From Hammer to Flashlight

From Hammer

to Flashlight

January 2017

A Decade of Data

in Education

The Data Quality Campaign is a nonprofit policy and advocacy organization leading the

eort to bring every part of the education community together to empower educators,

families, and policymakers with quality information to make decisions that ensure that

students excel. For more information, go to www.dataqualitycampaign.org and follow us

on Facebook and Twitter (@EdDataCampaign).

From Hammer

to Flashlight

A Decade of Data

in Education

CONTENTS

A Letter from Aimee 1

1. Making a Dierence with Data 2

2. Becoming a Data-Driven Sector 4

3. Policy Leadership Made It Happen 8

4. Challenges 11

5. Lessons and Recommendations 12

Appendix A: Information-Gathering Process to Inform

Development of This Paper 16

Appendix B: Expanded History of Building an

Education Data Infrastructure 17

Appendix C: Original DQC Managing and Endorsing Partners 21

Appendix D: DQC’s Policy Recommendations for States

from 2005 to Today 22

Appendix E: Acknowledgments 27

From Hammer to Flashlight: A Decade of Data in Education 1

A Letter from Aimee

Using data eectively is at the core of every successful institution—whether it be corporate, nonprofit,

or public. The world’s best companies have long used the wealth of information made possible through

advancing technology to refine their marketing, supply chain, and management strategies. And it is not

just big corporations using data for improvement. Individuals are using data in many ways, from health

devices that help tailor diet and exercise routines to personal finance soware that aggregates data across

accounts and lines of credit. Yet public sectors have been slow to embrace the power of information to inform

decisionmaking, tailor government services to meet people’s needs, and guide the allocation of scarce

resources into approaches that work.

People are beginning to recognize that government must use data to improve how it serves citizens. In 2015,

Congress created the Commission on Evidence-Based Policymaking to study how the federal government can

support the use of data throughout government operations. As the commission, policymakers, and thought

leaders at all levels focus on how to build a culture that values and uses data, the education sector provides a

worthy case study. Over the past decade, every state has built a longitudinal education data system, and most

are now taking steps to ensure that data is used to improve student achievement and system performance.

Indeed, the education sector has made great progress toward having a universal evidence-based culture that

serves all students, and this paper chronicles the journey it took to get here. From key events to challenges to

recommendations, we aim to provide insights for the field moving forward—because this work is not done.

Progress notwithstanding, education has not yet fully become an evidence-based sector. In our conversations

with the field, the word that most oen comes up to describe the education sector’s use of data is nascent. Our

hope is that this retrospective will not merely capture the past but help inform future eorts in education and

other public sectors alike.

The Data Quality Campaign’s partners used to joke that our informal goal was to “make data sexy”—shorthand

for getting people other than specialists to become passionate champions for the power of data to transform

education into a personalized, results-focused endeavor. As it turns out, that idea is one of the critical drivers

identified in this analysis. But I will not give away all the “ahas”—you will have to read the entire paper for that!

With best wishes for our collective eorts to build the culture of evidence our kids deserve,

Aimee Rogstad Guidera

Founder, President, and CEO, Data Quality Campaign

Data Quality Campaign2

CHAPTER 1

Making a Dierence with Data

While the corporate world has long been using data to streamline practices and meet business goals, the majority of public

sectors have lagged far behind in becoming evidence-based fields. Education is an exception. For years the public education

sector has been building data systems and developing a culture of data use to better inform and reach its goals. Although

much work remains before education becomes a truly evidence-based field, its incredible progress oers many lessons. This

paper seeks to highlight the successes and illuminate the challenges that have accompanied this progress, both for other

sectors to learn from and for the education field to examine as it continues to build the culture, capacity, and conditions to

use data to improve education for all students.

All states now have robust longitudinal education data systems

to provide information that is richer than ever to stakeholders

across the nation. This data infrastructure has made it possible

to shi education from a one-size-fits-all system built to meet

the demands of yesterday’s industrial economy to a model that

can provide every child a personalized learning experience that

is geared toward meeting the demands of today’s knowledge

economy.

This data infrastructure provides evidence to help answer

critical policy and practice questions to improve education.

How many high school graduates take remedial courses

in college? How many ninth graders end up maintaining

continuous enrollment and completing their high school

requirements on time? Which teacher preparation programs

produce teachers whose students have the strongest academic

growth? Before we had the infrastructure capable of collecting

and analyzing the relevant data, these questions and so many

more went unanswered and unaddressed. States must have

a clear picture of where their students are and the factors

that shape their performance to truly meet the needs of all

students—and having the right data is critical to fulfilling this

responsibility.

As a result of this infrastructure growth, many states are now

able to take the following actions:

 understand how each class of high school graduates fares

in postsecondary, like Hawaii does with its high school

feedback reports

 identify and support students who may be falling o

track to high school graduation before it is too late, like

Massachusetts does with its early warning indicator system

 analyze the eectiveness of teacher preparation programs

throughout the state, like Tennessee does with its teacher

preparation report card

And those are just a few examples. Thanks to the development

of data infrastructure in education, leaders at all levels can

better serve students, educators, and the public:

 States can support teachers to personalize learning for

every child by equipping teachers with tools to meet the

needs of all their students, even those who may require

more specialized instruction. Georgia combined local data

with state-level resources to help its teachers easily view

their students’ individual progress over time in various

subjects and create personalized learning activities that

build on strengths and fill gaps.

 School and system leaders can ensure that resources are

being allocated to support learning and the success of

every child. Chicago Public Schools has successfully used

data to keep high school freshmen on track by providing

every high school educator a set of on-track indicators,

including chronic absence, about their students. Between

2007 and 2014, the rate of students on track to graduate

in Chicago rose from 57 to 84 percent, which represents

7,000 additional students on track to graduate each year.

The district could not have achieved these results without

key resources to identify academic pathways and eective

interventions tailored to specific student needs.

 Policymakers can provide stronger accountability to

assure families and taxpayers that dollars are being spent

to prepare students for success beyond high school and

build trust by investing in richer, easier to understand

public report cards. Washington’s Education Research

and Data Center is a state-legislated, state-funded eort

that provides publicly available data and reports on its

website to answer questions about the health of the state

education system such as, “What percentage of high school

graduates enrolled in postsecondary education in the year

aer graduation?” and “Are Washington students earning a

postsecondary credential by age 26?”

The relatively new data infrastructure in education is not

perfect, but states have come a long way since 2005. With the

benefit of hindsight, the education sector has learned that

becoming a data-driven sector means not just building data

systems but also focusing on people. Leaders must engage

stakeholders to understand their needs and earn their trust,

prioritize using data as a critical tool in reaching education

From Hammer to Flashlight: A Decade of Data in Education 3

goals, and ensure that people have access to the data they

need and know how to use it.

Data itself does not improve teaching and learning. Too oen

in education data is seen as a hammer—a tool of accountability

to ensure that targets are being met. While accountability is

important, blame and shame oen follow when results fall

short. Shiing this paradigm and moving beyond accountability

opens the door to a vast array of opportunities to use data

as a flashlight, shining a light on what is working and fueling

continuous improvement. The culture of education is beginning

to embrace the true potential of data—not just to comply

with requirements but also to inform decisions and drive

improvement. This transformation has been the result of

shared, and oen coordinated, policy leadership at the federal

and state levels, with sustained support from philanthropic and

advocacy organizations.

The data infrastructure largely exists, but more work remains

to build the capacity, conditions, and culture to use data to

truly support success. Nowhere is data being used to its fullest

potential to illuminate challenges and identify solutions for

all students. Building data infrastructure and transforming

culture are diicult undertakings, but taking these steps has

never been more urgent to empower with information all those

invested in student success.

Though work still remains, the education sector has become

a leader in using data to drive results. Other public sectors

can learn from its successes and challenges along the way.

Based on insights from leaders across the education field,

this paper provides a brief history and substantive analysis of

the education sector’s journey of building data infrastructure

and beginning to develop a culture that values evidence for

improving decisionmaking, system performance, and individual

student success. (See Appendix A for a description of the

research process.) Recommendations are also provided on how

to build an evidence-based culture in any public sector based

on lessons from the education field.

Data Quality Campaign4

CHAPTER 2

Becoming a Data-Driven Sector

To understand how education has transformed into an emerging data-driven sector, one must look to the key events and

actors driving the change. State and federal policymakers, advocacy leaders, and others have played critical roles, and

progress has been animated by events from the publishing of A Nation at Risk in 1983 to the enactment of the Every Student

Succeeds Act (ESSA) in 2015. This section provides a summary of the major historical factors driving education toward

eective data use. (For a more detailed chronology, see Appendix B.)

1980–2004: States Take the Lead in Education

The quarter century from 1980 to 2004 was transformational

for the education sector, as state and federal leaders assumed

new roles and responsibilities in improving student outcomes.

A Nation at Risk: The Imperative for Educational Reform sounded

an alarm that American schools were failing, putting the

nation’s economic future at risk. The report is notable not only

for its candid discussion of an “emerging national sense of

frustration” about how “more and more young people emerge

from high school ready neither for college nor for work” but

also for its use of data to support its argument—from analyses

of SAT scores over time to results from a survey of teacher

preparation institutions.

For state and federal leaders, the charge was clear: ensure that

America’s schools deliver a globally competitive education to its

students. States like Kentucky and Massachusetts began taking

responsibility for improving student outcomes and passed

laws expanding the state role in education. Southern states

like Texas looked to the private sector as a model and began

collecting and using data to improve student performance.

Building on state momentum, the federal No Child Le Behind

Act (NCLB) was enacted in 2002, requiring states to annually

test students, disaggregate data by student subgroups, and

publicly report the results, attaching high-stakes accountability

to data. This shi refocused federal education priorities onto

states to use data to track goals and spur improvements in

education outcomes. These events set the stage for a period

of exponential growth in the development and use of data in

education.

2005–11: Policymakers, Philanthropy, and Advocacy Organizations Prioritize

Data Use Across the Nation

From 2005 to 2011 state and federal policymakers came

together with advocacy and philanthropic organizations to

define a shared vision for eective data use. The vision rested

on the premise that states were best positioned to take the lead

in developing and using high-quality data systems to answer

critical policy and practice questions. For example, in 2005 all

50 state governors signed the National Governors Association

(NGA) Graduation Rate Compact, agreeing to implement

a common formula for calculating high school graduation

rates in their states. This agreement has allowed for a more

accurate comparison of graduation rates across states and for

a consistent calculation of a national high school graduation

rate, which has been increasing each year and reached an all-

time high for the class of 2014–15.

During this time period, the federal government began

supporting states’ development and use of high-quality data

through a number of eorts:

 Since 2005 the congressionally created Statewide

Longitudinal Data Systems (SLDS) Grant Program has

helped states build, improve, and use their data systems.

By September 2016, 47 states, the District of Columbia,

Puerto Rico, the US Virgin Islands, and American Samoa had

successfully secured more than $500 million in grants for

their data eorts.

 Federal regulations in 2008 and 2011 clarifying NCLB and

the Family Educational Rights and Privacy Act (FERPA)

allowed states to continue to securely develop their data

infrastructure to provide meaningful and useful data.

One NCLB regulation built o of states’ Graduation Rate

Compact work and required states to submit a longitudinal

“ That’s where data comes in. Some places are keeping

electronic records of how a student does from one year to

the next and how a class does in any given year. This helps

students, parents, teachers, principals, and school boards

know what’s working and what’s not in the classroom. You

know, basketball coaches have a game tape for the team to

see what they did right and what they did wrong aer a tough

series—teachers and principals should have a way of doing the

same.”

— President Barack Obama, announcing the Race to the Top

competition in 2009

From Hammer to Flashlight: A Decade of Data in Education 5

statistic for the first time ever: the four-year adjusted cohort

graduation rate.

 The American Recovery and Reinvestment Act of 2009

(ARRA) established three separate funding mechanisms

for states to use in their eorts to build and use their SLDS.

This act included the Obama administration’s signature

education program, Race to the Top, which challenged

states to think dierently about how they would leverage

their SLDS in support of teaching and learning. The program

also elevated the conversation about data systems and use

beyond state education agency oicials to governors and

state legislatures.

To provide states support for building their data infrastructure

and eectively using data, the philanthropic community (e.g.,

the Bill & Melinda Gates Foundation, the Eli and Edythe Broad

Foundation, and the Michael and Susan Dell Foundation)

invested in education policy and advocacy organizations,

including the Data Quality Campaign (DQC), to prioritize data

and support evidence-based decisionmaking at all levels (see

sidebar on DQC’s role in the rise of data in education).

The maps below illustrate the tremendous advances states

made between 2005 and 2011 in building data systems to

collect, store, and use data to improve student achievement.

By 2011, most states had data systems that included all of

DQC’s 10 Essential Elements of Statewide Longitudinal Data

Systems (see Appendix D for more details). This radical change

in infrastructure allowed for opportunities to build tools that

put data to work.

Eective Data Use: State Progress

10 Essential Elements of Statewide Longitudinal Data Systems

2–3 0–1

4–5

6–7

8–9 10

2–3 Elements

0–1 Element

4–5 Elements

6–7 Elements

8–9 Elements

10 Elements

2–3 Elements

0–1 Element

4–5 Elements

6–7 Elements

8–9 Elements

10 Elements

2013: States and the Federal Government Focus on Student Data Privacy

Concerns

While states were making progress building and using

data systems with the support of federal, advocacy, and

philanthropic eorts, many other stakeholders, like families

and community members, were being le out of the

conversation. In 2013, inBloom, a nonprofit that oered

education data management and analysis services to states

and school districts to help them make better use of their data,

launched its trademark data platform with much fanfare at

the South by Southwest education conference. With funding

provided by the Gates Foundation and others, the product

promised a way to use hundreds of data points about students

to personalize instruction. However, poor communication;

lack of transparency; and a misunderstanding of how little the

public, especially parents, understood about data in education

led inBloom to dissolve a year later.

States were busy investing in education data, but they were

not communicating about the value of data and earning the

public’s trust. A wave of backlash from parents and the public

about the perceived intrusion of government and big business

into the lives of children through data prompted states and

the federal government to introduce hundreds of pieces of

legislation to protect student data privacy. Between 2013 and

2016, 410 bills were introduced in 49 states and the District

of Columbia, resulting in 36 states passing 74 student data

privacy bills into law. At the federal level, a number of student

data privacy bills have been introduced, including a proposed

update to the 40-year-old FERPA, but none has yet resulted in a

change to federal law.

2005 2011

Data Quality Campaign6

2015: New Federal Law Shines a More Powerful Light on Data Use

One federal education eort that did pass was ESSA in 2015.

ESSA represents a change in federal education policy that gives

states greater flexibility—and greater responsibility—to make

decisions about policies and practices to support all students’

success and close achievement gaps. ESSA maintains the

commitment of its predecessor, NCLB, to using data to examine

what is working for students—and what is not—to meet states’

education goals.

While powerful systems have been built and policies have

been created to protect data and guide its use, the education

field has not yet fully become an evidence-based sector.

Infrastructure has improved, but data systems will always

need maintenance to keep up with changes in technology and

people’s information needs. Equally important, the field is just

starting to address the conditions, capacity, and culture needed

to use the new information created by these data systems. As

the education sector now launches into the even harder work

of making data work for all students, the events outlined in

this chapter can provide insights that may inform other sectors

looking to make this transformation.

The Data Quality Campaign’s Role in the Rise of Data in Education

The Data Quality Campaign (DQC) was launched in 2005 by

14 advocacy and constituency organizations that recognized

the need for a national, collaborative eort to encourage

and support the use of high-quality, accessible data in

education. With the support of their funders, these founding

partners (see Appendix C) put aside their sometimes

conflicting policy agendas to align around the priority of

increasing the availability and use of data in education. To

ensure that the eort was truly collaborative, DQC was not

started as a separate nonprofit but rather was housed at the

National Center for Educational Achievement and managed

and run by the partner organizations.

Since its launch, DQC has been working to create a

sectorwide culture in which high-quality data is not only

collected but also used to inform action and improve

student achievement. DQC has used roadmaps for

state policy like the 10 Essential Elements of Statewide

Longitudinal Data Systems and 10 State Actions to Ensure

Eective Data Use to measure and celebrate state progress.

DQC became a fully independent nonprofit organization in

2011 and has continued to work with a growing network of

more than 100 partners to produce resources, messages,

and forums that will nurture the nascent culture of evidence

in the education sector.

The following strategies have guided DQC’s advocacy eorts

over the past decade and are a potential roadmap for other

sectors looking to build an evidence-based culture.

✔ Create a National Forum

Before DQC launched in 2005, there was no national voice

or advocate focused on education data policy and practice.

People were using data, but the culture of data use in

education was underdeveloped. Few states had quality

longitudinal data, much less eective or thoughtful policies

and practices for using the data. States were still adjusting

to the No Child Le Behind Act’s data requirements, and few

state policymakers knew what longitudinal data systems

were or how critical they were to informing and improving

student success. Yet leading states were already seeking

opportunities to build data systems and recognized the

need for data infrastructure as the federal government was

mandating and incentivizing improved data collection.

In this environment, DQC was launched as a national forum

to highlight those emerging practices and provide a place

for interested partners to convene, learn from one another,

and share information with the field. DQC’s founding

partners recognized the need for a full-time advocate that

could lead the way by thinking about education data full

time. Together these partners led eorts to build consensus

and collaboration as states developed their longitudinal

data systems. When concerns from the public around the

privacy and security of data increased in 2013, having a

national forum to gather, listen, learn, and address these

legitimate issues proved invaluable to the eorts to create

eective and trusted use of data to improve student

achievement.

“It is not every day that the National Education Association, the

American Federation of Teachers and the nation’s governors

agree on such important education policy. I urge you to take

advantage of this, state by state, while you can. Maybe it will

help to put education on the front burner and politics on the

back burner in our schools.”

—Senator Lamar Alexander (R-TN)

From Hammer to Flashlight: A Decade of Data in Education 7

✔ Create Evidence-Based Roadmaps and

Tools

DQC’s 10 Essential Elements of Statewide Longitudinal

Data Systems and 10 State Actions to Ensure Eective

Data Use provided a common language around education

data that was not technical or related to information

technology (IT). The purpose of these tools was to help

states use data for continuous improvement rather than

compliance. State policymakers now had clear, measurable

policy roadmaps that provided actions to take to build

data systems and ensure conditions and capacity for data

use. DQC measured and celebrated state progress on the

Elements and Actions and highlighted best practices in

implementation. The lessons learned were then distilled so

states could build on and improve their data infrastructures

as they worked to become leading states. By 2011, 36 states

had all 10 Essential Elements in place. By 2014, the final

year DQC surveyed states on their progress toward the 10

State Actions, three states had implemented all of them:

Arkansas, Delaware, and Kentucky.

✔ Advocate For and Support Changes in

Policy and Practice to Ensure That Data

Eectively and Securely Follows and

Serves the Individual

Changing the role of data in a sector depends not only

on eective IT strategy and practice but also on policy

leadership. Policymakers—especially state policymakers—

have been DQC’s target audience since day one. Eective

data systems and their use are essentially about meeting

people’s information needs—from parents to educators to

policymakers. When DQC launched, its 14 partners surveyed

their members for the questions they (governors, chief

state school oicers, legislators, state board members)

most needed to answer; DQC released the 10 Essential

Elements not as a standalone list of to-dos, but as actions

that were necessary to take so that policymakers could get

the answers to the very questions that most states found

impossible to answer for their stakeholders. This approach

has been a constant for the past 12 years: data as an end

in and of itself is useless, but when used as a means to

empower decisionmaking and fuel improvement, data gets

results. The critical components of building an eective and

user-friendly data system depend on state policy leadership:

 P–20/workforce governance systems that ensure data

can follow the individual across systems, sectors, and

states

 policy and practices that build transparency and privacy,

security, and trust

 capacity building to ensure that those using data have

the ability and training to do so

DQC has created momentum over the past decade that

has spurred states to continuously use data. In the next

generation of work, states will continue to develop their

data infrastructures so that data can be shared more

eectively across systems, sectors, and even states. Moving

forward, DQC will continue to advocate that states build

the capacity and culture needed to ensure that data is

used in service of student learning. To do so, states must

focus on people. When students, parents, educators, and

policymakers have the right information to make decisions,

students excel. Now that more quality information is

available than ever before, DQC will emphasize individuals

as the key users and beneficiaries of data by focusing more

on storytelling and building public understanding of the

value of using education data in service of student learning.

Data Quality Campaign8

CHAPTER 3

Policy Leadership Made It Happen

The previous chapter discussed key actors and events driving education to become an evidence-based sector. This section

will take a closer look at that transformation, exploring what made it possible in a relatively short period to build a robust

data infrastructure in every state and begin developing an education culture that values evidence to make decisions. How

did the education sector make this tremendous progress? The short answer is that policy leaders at all levels made data a

priority.

Champions across the spectrum—including federal, state, and

local leaders and those working with policy leaders, such as

advocates, constituencies, and philanthropies—were actively

engaged in making this work a priority despite a long list of

competing demands. State leaders began to harness the power

of data in meeting education goals, and they increased state

eorts to eectively use data to improve teaching and learning.

In response, the federal government provided funding and

set policy to support, promote, and incentivize state action.

Advocates, with philanthropic investments, provided a national

forum to communicate and share ideas, created resources

and tools for policymakers, and advocated continuously for

investments in data.

None of these changes happened in isolation, but policymakers

took the lead by embracing a focus on improving outcomes for

students—which required comprehensive, high-quality data

to measure. This focus on outcomes took data beyond the

information technology (IT) department and into board room

conversations for the first time.

Leading policymakers began making the case that data

was, in fact, integral to policy success and should not be an

aerthought to policy decisions, which it had been before.

Reflecting on the strategies that policymakers used to

transform education into a data-driven sector reveals three

major conclusions: policies incentivize eective data use,

coordinated advocacy supports change, and money matters.

Policies Incentivize Eective Data Use

Policies (legislation or agency initiatives) can focus not only on

supporting data systems but also on creating the conditions

necessary to support data use. At the federal level, NCLB and

ESSA created the framework for data collection and moved

the education sector forward by requiring data to be used

for accountability. However, the Race to the Top competitive

grant program marked the first time that federal policy

called specifically for using data for continuous improvement

(e.g., delivering student growth data to teachers) rather

than for building systems and using data for accountability

(e.g., measuring annual yearly progress toward academic

proficiency) and reporting (e.g., high school graduation rate).

This critical innovation provided an opportunity for state

policymakers to think strategically about empowering people

with data, shiing the conversation from building data systems

to fostering data use.

Other federal programs contributed to the growth of data

infrastructure by acknowledging the importance of evidence.

For example, the Reading First program put evidence-based

methods of early reading instruction into classrooms. The

program provided states and districts support to apply

scientifically based reading research to teaching, including

proven instructional and assessment tools consistent with the

research.

“Longitudinal data is not just a K–12 issue; it requires gubernatorial commitment because all of our systems—from early childhood, to

K–12 education, to colleges and universities, to workforce development, to employment databases—must work together to make data

collection possible. And we need to do more to make the data useful because even the best data collection system is worthless if it

does not change what goes on in the classroom.”

—Ed Rendell, former governor of Pennsylvania, 2009

“It is our hope that states and districts will take a serious and thoughtful approach about how they can use this data to help

improve student learning.”

— Rep. George Miller (D-CA), then-chairman of the House Education and Labor Committee, on Race to the Top

From Hammer to Flashlight: A Decade of Data in Education 9

Spurred by new data capacity and a focus on continuous

improvement and evidence, states began to roll out innovative

uses of data driven not by IT departments but by policy

priorities:

 Arkansas provided one of the first direct student benefits by

using its state data system to determine student eligibility

for the Arkansas Challenge Scholarship.

 Colorado garnered widespread praise with its student

growth model, introducing it to the nation with a data

visualization tool that helped nonstatisticians understand

the value of a complex data model at a glance.

 Delaware required every school to have 90 minutes of

weekly collaborative planning time so teachers could have

data-informed conversations about how to best support

every student.

 Prompted by state legislation, Illinois redesigned the

Illinois School Report Card with new indicators, including

school characteristics, curriculum, student outcomes and

predictors, and school environment, as well as methods of

data display (e.g., comparisons to similar schools) to meet

the information needs of families and communities.

 Kentucky linked its K–12 and postsecondary data to

provide better information about how students fared in

postsecondary institutions.

State policymakers spurred progress by expecting data use to

be embedded directly into policy and practice at every level.

Coordinated Advocacy Supports Change

While policymakers worked to make data a policy priority, they

could not have done it alone. Policy leaders collaborated with

and benefitted from the support of philanthropy, constituency,

and advocacy organizations. Critical to this eort were both

broad-based constituency organizations (like the Council of

Chief State School Oicers, National Association of State Boards

of Education, National Conference of State Legislatures, and

NGA) and advocacy organizations (like the Alliance for Excellent

Education, The Education Trust, and Achieve). Funding from

foundations (like the Bill & Melinda Gates Foundation, the Eli

and Edythe Broad Foundation, and the Michael and Susan

Dell Foundation) helped these organizations set aside their

oen competing agendas to unite around the one issue they

could all agree on—high-quality data. To kick this eort o,

policymakers and thought leaders came together to create DQC

(see Appendix C for a list of DQC’s original managing partners).

Coordinated advocacy eorts supported state policymakers

in their eorts to incentivize eective data use by holding

state policymakers accountable for progress, providing

encouragement, highlighting success, identifying challenges

and opportunities, and convening policymakers to learn from

each other. Advocates amplified policy voices by creating a

shared vision, language, and forum for discussion. State and

federal policymakers relied on advocacy, constituency, and

philanthropic organizations as key resources in their eorts to

secure funding from all sources and to better understand how

to maximize these dollars.

DQC and its partner organizations produced evidence-based

roadmaps and tools, including the 10 Essential Elements of

Statewide Longitudinal Data Systems and the 10 State Actions

to Ensure Eective Data Use, to provide guidance and a shared

vision in the broader eort toward an evidence-based sector.

Critical to both the supply and demand for funding was the

eort to measure state progress and highlight best practices.

The sector’s ability to demonstrate success ensured sustainable

funding at all levels.

Advocates framed the need for data systems and use in terms

of policy benefits, allowing them to meet policymakers where

they were rather than trying to engage in a technical discussion.

For example, one of the thorniest issues policymakers faced

was linking data across systems, specifically linking K–12 with

postsecondary data. Approaching a state school chief with a

strategy to support the state’s goal of increasing postsecondary

enrollment is much easier than starting a conversation about

unique identifiers and interoperable systems. Advocacy eorts

demonstrated the value of linking K–12 and postsecondary data

by providing use cases, proof points, and stories of success, and

they built the political will to break down data silos between

K–12 and postsecondary. This eort helped policymakers at the

federal and state levels understand their individual roles and

responsibilities in each of these areas, and today, almost every

state has linked its K–12 and postsecondary data systems.

“ In Idaho, we now will have current, accurate data to make

better informed decisions at all levels and to give classroom

teachers the data they need to guide instruction every day.”

— Tom Luna, former Idaho superintendent of public instruction

(elected-R)

Data Quality Campaign10

Money Matters

System building and data use require investments to pay

for people, technology, training, and maintenance. Before

an evidence-based culture existed in the sector, states had

diiculty securing funding for data systems and data use for

purposes other than compliance reporting. The infusion of

federal dollars from the SLDS Grant Program was critical to

securing state policymakers’ interest, helping them move their

systems from emerging tools to robust sources of information.

States responded to this “seed funding” by increasing their own

investments to ensure long-term sustainability. While just a

handful of states were funding their systems in 2009 when the

bulk of the federal grant funds were distributed, 41 states were

funding their data systems by 2014.

As a part of the federal stimulus bill (ARRA), every state received

money from the State Fiscal Stabilization Fund program. While

no funds were allocated directly toward building data systems,

states were required to commit to building an SLDS and using

it to report a series of new indicators. Every state had to report

on the percentage of high school graduates enrolling in an

institution of higher education for the first time. More than 40

states are still reporting this indicator publicly five years aer

the requirement expired.

From Hammer to Flashlight: A Decade of Data in Education 11

CHAPTER 4

Challenges

Becoming a data-driven sector has not been an easy process, and the education field still has work ahead to fully reach its

goal of eective data use to improve outcomes for all students. Even with the leadership, policies, funding, and advocacy in

place, progress at times has been halting and beset by challenges. This section will explore those challenges, both to provide

insights to other public sectors and to caution the education field against repeating past mistakes.

Data Was Used As a Hammer Instead of a Flashlight

Policy, both federal and state, can be a double-edged sword.

Before there was demand for data at the local level, federal

involvement was a key policy driver, start-up funder, and

demand builder. However, NCLB, while a critical groundbreaker

for education to become a more data-driven sector, helped

perpetuate a static view of data as a tool for accountability

before people like teachers and school leaders really got value

from the data. Under NCLB, state data systems were initially

built to satisfy federal data reporting requirements. In turn,

both federal and state governments used data primarily as

a hammer for school accountability. At the state level, for

example, data was used to evaluate teachers based on student

performance, a requirement for states receiving federal Race

to the Top funds. Without state and local leadership pushing

to expand the role of education data beyond compliance,

embracing the power of data as a sector seemed impossible.

Teacher evaluations based on student test scores led teachers

to become wary of the very assessments that could potentially

help them gauge impact and pinpoint areas requiring

additional focus. When their livelihoods were at stake based

on data, the greatest potential champions of data were steered

into a position of opposition and defense.

People’s Needs Were Not at the Forefront

While leaders at the state and federal levels were working

to develop data systems with the support of advocacy

and philanthropy organizations, many stakeholders were

excluded from the conversation. Teachers were not asked

what data they needed—and in what format—to dierentiate

instruction, increase student achievement, and reflect on their

own practice. Teachers did not have access to data to help

them improve teaching and learning, yet it was being used to

evaluate their performance in the classroom.

Parents were also le out. States and the federal government

were not transparent about their eorts to build and use data

systems, including how students’ privacy was being protected

and students’ data kept safe. Families were not engaged with

tangible proof points, useful tools, or compelling use cases

to reinforce the value of data. Indeed, many parents still rely

on a single paper report card at the end of the school year to

understand their child’s progress in school. And parents in most

states are not provided tools or resources that put together

data over time to provide a rich history of their child’s learning,

information that could help them provide better support, make

better decisions, and be better advocates for their children.

Because of this information vacuum, many parents have

grown distrustful of data in education and concerned for their

children’s privacy.

People Were Not Provided the Conditions, Capacity, and Support to Use Data

While state and federal policies incentivized data use, the

focus was initially on building systems for compliance and

accountability. To truly achieve an evidence-based culture,

leaders need to understand the value of data beyond

compliance, and they need to have the skills to analyze and

act on data when they get it. Too oen school and district

administrators do not know how to properly use data to drive

results. Until recently, teachers, counselors, and others who

work with students did not have data tools that could help

them to do their jobs. Those who could benefit the most

from data use at the local levels have not been trained to

understand and use data to its fullest potential to support all

students. This lack of training has generated frustration and

misunderstanding among educators, which in turn has led to

fears of data misuse among parents and communities.

Data Quality Campaign12

CHAPTER 5

Lessons and Recommendations

Education’s progress toward becoming a data-driven sector was not the result of any one actor or action, but instead the

convergence of critical drivers. Based on what DQC has learned from the education field’s successes and challenges, the

following lessons and recommendations are a guide for other public sectors as well as the education sector. Leaders from

every public sector can benefit from these insights as they seek to build an evidenced-based culture to harness the power

of data to inform decisionmaking, meet individuals’ needs, and provide a measurable return on taxpayer investment. The

education sector can also benefit from reflecting on these lessons as it pivots from building data infrastructure to ensuring

that data is meeting people’s needs.

Leadership Matters

Data as a goal in and of itself does not inspire anyone (except

maybe self-proclaimed “data geeks”). But when leaders talk

about data as a crucial element for greater transparency,

empowered citizens, better decisionmaking, and improved

outcomes, it becomes central to everyone’s agenda.

National leaders made data use a priority, and now the

foundation has been laid for an education culture that values

and uses evidence to fuel improvement. When leaders prioritize

data use, people receive the support they need to spend the

time, resources, and energy required to make data work for

people.

Recommendations

✔ Encourage leaders to use their political capital to champion

data.

✔ Support leaders with tools, evidence, roadmaps, proof

points, and messaging points to make the case for the

eective use of data as a primary strategy to achieve policy

goals.

✔ Celebrate success stories to inspire other leaders and build

a movement.

✔ Reinforce the critical role of chief information oicers as

part of leadership teams in agencies to ensure that data

is not an end in itself but a source of information to meet

people’s needs.

It Is All about the People

Builders of data systems who once believed in the “Field of

Dreams” approach—“If you build it, they will come”—have

learned the hard way that is not true. People will not use

data that they do not find valuable. Creating a rich data

infrastructure must be continuously grounded in the service of

individuals who need information.

Eective data systems answer people’s questions. In education

it was hard, if not impossible, before data systems were

developed to answer questions about the success of high

school graduates, what education programs were the most

eective for which students, or what indicators could alert

teachers that a student was falling o track. Data kept in

silos cannot meet people’s needs. State agencies and other

responsible entities need to rethink their understanding of

data systems as solely an IT project and instead focus on the

needs of the people they are aiming to serve. Basing every data

conversation on people’s information needs also helps limit

data collection to only what is required and useful to answer

key questions.

Recommendations

✔ Build data systems to serve people’s information needs.

✔ Prioritize policies to create a culture that supports people

using data for improvement and build the conditions and

capacity to sustain that culture.

✔ Link key data across systems and sectors that serve

students, including early childhood, K–12, postsecondary,

workforce, and other sectors, like child welfare, to ensure

that data follows individuals and can be shared to support

students throughout their education journey.

From Hammer to Flashlight: A Decade of Data in Education 13

Earn Trust

People who need the data—in education, everyone from

parents and teachers to policymakers—have to understand

why specific data is being collected, who has access to it, and

how it is used and protected. This level of transparency and

understanding is critical to building trust. Stakeholders need

to be part of the development and constant review of policies

around data access, use, and protection. As the education

sector has learned from the implosion of inBloom, data is more

likely to be useful and used if those who need it have a say in

the information delivery process.

Recommendations

✔ Be transparent and communicate clearly and continually

about what data is collected, why it is beneficial to

individual users, and how it is protected.

✔ Ensure that everyone with a stake in the collection and

use of the sector’s data is engaged in the development

and constant refinement of data systems so that the

infrastructure is built to meet the needs of the end user.

✔ Provide the time and space for people to use data to

improve, rather than just examine data and put it on a shelf.

No One Can Do This Alone

Everyone has a role in making sure data is used in the service of

people. Ensuring that everyone with a stake in education saw

how quality information helped them meet their goals was a

critical part of creating an evidence-based sector. Policymakers,

philanthropy, and advocacy groups worked together to

ensure that infrastructure was built and stakeholder needs

were met. State and federal government played a central role

in providing financial support, enacting policies, and using

policymaker leadership to incentivize the building of state data

infrastructure across the country. However, without the critical

actions of advocacy and membership organizations (made

possible by philanthropic support), the creation of longitudinal

data systems in every state likely would have taken longer and

resulted in more compliance-oriented, lower quality, and less

widespread systems.

Coordinated advocacy made building and using data systems a

priority in policymaking, provided roadmaps, highlighted and

celebrated emerging best practices and success stories, and

responded to public concerns. These eorts ensured that data

was an integral part of the broader education agenda of the

past decade. A leading advocacy organization (as DQC has been

in the education sector) is instrumental in helping coordinate,

convene, share knowledge among, and continuously expand

the network of data champions.

Recommendations

✔ Build eective collaboration across government, advocates,

and constituency and membership groups to ensure that

everyone is pulling in the same direction to build and use

data systems to meet stakeholder needs.

✔ Create and support an advocacy organization working on

behalf of the field that is completely focused on building the

value proposition and public understanding of the need for

better data in the sector.

✔ Allocate adequate financial support to build and regularly

upgrade data infrastructure and improve people’s capacity

to use information to drive results. Both government and

philanthropy can provide funding to build this capacity.

Governance Is Critical

Developing high-quality data systems and processes is not a

one-time project. The work is never done, as systems need to

be continually maintained and updated and be flexible enough

to meet changing information needs. To make informed policy

decisions across agencies (e.g., state education agencies and

early childhood, higher education, and workforce agencies),

cross-agency data governance is needed. Data governance

is more than an IT issue. States can think broadly about data

governance as a base on which to build the relationships

and trust needed to securely share data across agencies to

answer questions such as, “How well do state higher education

institutions’ educational programs and capacity align with the

state’s workforce needs?” States can use a forward-looking data

governance body to lead proactive thinking about data, rather

than just reacting to compliance requirements.

Recommendations

✔ Establish a sustainable, multi-tiered cross-agency data

governance committee that establishes the vision and

mission of the cross-sector data governance work, sets

policy, and ensures that the policy and data work are

carried out.

✔ Empower the cross-agency data governance committee

and hold it responsible for developing and implementing

processes for data access, protection, and use.

Data Quality Campaign14

Build Data Literacy

Potential users of the information produced by these enhanced

data systems need the capacity and conditions to use data

eectively to make decisions. Thoughtful policies and practices

should be implemented to ensure that everyone expected

to use data is able to do so. This includes ensuring that

organizations and institutions on the ground (e.g., districts and

schools in the education sector) have the flexibility and people,

time, tools, money, and technology to use data to inform action

and improve outcomes. Making sure that every potential user

of data is “data literate” is critical to creating a culture that

values evidence. Data literacy is not just a buzzword; it can be

transformational to conversations, decisions, behaviors, and

actions to get results.

Recommendations

✔ Provide timely data and analyses in a format that people

can use to take action.

✔ Provide people the necessary training to use data

continuously, eectively, and ethically.

✔ Provide a forum for people to learn from each other, transfer

knowledge, and share best practices.

These lessons are useful to any sector or organization

interested in using data as a tool for improvement. The

education sector, for all of its progress over the past decade,

has only just begun its journey to develop a culture that values

and uses data. Building the infrastructure was the easy part.

The more diicult part remains—truly making data work for

students.

In moving from using data solely as a hammer—a tool of

compliance and accountability—to using data as a flashlight—a

tool to shine a light on what is working and what is not—the

education sector has learned something. While the hammer

can get the field moving, the flashlight is needed over the

long term. Mandates from the federal government pushed the

education field to start becoming an evidence-based sector, but

sustaining a culture of data use that works for people will take

more than federal mandates.

Eorts going forward must focus on demonstrating the value

of data and helping the people closest to students eectively

use data for continuous improvement. The flashlight will

provide transparency about how well the education system,

from policymakers to teachers, is serving students and how to

improve. This transparency, in turn, will increase demand for

data and push the sector closer to an evidence-based culture

that better helps students succeed.

Data in education must be used to create opportunities for all

students—ensuring that no child is lost on his or her education

journey. And in every public sector, the focus must always be on

using data to meet people’s needs. When people have the right

information to make decisions, everyone succeeds.

From Hammer to Flashlight: A Decade of Data in Education 15

Continuing to Build an Evidence-Based Culture in Education

The education sector is at a unique moment with a lot of

promise. The new Every Student Succeeds Act has shied

power back to states, which have made tremendous progress

over the past decade in building longitudinal data systems.

Today every state has the technical capacity to empower

people with information. High-quality data is more available

and transparent than ever before, but information is oen

hard to find, access, and understand. Now it is time for the

education sector to pivot from a focus on building data systems

to using data in service of students’ lifelong learning at all

levels and across sectors, from early childhood through K–12,

postsecondary, and the workforce.

The Data Quality Campaign collaborated with leaders from

across the education field to develop a set of recommendations

to help states enact policies that are critical to ensuring that

data is used to support student learning. The following Four

Policy Priorities to Make Data Work for Students build upon

the lessons learned that are detailed in this paper and will help

guide the education sector as it continues to become more

evidence based.

1. Measure What Matters

Be clear about what students must achieve and have the data

to ensure that all students are on track to succeed. Currently

data has a bad reputation and oen is not useful to educators

and families. Data systems were built within states and within

sectors, which makes it diicult to create data linkages and

allow data to follow individuals as they move from school

to school. This recommendation is not about collecting

more data. It is about meeting people’s needs. Aligning data

systems and indicators to critical policy and practice questions

makes data relevant and valuable to everyone with a stake

in education. Linking and governing data across all agencies

critical to student success, from early childhood and K–12

to postsecondary and the workforce, ensures that systems

are built to serve the individual and clarifies the roles and

responsibilities of everyone involved to institutionalize the

commitment to data quality and use.

2. Make Data Use Possible

Provide teachers and leaders the flexibility, training, and

support they need to answer their questions and take action.

Few teachers support the current uses of data because they

are rarely given the tools and training to make data work for

them and their students. Instead, data use is seen as a mandate

from administrators and policymakers, who themselves are

not supported in turning data into useful information to make

decisions. The path from data to evidence is complex and

requires sharing and linking data, using data to create evidence,

and using evidence to inform policymaking. Data is not useful

without strong analytics and research capacity. Leaders must

use the bully pulpit and allocate resources to prioritize using

data to inform decisionmaking at the state level.

3. Be Transparent and Earn Trust

Ensure that every community understands how its schools and

students are doing, why data is valuable, and how it is protected

and used. The existing culture of compliance in education has

stifled data use for transparency, support, and empowerment.

No one will use data if they do not trust it and find it useful.

Citizens must be empowered with quality information to act in

their communities and ensure that all students’ needs are met—

and to hold policymakers and public agencies accountable

for results. The public also deserves to know what data is

collected, how it is used to support students, and how it is

protected. Clear, steady communication about data will foster

public understanding and trust in the state as a good steward of

student information.

4. Guarantee Access and Protect Privacy

Provide teachers and parents timely information about their

students and make sure it is kept safe. Currently those closest

to students—especially parents—are not getting enough value

from the student data that is collected. Students will not be

successful unless the individuals closest to them have timely,

tailored access to information that answers their questions.

States must ensure that people who need access to data have

it—and that those with no business seeing confidential personal

information are kept away from it.

Data has the power to transform education into a personalized

enterprise that meets the needs of individuals and ensures that

no student is lost along the way. But for this transformation

to happen, the focus needs to pivot from collecting data to

prioritizing the eective use of data at all levels. The Four

Policy Priorities focus on people—meeting their information

needs, providing them the conditions to use data, providing

them greater transparency, and guaranteeing that they will

have access to data that is also kept safe. Without these pieces

in place, the power of data to support student learning and

the success of every student will never be realized, and the

education sector cannot become truly evidence based.

Data Quality Campaign16

APPENDIX A

Information-Gathering Process to

Inform the Development of This Paper

The contents of this paper (key events, analysis of progress, and

challenges and recommendations) were based on institutional

knowledge and informed by feedback and ideas from hundreds

of voices from both within and outside of the education

sector. The Data Quality Campaign (DQC) began this reflection

through its strategic planning process in 2015, during which the

organization collaborated with education leaders to produce a

vision for the field and Four Policy Priorities to Make Data Work

for Students. Specific to this data retrospective project, DQC

conducted additional outreach beginning in March 2016 and

collected information in the following ways. A list of participating

individuals can be found in Appendix E.

1. Working Meeting with Other Sectors: On March 15, 2016,

DQC hosted a small working group meeting with leaders

from sectors other than education such as health, housing,

and the workforce. The meeting helped shape this project

to ensure maximum value to multiple stakeholders. For

example, participants were asked what they would like

to know or better understand about the development of

education data infrastructure.

2. Conference Session with Education Researchers: On

April 7, 2016, DQC led a conference session of leading

education researchers at the third meeting of the National

Science Foundation Network on the Use of Administrative

Data for Education Research and Practice. Participants

discussed the progress made in education thus far and the

remaining challenges for transforming education into a

sector that truly values and uses evidence.

3. Survey of the Education Field: In June 2016, DQC

conducted an anonymous survey of approximately

190 people who have been critical players in the data

movement of the past decade. The survey, which had a 30

percent response rate, included both open- and closed-

ended questions about the changes, challenges, and

opportunities surrounding the development and use of

quality data in education.

4. One-on-One Interviews: In summer 2016, DQC conducted

one-on-one interviews with a select group of critical

players in the data movement to delve deeper into the

survey questions.

5. Data Conference Gathering: On July 13, 2016, DQC hosted

a gathering following the US Department of Education’s

2016 National Center for Education Statistics’ STATS-DC

Data Conference. Attendees at the conference included

data managers, chief information oicers, and policy and

thought leaders. Those who attended the reception were

given anonymous comment cards that included three of

the questions from the June survey about how they would

describe the current data culture and what challenges and

promising outcomes they foresee.

6. DQC Original Managing Partners Dinner: On July 18,

2016, DQC hosted a dinner for former managing partners

who founded DQC to reflect on the past decade of building

a data infrastructure. Attendees discussed the most

important drivers of that progress, what should have been

done dierently, and what should be done in the future to

leverage the progress to date.

7. Working Meeting with Key Education Leaders: On

October 3, 2016, DQC hosted a small working meeting

with a select group of education leaders and partners

representative of key voices in the field, most of whom

had not yet been involved in the project, to hear their

feedback on the initial analysis and dra of the paper and

suggestions for improving its messages and dissemination.

From Hammer to Flashlight: A Decade of Data in Education 17

APPENDIX B

Expanded History of Building an

Education Data Infrastructure

1980 to 2004

1980s–1990s: States Take the Lead and Focus

on Student Outcomes

Following the 1983 release of A Nation at Risk: The Imperative

for Educational Reform and the 1989 convening of the nation’s

governors for an education summit in Charlottesville, VA,

states demonstrated a remarkable shi in their approach to

education. Policies had traditionally been the sole purview

of local oicials, but governors and legislators began to set

policy focusing on improving student outcomes and rejected

the notion that their role should be limited to compliance

and focused on inputs. For example, the Kentucky Education

Reform Act of 1990 (KERA, House Bill 940) expanded the state

role in education by mandating financial, curricular, and

governance reforms, and the Education Reform Act of 1993 in

Massachusetts created state aid for schools, established higher

standards, and required more accountability from all levels of

education.

Southern states were brought together by the Southern

Regional Education Board (SREB) to develop policies and

share best practices as they sought to increase their standing

among peers to attract employers to their states. SREB was

created by regional state policymakers and chaired during

this pivotal time by governors who were prioritizing data use.

Aided by new technologies, these states in particular homed

in on an emerging practice in the private and health care

sectors—the use of data to rapidly improve performance. Texas

Governor George W. Bush was an early pioneer of this strategy

(by 2005 Texas had 9 of the Data Quality Campaign’s [DQC] 10

Essential Elements in place), which would become a hallmark

of his presidency. He prioritized the investment of dollars into

collecting student-level data and linking it longitudinally to

better understand if his education policies were having the

desired impact on outcomes. Texas was joined by states such

as Tennessee, Louisiana, Georgia, and Florida, which all led the

nation by implementing 8 or more of the 10 Essential Elements.

Education policymakers from around the nation, as well as

those sitting in Washington, DC, took note of this impressive

innovation.

2001: Congress Requires States to Use Data

for Accountability

The No Child Le Behind Act of 2001 (NCLB) required states

to annually test students nationwide, disaggregate data by

subgroups, and publicly report the results through state report

cards. With a federal framework that attached high-stakes

decisions to data, states were highly motivated to ensure that

they had high-quality data. However, at the time, districts

used varying data definitions and dierent numerators/

denominators in their calculations. If schools were going to be

compared and judged, states could not rely solely on district-

reported aggregate data. As a result, leaders turned toward

student-level data collections as a solution. By the time the

accountability provisions took eect a few years later, states

were prepared to leverage these new collections to meet the

reporting requirements of the law.

2003: The US Department of Education Looks

to Data to Manage for Results

With the passage of NCLB came the need for the US Department

of Education (USED) to better manage and use the data it

was required to collect from states—and to move away from

gathering state data using paper forms that would sit in

drawers. The Performance Based Data Management Initiative

(PBDMI) was a large-scale eort within USED to leverage

technology and combine more than a dozen separate data

collections into a single system. PBDMI established a process

for states to electronically submit elementary and secondary

education data from the state, district, and school levels to

USED. The goal of PBDMI was to improve the use of data by

USED and focus the information it requested from states by

eliminating duplication, conflicting definitions, and information

that was not useful for the evaluation of its programs. The

initiative was also a large-scale undertaking for state education

agencies, which volunteered to help develop uniform data

and test the new data collection system. PBDMI evolved into

EDFacts, a USED initiative to put performance data at the

center of policy, management, and budget decisions for all

K–12 educational programs. EDFacts centralizes aggregate K–12

performance data supplied by states with other data, such

as financial grant information, within USED to enable better

analysis and use in policymaking.

Data Quality Campaign18

2005 to 2008

2005: Governors Sign the National Governors

Association Graduation Rate Compact

In 2005, all 50 state governors signed the National Governors

Association (NGA) Graduation Rate Compact, agreeing to

implement a common formula for calculating high school

graduation rates in their states. This agreement was significant

in that it was the first time the nation’s leaders agreed on a

comparable statistic across states. The key commitments of

the compact included improving data capacity and reporting

annual progress. At the time of signing, 34 states were

collecting outcome data at the student level, but just 14 had

the necessary infrastructure in place to actually calculate the

rate as promised that day.

2005: Congress and USED Support States

with Grant Funding

That same year, the Institute of Education Sciences, the

statistics, research, and evaluation entity for USED, awarded

grants to 14 states to build or improve their statewide

longitudinal data system (SLDS). This congressionally created

grant program encouraged states to generate and use “accurate

and timely data to meet reporting requirements; support

decision-making at State, district, school, and classroom levels;

and facilitate research needed to eliminate achievement gaps

and improve learning of all students.” The federal government

did not invent state data systems, but it did support and

incentivize their development by following the lead of

states that had demonstrated implementation success (e.g.,

Florida, Texas, Tennessee) and a desire to build (e.g., the NGA

Graduation Rate Compact). As of September 2016, 47 states,

the District of Columbia, Puerto Rico, the US Virgin Islands, and

American Samoa had successfully secured at least one grant

from the SLDS Grant Program, totaling more than $500 million.

2005: Philanthropic Organizations Provide

Investments to Prioritize and Coordinate

Advocacy Eorts

When the SLDS Grant Program began, “data-driven

decisionmaking” was still technical jargon to most educators

and policymakers, and little consensus existed as to what

it would look like in education. To address this issue, the

philanthropic community (e.g., the Bill & Melinda Gates

Foundation, the Eli and Edythe Broad Foundation, and the

Michael and Susan Dell Foundation) invested in education

policy and advocacy organizations (e.g., Council of Chief State

School Oicers, DQC, The Education Trust, and American

Federation of Teachers) to prioritize data and support evidence-

based decisionmaking at all levels. These investments served

to expedite and fuel states’ progress as they built their data

infrastructures and began the shi from compliance to service.

2005: DQC Launches to Change the Data

Conversation from Hammer to Flashlight

DQC was launched in 2005 by 14 national partners (see

Appendix C) that formed a coalition of advocacy and

membership-based organizations to advocate for policies in

support of eective data use (for more information, see page

6). Rejecting the idea that data was simply for compliance and

was, therefore, an information technology (IT) project, DQC

chose to focus the conversation around 28 crucial education

policy questions that state policymakers could not answer

because they had not built the systems to answer them. To

that end, DQC identified the 10 Essential Elements of Statewide

Longitudinal Data Systems (for more information, see Appendix

D), highlighted promising practices in implementation,

convened partners and states to support advocacy eorts,

and measured (and celebrated) state progress. The

importance of these organizations coming together around

a shared vision cannot be overstated. These partners oen

represented conflicting policy agendas, but all believed that

their constituencies and networks needed better information

regardless of policy positions.

2006: The National Center for Analysis of

Longitudinal Data in Education Research Puts

Longitudinal Data to Work

The National Center for Analysis of Longitudinal Data in

Education Research (CALDER) is one of the National Research

and Development Centers funded by the federal government

and private foundations. CALDER is a joint eort of American

Institutes for Research and scholars at Duke University,

Stanford University, the University of Florida, the University

of Missouri, the University of Texas at Dallas, the University

of Virginia, and the University of Washington. In partnership

with states, CALDER uses individual-level longitudinal student

and teacher data to examine the eects of real policies and

practices on the learninggains ofstudents in a district or

stateover a number of years. CALDER pays particular attention

to how outcomes dier for dierent subgroups of students.

CALDER was among the first organizations to use longitudinal

data sets to conduct research and demonstrate the value of

data systems in education.

2008: USED Regulations Clarify the Value of

SLDS

US Secretary of Education Margaret Spellings announced

regulations in 2008 focused on strengthening and clarifying

certain provisions of NCLB. One regulation required states to

submit a longitudinal statistic for the first time ever—the four-

year adjusted cohort graduation rate. That same year, based

on requests for clarification by the states, USED also issued

From Hammer to Flashlight: A Decade of Data in Education 19

regulations on the Family Educational Rights and Privacy

Act (FERPA). Among other things, the regulations specifically

clarified that states could share data on behalf of school

districts without violating the prohibition against redisclosure.

With this clarification and a more powerful statistic in their

toolbox, states were poised to deliver real value (e.g., tools,

research, and analytics) to schools on the data they were

already collecting.

2009 to 2012

2009: DQC Changes the Conversation

from Systems to Policies with a New Set of

Recommendations

DQC released the 10 State Actions to Ensure Eective

Data Use to provide a set of clear, measurable steps for

state policymakers to take as they began to use their new

longitudinal data systems. DQC’s 2009 report, The Next Step:

Using Longitudinal Data Systems to Improve Student Success,

outlined three bold imperatives that states must embrace if

they were going to successfully move from data for compliance

to data for action. Specifically, states were encouraged to

prioritize linking their K–12 systems to early education,

postsecondary education, and the workforce, along with

other state social service systems, to create richer pictures of

student pathways and success, provide access to appropriate

individuals and the public, and build the capacity of those

using data to use it well.

2009: The American Recovery and

Reinvestment Act Provides Unprecedented

Funding to Build and Use SLDS

The American Recovery and Reinvestment Act of 2009 (ARRA)

established three separate funding mechanisms for states

to use in their eorts to build and use their SLDS. In addition

to the incredible funding opportunity, ARRA’s emphasis on

data systems and use brought to the conversation many new

stakeholders and audiences who had, to date, largely been

absent.

 State Fiscal Stabilization Fund: USED awarded $48.6

billion to states that committed to specific education

reforms, including the implementation of SLDS. While none

of the funding was to be used to build data systems, the

mere inclusion of the requirement was a strong signal of the

importance of data use as a strategy for improving student

outcomes.

 SLDS Grant Program: Twenty states (Arkansas, Colorado,

Florida, Illinois, Kansas, Maine, Massachusetts, Michigan,

Minnesota, Mississippi, New York, Ohio, Oregon,

Pennsylvania, South Carolina, Texas, Utah, Virginia,

Washington, and Wisconsin) won grants totaling $250

million and ranging from $5.1 million to $19.7 million each

to build their SLDS with the 12 Required Elements of a P–16

Education Data System

listed in the America COMPETES

Act (ACA). (The 12 ACA elements align closely with DQC’s 10

Essential Elements. For more information about DQC’s 10

Essential Elements, see Appendix D.) This round of grants

marked the first time the program required these elements

to secure funding and brought the total number of states

with SLDS grants to 41. (Forty-one states and the District of

Columbia had received at least one SLDS grant since 2005.)

 Race to the Top: The administration launched an ambitious

competitive grant program to incentivize states to tackle

complex education challenges that rarely find their way to

the top of state priorities. Part of this new program asked

states to think dierently about how they would leverage

their SLDS in support of teaching and learning. The program

also served to elevate the conversation about data systems

and use to governors and state legislatures; these systems

were no longer simply the purview of state education

agency oicials. Aer three rounds, grants ranging from $17

million to $700 million had been awarded to 18 states and

the District of Columbia in support of eective data use.

ARRA’s provisions helped reinforce that data systems and use

are critical for states and are far from just an IT project.

2009: The Call for Common Education Data

Standards Spreads

As state education agencies continued building SLDS,

states, national organizations, and federal oices began to

call for common education data standards to help states

improve data quality. Supported by the National Center for

Education Statistics, the Common Education Data Standards

(CEDS) project is a national collaborative eort to develop

voluntary, common data standards for a key set of education

data elements to streamline the exchange, comparison,

and understanding of data within and across early learning,

K–12, postsecondary, and workforce (P–20W) institutions and

sectors. Versions 1 through 5 of the standards were developed

by a combination of a CEDS stakeholder group (including

representatives from across the P–20W field) and open

meetings and conversations. Starting with Version 6, CEDS is

developed and maintained by an open community that allows

anyone to participate. According to DQC’s Data for Action

2014 state survey, only six states reported not implementing

CEDS. Other states are at dierent levels of implementation,

from making a formal decision to adopt CEDS to operationally

sharing data using CEDS.

Data Quality Campaign20

2013 to 2016

The incredible progress of the prior decade brought increasing

numbers of new stakeholders to the conversation, including

those voicing concerns that transparency had been lost in the

name of progress.

The general public was largely absent from a data conversation

that had been mostly taking place in state capitols. This

situation changed overnight in winter 2013 when inBloom

launched, touting its ability to aggregate and make available

vast amounts of student data to inform tools and dashboards

for educators. The arrival of inBloom exposed the lack of

attention policy leaders, advocates, and educators had paid to

ensuring that the public, and parents in particular, understood

the value of data and the means by which it was collected,

used, and kept private. The absence of basic facts about

data collection, use, and protection from the webpages of

prominent champions of education data use—and states and

districts themselves—was a glaring oversight that gave rise to

myths and fed into the rising concerns that privacy had been an

aerthought in the desire to use data. Aer more than a year of

negative publicity and successful moves by grassroots activists

to force school districts and states to abandon using inBloom,

the company closed its doors in June 2014 due to a lack of

customers, but its eect on the education landscape was just

beginning to be understood.

The growing chorus of opposition to data use practices was

heard in state legislatures, and in 2014, 36 states introduced

110 bills resulting in 27 new laws addressing student data

privacy; the prior year saw just one state action in this area.

Legislators primarily focused on limiting the scope of district

and state data collection and use in 2014. However, in 2015

privacy concerns evolved to include third-party data sharing,

particularly with online service providers (47 states introduced

188 bills resulting in 28 new laws addressing student data

privacy). By the 2016 legislative session the number of new laws

passed decreased (112 bills resulted in 19 new laws in 15 states)

as states began to implement the privacy protections passed

in recent years and began to grapple with implementing the

new federal K–12 education law, the Every Student Succeeds

Act (ESSA). At the conclusion of three years of legislating, the

legal landscape around student data privacy had significantly

changed with all states but Vermont introducing at least one

bill and 36 states putting new laws in place.

Frustrated by an outdated 40-year-old federal privacy law

(FERPA), members of Congress responded to student privacy

concerns by holding hearings and introducing several new

bills in 2015. The most significant attempts to legislate were

bipartisan and far reaching in scope. Senators Edward Markey

(D-MA) and Orrin Hatch (R-UT) were first (in 2014) with a bill

to amend FERPA, followed quickly by Representatives Luke

Messer (R-IN) and Jared Polis (D-CO) introducing the Student

Digital Privacy and Parental Rights Act (SDPPRA). While the

fate of SDPPRA is unknown (as of this publication date), the bill

was supported by a broad coalition of organizations. In public

statements, supporters applauded legislative provisions that

would protect student privacy while ensuring that educators

and families could still use data and education technology

to improve outcomes. On the importance of this balance,

Representative Polis stated: “Our bipartisan bill is a much-

needed first step in providing a framework that can address

these concerns of parents and educators while at the same time

allowing for the promise of education technology to transform

our schools.” At the same time, Senators Richard Blumenthal

(D-CT) and Steve Daines (R-MT) introduced the SAFE KIDS Act

with similar provisions to the bill introduced by Representatives

Polis and Messer, and Representatives Marcia Fudge (D-OH)

and Todd Rokita (R-IN) introduced an amendment to FERPA

to essentially rewrite the law to reflect the data realities of the

21st century. To date, none of these attempts have resulted in

changes to federal law. There were some attempts to include

privacy provisions in ESSA; however, all but one (the inclusion

of privacy training as an allowable use of Title II funding) failed

in conference.

2015: Congress Reauthorizes NCLB and

Shines a More Powerful Light on Data Use

ESSA preserves what was widely reported as the biggest

success of its predecessor, NCLB—the collection and public

reporting of disaggregated student performance data. In fact,

Congress went further by adding new levels of disaggregation

(e.g., foster and military youth), new types of information

to be collected and reported (e.g., chronic absence and

school funding), and new indicators of success including

postsecondary enrollment statistics. ESSA’s ultimate impact

on data use is unclear; however, the new law makes a strong

statement that data use is here to stay in education.

From Hammer to Flashlight: A Decade of Data in Education 21

APPENDIX C

Original DQC Managing and Endorsing

Partners

The Data Quality Campaign (DQC) was launched in 2005 by

14 advocacy and constituency organizations that recognized

the need for a national, collaborative eort to encourage and

support the use of high-quality, accessible data in education.

With the support of their funders, these founding partners

put aside their sometimes conflicting policy agendas to align

around the priority of increasing the availability and use of data

in education. To ensure that the eort was truly collaborative,

DQC was not started as a separate nonprofit but rather was

housed at the National Center for Educational Achievement

and managed and run by the partner organizations, all of which

are listed below.

Managing Partners

 Achieve

 Alliance for Excellent Education

 Council of Chief State School Oicers

 Education Commission of the States

 The Education Trust

 National Association of State Boards of Education

 National Association of System Heads

 National Center for Educational Achievement

 National Center for Higher Education Management Systems

 National Conference of State Legislatures

 National Governors Association Center for Best Practices

 Schools Interoperability Framework Association

 State Educational Technology Directors Association

 State Higher Education Executive Oicers

Data Quality Campaign22

APPENDIX D

DQC’s Policy Recommendations for

States from 2005 to Today

28 Policy Questions

Central to the 2005 launch of the Data Quality Campaign (DQC)

was highlighting a list of 28 policy questions that could not

be answered without a statewide longitudinal data system

(SLDS) consisting of 10 Essential Elements (see page 24 for the

10 Essential Elements). These questions framed the need for

SLDS by showing state policymakers that they were not able to

answer the priority questions that they had identified as being

most important to meeting their education goals. Investing in

high-quality longitudinal data helps policymakers, educators,

researchers, community leaders, and others have the data to

answer critical questions, inform pressing policy discussions,

and make practical decisions to support student success.

TOPIC QUESTIONS

Predicting Success in

Later Grade Levels

1. What is the impact of preschool on later academic achievement (e.g., third-grade test results)?

2. Do the eects of our early interventions “fade out” later?

3. Are students academically prepared for high school?

4. Which elementary and middle schools in the state are consistently highest performing in preparing

dierent student populations for high school?

5. Which elementary and middle schools produce the strongest academic growth among initially

poorly prepared students and among initially well-prepared students?

Academic Growth 6. How many students are achieving at least one year’s academic growth every year?

7. How many of the students who started out below grade level are achieving more than a year’s

growth?

Achievement Levels

in Early Grades As

Indicators of Later

Success

8. What achievement levels in grades three though seven indicate that a student is “on track” for later

success?

Impact of Grade-Level

Retention

9. What eect does early grade retention have on the later academic success of students who were

retained in the early grades?

Course Rigor 10. What eighth-grade achievement levels indicate that a student is well prepared to succeed in

challenging courses in high school?

11. Have students taken the coursework to prepare them for college and work—both in years of study

and rigor of content?

12. What evidence exists that students who take and pass the courses have learned the course content?

Sustaining Enrollment

in Early Grades

13. What students are being lost in transition between middle and high school?

14. What proportion of the students who enter elementary school maintain continuous enrollment and

complete eighth grade in a timely manner?

Consistently High-

Performing Schools

15. Which elementary and middle schools in the state are consistently highest performing in preparing

dierent student populations for high school?

College Preparation 16. Are students academically prepared to graduate from high school and enter college?

From Hammer to Flashlight: A Decade of Data in Education 23

TOPIC QUESTIONS

High School

Indicators of College

Preparedness

17. What high school achievement levels indicate that a student is college and work ready?

18. Are students academically prepared to enter college and complete their program or degree in a

timely manner?

19. What is the relationship between students’ performance on state assessments (high school exit

exam, end-of-course exams) and subsequent postsecondary performance and graduation?

College Remediation 20. What percentage of high school graduates who go on to college take remedial courses?

High School

Completion Rates

21. What proportions of the students who enter ninth grade maintain continuous enrollment and

complete their high school requirements in a timely manner?

High-Performing

Schools: College

Preparation of

Subgroups

22. Which high schools in the state are consistently highest performing in preparing dierent student

populations for college and work?

Academic Growth by

Prior Performance

Subgroup

23. Which high schools produce the strongest academic success among initially poorly prepared

students and among initially well-prepared students?

College Success of

K–12 Students

24. In what content areas do students require remediation?

25. What are the retention and degree completion rates of students who are placed in remedial

coursework?

Dual Enrollment 26. How do dual enrollment and Advanced Placement programs in high school aect students’ success

in college?

Graduation Rates by

Subgroup and Prior

Performance

27. Which institutions are doing the best job of graduating students on time, based on those students’

prior preparation and level of economic disadvantage?

Teacher Eectiveness

and Preparation

Programs

28. Which teacher preparation programs produce the graduates whose students have the strongest

academic growth?

Data Quality Campaign24

10 Essential Elements of Statewide Longitudinal Data Systems

In 2005 DQC identified the 10 Essential Elements of Statewide

Longitudinal Data Systems. The 10 Essential Elements, listed

below, provided a roadmap for states as they built statewide

longitudinal data systems to collect, store, and use longitudinal

data to improve student achievement.

1. A unique student identifier. A single, unduplicated

number assigned to an individual student that remains

with that student from kindergarten through high school

and connects student data across key databases across

years.

2. Student-level enrollment, demographic, and program

participation informationincluding information such

as attendance, special education status, gied and

talented education status, career and technical education

participation, and free or reduced-price lunch status.

3. The ability to match individual students’ test records

from year to year to measure academic growth and the

ability to disaggregate the results by individual test item

and objective.

4. Information on untested studentsand the reasons why

they were not tested.

5. A teacher identifier systemwith the ability tomatch

teachers to studentsby classroom and subject.

6. Student-level transcript data, including information on

courses completed and grades earnedfrom middle and

high school.

7. Student-level college readiness test scoressuch as

scores on SAT, SAT II, ACT, Advanced Placement, and

International Baccalaureate exams.

8. Student-level graduation and dropout data.

9. The ability to match student records between the P–12

and postsecondary systems.

10. A state data audit system assessing data quality,

validity, and reliability.

To read the report creating the 10 Essential Elements of Statewide Longitudinal Data Systems, visit http://dataqualitycampaign.org/

resource/creating-a-longitudinal-data-system/.

2–3 0–1

4–5

6–7

8–9 10

2–3 Elements

0–1 Element

4–5 Elements

6–7 Elements

8–9 Elements

10 Elements

2–3 Elements

0–1 Element

4–5 Elements

6–7 Elements

8–9 Elements

10 Elements

2005 2011

From Hammer to Flashlight: A Decade of Data in Education 25

10 State Actions to Ensure Eective Data Use

In 2009 DQC released its 10 State Actions to Ensure Eective

Data Use. The 10 Actions called for states to move from only

collecting data for compliance and accountability purposes

to using data to answer critical policy questions, inform

continuous improvement, and ultimately support students on

their paths to success.

1. Link state K–12 data systems with early learning,

postsecondary, workforce, and other critical state agency

data systems.

2. Create stable,sustainedsupportfor longitudinal data

systems.

3. Developgovernancestructures to guide data collection

and use.

4. Build state data repositories.

5. Provide timely,role-basedaccessto data.

6. Create progressreportswith student-level data for

educators, students, and parents.

7. Createreportswith longitudinal statistics to guide system-

level change.

8. Develop a purposeful research agenda.

9. Implementpoliciesand promotepracticesto build

educators’ capacity to use data.

10. Promote strategies to raise awareness of available data.

Note: The maps do not show results from 2009, the first year DQC surveyed states on the 10 Actions, because the criteria for one Action

were changed for the 2011 survey. Therefore, results from 2009 and 2014 cannot be compared.

To read the report creating the 10 State Actions to Ensure Eective Data Use, visit http://dataqualitycampaign.org/resource/next-

step-using-longitudinal-data-systems-improve-student-success/.

1–3 Actions

0 Actions

4–5 Actions

6–7 Actions

8–9 Actions

10 Actions

1–3 Actions

0 Actions

4–5 Actions

6–7 Actions

8–9 Actions

10 Actions

Did not participate

1–3 Actions

0 Actions

4–5 Actions

6–7 Actions

8–9 Actions

10 Actions

2011 2014

Data Quality Campaign26

Four Policy Priorities to Make Data Work for Students

In 2016 DQC partnered with leaders from across the education

field to develop Four Policy Priorities to Make Data Work for

Students. These policy priorities are a set of recommendations

that build on the foundation of DQC’s 10 Essential Elements

and 10 State Actions, evolving further to reflect a changing

focus at DQC—and in states and classrooms—from systems to

people.

 Measure What Matters. Be clear about what students must

achieve and have the data to ensure that all students are on

track to succeed.

 Make Data Use Possible. Provide teachers and leaders the

flexibility, training, and support they need to answer their

questions and take action.

 Be Transparent and Earn Trust. Ensure that every

community understands how its schools and students are

doing, why data is valuable, and how it is protected and

used.

 Guarantee Access and Protect Privacy. Provide teachers

and parents timely information on their students and make

sure it is kept safe.

To read the full report outlining the Four Policy Priorities to Make Data Work for Students, visit http://dataqualitycampaign.org/

resource/time-to-act/. In addition to this report, DQC also released District and Federal Actions to Make Data Work for Students.

SCHOOL

People—like parents and teachers—need tailored

information that they can trust to ensure all

students’ individual needs are met. A culture

of effective data use means putting

students at the center.

DATA IN SERVICE

OF LEARNING

MEASURE WHAT MATTERS

MAKE DATA USE POSSIBLE

BE TRANSPARENT

AND EARN TRUST

GUARANTEE ACCESS

AND PROTECT PRIV

ACY

MEASURE

HAT MATTERS

AKE DATA USE POSSIBLE

BE TRANSPARENT

AND EARN TRUST

D PROTECT PRIVAC

From Hammer to Flashlight: A Decade of Data in Education 27

APPENDIX E

Acknowledgments

The Data Quality Campaign (DQC) thanks the Laura & John Arnold Foundation for funding and supporting this project.

DQC also thanks the following people and organizations that provided expert advice and guidance during the research process and

writing of this paper. See Appendix A for more details on these information-gathering sessions.

Working Meeting with Other Sectors

March 15, 2016, Washington, DC

 Amanda Cash, US Department of Health and Human

Services

 Emmalie Dropkin, National Head Start Association

 Robin Ghertner, US Department of Health and Human

Services

 Cynthia A. Guy, PhD, Annie E. Casey Foundation

 Natalie Evans Harris, Oice of Science and Technology

Policy

 Brian Harris-Kojetin, National Academies of Sciences,

Engineering, and Medicine

 Carter Hewgley, Johns Hopkins University

 Matthew Hill, University of Pennsylvania

 Sallie Keller, Biocomplexity Institute of Virginia Tech

 Therese Leung, MDRC

 Shelley Metzenbaum, Volcker Alliance

 Paul Perez, IBM

 Erika Poethig, The Urban Institute

 Rachel Rosen, MDRC

 George Schoeel, National Academies of Sciences,

Engineering, and Medicine

 Robert Shea, Grant Thornton LLP

 Stephanie Shipp, Biocomplexity Institute of Virginia Tech

 Jack Smalligan, Oice of Management and Budget

 Lauren Supplee, Administration for Children and Families

 Jennifer Thornton, The Pew Charitable Trusts

 Anna White, National Head Start Association

 Rachel Zinn, Workforce Data Quality Campaign

Conference Session with Education Researchers

April 7, 2016, National Science Foundation Network on the Use of Administrative Data for

Education Research and Practice, Washington, DC

 Organizers: Kenneth Dodge, Duke University, and

David Figlio, Institute for Policy Research, Northwestern

University

 Atila Abdulkadiroglu, Duke University

 Emma Adam, Northwestern University

 Elaine Allensworth, University of Chicago

 Peter Arcidiacono, Duke University

 Tony Bennett, Strategos

 Thomas Brock, US Department of Education

 Carycruz Bueno, Georgia State University

 Jennifer Carinci, Johns Hopkins University

 Chelsea Clinton, Oregon Department of Education

 Carrie Conaway, Massachusetts Department of Elementary

and Secondary Education

 Thurston Domina, The University of North Carolina at

Chapel Hill

 Greg Duncan, University of California, Irvine

 Susan Dynarski, University of Michigan

 KC Elander, North Carolina Department of Public

Instruction

 Emerson Elliott, Council for the Accreditation of Educator

Preparation

Data Quality Campaign28

 Sebastian Gallegos, University of Chicago

 Neal Gibson, Arkansas Research Center

 Christina Gibson-Davis, Duke University

 Dan Goldhaber, American Institutes for Research

 Edith Gummer, Ewing Marion Kauman Foundation

 Jonathan Guryan, Northwestern University

 Jane Hannaway, National Center for Analysis of

Longitudinal Data in Education Research, American

Institutes for Research

 Rick Hanushek, Stanford University

 Tracy Hunt-White, US Department of Education

 Clement Jackson, Northwestern University

 Ariel Kalil, University of Chicago

 Venessa Keesler, Michigan Department of Education

 Helen Ladd, Duke University

 Joy Lesnick, US Department of Education

 Susanna Loeb, Stanford University

 Hugh Macartney, Duke University

 Joseph Martineau, National Center for the Improvement of

Educational Assessment

 Joel McFarland, US Department of Education

 Clara Muschkin, Duke University

 Chelsea Owens, US Department of Education

 Andrew Penner, University of California, Irvine

 Emily Penner, Stanford University

 Karl Pond, North Carolina Department of Public Instruction

 Sam Rauschenberg, Georgia Governor’s Oice of Student

Achievement

 Margaret Raymond, Center for Research on Education

Outcomes at Stanford University

 Allen Ruby, US Department of Education

 Janelle Sands, US Department of Education

 Ross Santy, US Department of Education

 Tim Sass, Georgia State University

 Diane Schanzenbach, Northwestern University

 Elizabeth Setren, Massachusetts Institute of Technology

 Nancy Sharkey, US Department of Education

 John Singleton, Duke University

 Tony Smith, Illinois Superintendent of Education

 Lucy Sorenson, Duke University

 Kathy Stack, Laura & John Arnold Foundation

 Russ Whitehurst, Brookings Institution

 Governor Bob Wise, Alliance for Excellent Education

DQC Original Managing Partners Dinner

July 18, 2016, Washington, DC

 Matt Gandal, Education Strategy Group

 Daria Hall, The Education Trust

 Richard Laine, formerly National Governors Association

 Scott Palmer, Education Counsel

 Ryan Reyna, Education Strategy Group

 Stefanie Sanford, College Board

 Ross Santy, US Department of Education

 Amy Starzynski, Foresight Law & Policy

 David Wakelyn, Union Square Learning

 Maureen Wentworth, Council of Chief State School Oicers

From Hammer to Flashlight: A Decade of Data in Education 29

Working Meeting with Key Education Leaders

October 3, 2016, Washington, DC

 John Bailey, consultant, formerly Foundation for Excellence

in Education

 Jennifer Engle, Bill & Melinda Gates Foundation

 Kathy Gosa, SLDS State Support Team, National Center for

Education Statistics

 Jane Hannaway, National Center for Analysis of

Longitudinal Data in Education Research, American

Institutes for Research

 Bethany Little, Education Counsel

 Michael J. Petrilli, Thomas B. Fordham Institute

 Kathy Stack, Laura & John Arnold Foundation

 Bob Swiggum, Georgia Department of Education

 Bill Tucker, Bill & Melinda Gates Foundation

 Jerey Wayman, Wayman Services, LLC

 Martin West, Harvard Graduate School of Education,

Education Policy and Governance

 Rachel Zinn, Workforce Data Quality Campaign

DQC also thanks the many individuals from states and organizations willing to take the time to complete the survey, participate in a

one-on-one interview, or attend the data conference gathering.

DQC thanks our funders; in particular we thank the Laura & John Arnold Foundation for their support in the yearlong eort to

develop this report. Our work is made possible by philanthropic grants and contributions from the Annie E. Casey Foundation,

Bill & Melinda Gates Foundation, Carnegie Corporation of New York, Charles Stewart Mott Foundation, Ewing Marion Kauman

Foundation, ExxonMobil, Laura & John Arnold Foundation, Michael and Susan Dell Foundation, and the Walton Family Foundation.