personal communication, November 2, 2018.

53 For example, “Linking the PARCC Assessments to NWEA MAP Growth Tests,” NWEA, November 2016, nwea.org/content/
uploads/2017/07/PARCC-MAP-Linking-Study_2016.pdf.

54 Every Student Succeeds Act (ESSA), section 1204, “Innovative Assessment and Accountability Demonstration Authority,” 129
STAT. 1885.

55 Ibid.

56 US Department of Education Office of Elementary and Secondary Education, “A State’s Guide to the U.S. Department of
Education’s Assessment Peer Review Process,” September 24, 2018.

57 Every Student Succeeds Act (ESSA), section 1204, “Innovative Assessment and Accountability Demonstration Authority,” 129
STAT. 1885. Note that USED has permitted an additional two year grace period.

58 “Innovative Assessment Demonstration Authority (IADA),” U.S. Department of Education, September 28, 2018, ed.gov/
admins/lead/account/iada/index.html.

59 Ibid.

60 Ibid.

61 Ibid.

62 “Explore the Latest NAEP Mathematics Results,” The Nation’s Report Card, nationsreportcard.gov/math_2017?grade=8.
Accessed February 7, 2019.

63 J. Margolis, “Three-Year MAP Growth at Schools using Teach to One: Math,” MarGrady Research, 2019, margrady.com/tto/.
Accessed March 5, 2019. Note that a broader group of students, including those not continuously enrolled, showed average
three-year gains of 13 percentile points.

64 Ibid.

65 Ibid.

66 D. Ready et al., Final Impact results from the i3 implementation of Teach to One: Math. Columbia University, Consortium for
Policy Research in Education, New York, 2018.

67 Schmidt and McKnight (2012); William Schmidt, “At the Precipice: The Story of Mathematics Education in the United States,”
Peabody Journal of Education 87, no. 1 (February 1, 2012): 133–156, doi.org/10.1080/0161956X.2012.642280; Morgan Polikoff, “The
Redundancy of Mathematics Instruction in US Elementary and Middle Schools,” Elementary School Journal 113, no. 2 (2012): 230–
251, journals.uchicago.edu/doi/abs/10.1086/667727.

68 Tom Loveless, “The Misplaced Math Student: Lost in Eighth Grade Algebra,” Brookings Institution, September 2008,
brookings.edu/wp-content/uploads/2016/06/0922_education_loveless.pdf.

69 Charles T. Clotfelder, Helen Ladd, and Jacob Vigdor, “Algebra for Eighth Graders: Evidence on Its Effects from 10 North
Carolina Districts,” The Calder Center, February 2013, caldercenter.org/sites/default/files/wp87-2.pdf.

70 “School Quality Rating Policy (SQRP) Handbook,” Chicago Public Schools, September 27, 2017, cps.edu/Performance/
Documents/SQRPHandbook.pdf.

“2017–18 School/Parent Guide to the Elementary School Promotion Policy,” Chicago Public Schools, 2017, cps.edu/
SiteCollectionDocuments/ElemPromotionPolicy_English.pdf.
ENDNOTES 66

71 “2017–18 School/Parent Guide to the Elementary School Promotion Policy,” Chicago Public Schools, 2017, cps.edu/
SiteCollectionDocuments/ElemPromotionPolicy_English.pdf.

72 Julia Gwynne and Sarah Cashdollar, “Changes in Math Instruction and Student Outcomes Since the Implementation of the
Common Core State Standards in Chicago,” University of Chicago, Consortium on School Research, September 2018,
consortium.uchicago.edu/sites/default/files/publications/Changes%20in%20Math%20Instruction-Sep2018-Consortium.pdf.

73 Alexandra Arriaga, “Record Number of CPS Students Graduating with College, Career Credentials,” Chicago Sun-Times,
August 31, 2018, chicago.suntimes.com/news/cps-chicago-public-high-schools-graduating-college-career-credentials-test-
scores/.

74 Gwynne and Cashdollar, “Changes in Math Instruction.”

75 Tara Garcia Mathewson, “Psst! When Teachers Get Useful, Timely Data, They Use It,” The Hechinger Report, November 1, 2017,
hechingerreport.org/psst-teachers-get-useful-timely-data-use/.

76 Interview and communications with TPS officials.

77 “Voices from the Classroom: A Survey of America’s Educators,” Educators for Excellence, 2018, e4e.org/news/voices-
classroom-survey-americas-educators.

78 Eric M. Anderman et al., “Value-Added Models of Assessment: Implications for Motivation and Accountability,” Journal
Educational Psychologist 45, no. 2 (April 2010): 123–137, doi.org/10.1080/00461521003703045.

79 Frank Borgan, “Dear Colleague ESEA Letter on State Plan Amendments,” November 14, 2018, ed.gov/policy/elsec/leg/essa/
dclessaspamendmentprocessltr.pdf.

80 Chirstina Quattrocchi, “How Lindsay Unified Redesigned Itself from the Ground Up,” June 17, 2014, edsurge.com/news/
2014-06-17-how-lindsay-unified-redesigned-itself-from-the-ground-up. Accessed February 7, 2019.

81 Texas Education Agency (TEA), “Math Innovation Zones,” tea.texas.gov/Academics/Learning_Support_and_Programs/

Math_Innovation_Zones/. Accessed November 1, 2018.

82 Ibid.

83 Barshay, "Gifted Classes May Not Help Students Move Ahead Faster," The Hechinger Report, April 15, 2019,
hechingerreport.org/gifted-cla sses-may-not-help-talented-st udents-move-ahead-faster/

84 “Linking the PARCC Assessments to NWEA MAP Tests for Illinois,” Northwest Evaluation Association (NWEA), March 2016,
nwea.org/content/uploads/2017/06/Illinois_PARCC_Linking_Study_MAR2016.pdf.

85 Ira Nichols-Barrer et al., “Predictive Validity of MCAS and PARCC: Comparing 10th Grade MCAS Tests to PARCC Integrated
Math II, Algebra II, and 10th Grade English Language Arts Tests,” Mathematica Policy Research Report, October 5, 2015,
mathematica-mpr.com/download-media?MediaItemId=%7B8EF64B21-27FC-48D0-B91F-B8BA3FA95E47%7D.

86 “A Nation at Risk: The Imperative for Educational Reform: A Report to the Nation and the Secretary of Education, United
States Department of Education,” National Commission on Excellence in Education, 1983, ed.gov/pubs/NatAtRisk/risk.html.

87 103rd Congress, “Improving America’s Schools Act of 1994,” October 20, 1994, congress.gov/bill/103rd-congress/house-bill/
6/text.

88 81st Congress, “Elementary and Secondary Education Act of 1965,” December 10, 2015, ed.gov/about/offices/list/oii/
nonpublic/eseareauth.pdf.

89 “Summary of the Improving America’s Schools Act,” Education Week, November 9, 1994, edweek.org/ew/articles/
1994/11/09/10asacht.h14.html.

90 Dianne Piche, “Closing the Deal: A Preliminary Report on State Compliance with Final Assessment and Accountability
Requirements under the Improving America’s Schools Act of 1994,” Citizens Commission on Civil Rights, March 1, 2001,
files.eric.ed.gov/fulltext/ED460200.pdf.

91 107th Congress, “No Child Left Behind Act of 2001,” January 8, 2002, congress.gov/bill/107th-congress/house-bill/1/text.

92 NCLB, 2001.

93 Alyson Klein, “No Child Left Behind: An Overview,” Education Week, April 10, 2015, edweek.org/ew/section/multimedia/no-
child-left-behind-overview-definition-summary.html.
ENDNOTES 67

94 David Conley, “The Common Core State Standards: Insight into Their Development and Purpose,” Council of Chief State
School Officers, June 2014, p. 3, inflexion.org/ccss-development-and-purpose/.

95 “Ensuring U.S. Students Receive a World-Class Education,” National Governors Association, the Council of Chief State School
Officers, and Achieve, 2008, p. 28, edweek.org/media/benchmakring%20for%20success%20dec%202008%20final.pdf.

96 Conley, “The Common Core State Standards,” p. 5.

97 “President Obama, U.S. Secretary of Education Duncan Announce National Competition to Advance School Reform,” press
release, U.S. Department of Education, July 24, 2009, ed.gov/news/pressreleases/2009/07/07242009.html.

98 Common Core State Standards Initiative, “About the Standards: Development Process,” corestandards.org/about-the-
standards/development-process/. Accessed October 23, 2018.

99 Daniel Hamlin and Paul Peterson, “Have States Maintained High Expectations for Student Performance?” Education Next 18,
no. 4 (2018),educationnext.org/have-state s-maintained-high-expectations-student-performance-analysis-2017-proficiency-
standards/.

100 Andrew Ujifisa, “Resistance to the Common Core Mounts,” Education Week, April 21, 2014, edweek.org/ew/articles/
2014/04/23/29cc-backlash.h33.html.

101 ESSA text—ESEA 1111(b)(1).

102 ESSA text—ESEA 1111(b)(1).

103 ESSA text—ESEA 1111(b)(1)(G)(ii).

104 State websites.

105 Solomon Friedberg et al., “The State of the Standards Post-Common Core,” Thomas Fordham Institute, August 2018, edex.s3-
us-west-2.amazonaws.com/publication/pdfs/(08.22)%20The%20State%20of%20State%20Standards%20Post-
Common%20Core.pdf.

106 Ashley Jochim and Patrick McGuinn, “The Politics of the Common Core Assessments,” Education Next 16, no. 4 (2016),
educationnext.org/the-politics-of-common-core-assessments-parcc-smarter-balanced/.

107 Alex Johnson, “Majority of States Line Up to Ditch No Child Left Behind,” NBC News, September 30, 2011, nbcnews.com/id/
44693695/ns/us_news-education_nation/t/majority-states-lining-ditch-no-child-left-behind/#.W1XUfapKjIU.

108 Michele McNeil and Alyson Klein, “Obama Offers Waivers from Key Provisions of No Child Left Behind,” Education Week,
November 27, 2011, edweek.org/ew/articles/2011/09/28/05waiver_ep.h31.html.

109 “ESEA Flexibility,” Laws & Guidance/Elementary & Secondary Education, US Department of Education, ed.gov/policy/elsec/
guid/esea-flexibility/index.html. Accessed October 23, 2018.

110 “White House Report: Every Student Succeeds Act,” The White House Office of the Press Secretary, December 10, 2018,
obamawhitehouse.archives.gov/the-press-office/2015/12/10/white-house-report-every-student-succeeds-act.

111 ESSA text—ESEA 1111(b)(2)(J); ESSA permits states to exempt eighth-grade students taking advanced math courses from the
statewide math assessment used for to eighth-grade students. Students in such courses can instead take the corresponding end-
of-course assessment, so long as their results on the end-of-course assessment are included in accountability determinations for
that year, and they take another more advanced mathematics assessment in high school and that score is included in
accountability determinations for students’ high schools.

112 “FAQ: The Every Student Succeeds Act: Assessment Flexibility Under Title I Assessments Section 1111(b)(2),” Knowledge
Works, accessed October 23, 2018, knowledgeworks.org/wp-content/uploads/2018/01/essa-faqs-title1-assessments.pdf.

113 Alyson Klein, “Louisiana, New Hampshire, and Puerto Rico Apply for ESSA Innovative Testing Pilot,” Education Week, April 3,
2018, blogs.edweek.org/edweek/campaign-k-12/2018/04/ESSA_testing_pilot_louisiana_new_hampshire_and_puerto_rico.html.

114 “What’s in the Every Student Succeeds Act?—Accountability,” Ed Trust, January 13, 2016, edtrust.org/resource/whats-in-the-
every-student-succeeds-act-accountability/.

115 Lyons, D’Brot, and Landl, “State Systems of Identification and Support under ESSA.”

116 D’Brot, “Considerations for Including Growth in ESSA State Accountability Systems,” p. 4.

117 Alyson Klein, “ESEA Reauthorization: The Every Student Succeeds Act Explained,” Education Week, November 30, 2015,
blogs.edweek.org/edweek/campaign-k-12/2015/11/esea_reauthorization_the_every.html