Technology Counts - March 14, 2013 - 26

Data Development

Data-Sharing

GOALS

DRIVES CHANGE
Districts need to build better information intelligence

By_Katie Ash

he fragmented nature of data
systems in school districts, a
lack of common data standards
across states, and the financial challenges of providing
professional development to data users
in schools combine to leave many districts
and states struggling to provide meaningful, real-time data about student performance to educators.
And that reality, experts say, is a major barrier for districts working to transform themselves into organizations that maximize the
effectiveness of new technologies.
“It’s hard to get machines to talk to one another,” says Darrell West, the vice president
and director of governance studies and the
founding director of the Center for Technology Innovation at the Brookings Institution,
in Washington.
“A lot of school data are siloed. You may
have academic-performance data in one
place, administrative data someplace else,
and disciplinary data somewhere else,” he
says.
Complying with privacy laws around student data, such as the Family Educational
Rights and Privacy Act, or ferpa, also presents challenges, he says.
While protecting student information
and providing educators with meaningful,
timely data are important goals, “right now
the balance is skewed very much in favor of
privacy over data-sharing, so we’re not able
to get the benefits that would come from integrating information,” he says.

STEPS TO SHARING
Kathleen Berry, the coordinator of research, evaluation, and assessment for the
Monroe County Intermediate School District
in Michigan—which provides special education and professional development services
to nine districts, two charter schools, and 15
private schools in the county—says her state
is a prime example of how hard it can be to
share data between districts. That difficulty
impedes comparisons of instructional techniques and keeps teachers from accessing records for students who have transferred from
elsewhere in the state.
“Through the mid-1990s, each school district really operated independently of each
other,” says Berry. Because of that local control, districts built their own data systems,
creating a hodgepodge of data warehouses,
she says.
In 2009, the state received an $11.6 million
federal grant under the American Recovery
and Reinvestment Act to help bridge the
gaps in the data warehouses that districts
used and facilitate information-sharing
across districts. Under the grant, the districts
were broken into eight consortia, each of
which held face-to-face meetings with members to talk about the challenges of their different data systems.
The grant specifically focused on districts’
assessment-data warehouses, and while it did
not provide money for all members of each
26

EDUCATION WEEK

consortium to switch to the same system, it
did help establish such systems in districts
that did not have their own assessment-data
warehouses previously, says Berry.
“Even though we couldn’t share a lot of
actual data, the information about what
strategies people are using—professional
development or user supports—was hugely
helpful and has saved some time and
money,” she says.
The grant also required districts to provide at least four days of professional development over the course of two years for
teachers to learn how to interpret and use
the data they had access to.
However, when it came to actually facilitating data exchange between districts, not
much progress was made, Berry says.
“We found out almost immediately that
this would prove to be somewhat challenging” to accomplish district-to-district datasharing with the amount of money available, she says.
The state has also taken steps on its own
to help facilitate information-sharing, says
Berry, by creating a school data portal that
provides information to the public and—
through a password-protected login—more
detailed information about individual students to teachers.
But even though Michigan has a set of
criteria for data standards, many districts
do not have the money for a dedicated data
manager, so inputting the data into the system often falls to staff members who do not
have formal training in that area, says Berry.
“They do what they think is right, or what
they’ve always done, and we end up with
ten different ways to code an excused absence,” she says.
Educators in Texas have similar challenges
at the district level, says Melody Parrish, the
director of statewide education data systems
for the Texas Education Agency.
As the system now stands, teachers have
to log in to multiple systems to see different
kinds of data, and they can’t link the data
in one silo to the contents of another, making it difficult to analyze the information in
meaningful ways, she says. For example, the
system cannot compare attendance data with
academic-performance data to predict which
students may be at risk of falling behind.

ROLLING OUT A SYSTEM
But the state is moving to a new statewide data system that will produce feedback reports that can integrate such data
for teachers, Parrish says.
In addition, districts can opt to receive access to teacher dashboards that link educators to a collection of reports about student
academic performance, she says.
“We went out and ... gathered information
on what teachers and campus administrators
and district administrators would need” in
order to develop the dashboards, says Parrish. Teachers expressed a desire for usable,
real-time data that would help them be able
to group and differentiate instruction for
their students, she says.
The system, which is being piloted in a
handful of Texas districts, will be rolled out

>> MARCH 14, 2013

statewide over the next three or four years.
The teacher dashboards have been developed in partnership with the Michael and
Susan Dell Foundation, based in Austin,
Texas, which rolled out its Ed-Fi data standards in July 2011 to help standardize data
across districts and states. Fourteen states
have committed to adopting the Ed-Fi standards so far, says Lori M. Fey, the president
of the Ed-Fi Alliance.
The Ed-Fi data standards are aligned to
the common education data standards developed by the National Center for Education
Statistics, which released the third version of
the standards in January 2013. The common
education data standards, or ceds, are helping facilitate data-sharing across districts
in Georgia and potentially across states,
says Bob Swiggum, the chief information officer for the Georgia education department.
Swiggum has earned recognition in the
education data sector for building a data
“tunnel” that allows teachers in districts
throughout Georgia to access state-level
data and reports analyzing data across districts, without ever having to log in to more
than one system. The state-level reports
provide detailed feedback to teachers about
their own classrooms as well as how their
students compare with students in other
schools and districts in the state.
Before the creation of the tunnel, only 300
out of a potential 150,000 users were engaging with the statewide data system, says
Swiggum. Now, that number is about 60,000.
The tunnel automatically transfers the educators’ login information to the state system,
verifying their credentials and levels of access
to data, and allowing those educators to enter
the state system without having to go through
a second login process.
In addition, the tunnel mirrors the look
and feel of the district-level data system so
that even though educators leave their district systems to view the state-level reports,
it doesn’t feel as if they are navigating to an
entirely new system, Swiggum says.
To build the tunnel, Swiggum worked
with 12 different vendors that provide data
services to Georgia districts to add several
lines of code to trigger the login authentication and website mirroring.
The state has added tools within its system—based on feedback from educators
about what they would find most helpful—to include growth models, longitudinal
data, individualized education programs for
students, and tools to link resources to the
Common Core State Standards.
Listening to feedback from educators was
essential in creating the solution, Swiggum
says.
“You have to know who your audience is
and listen to your audience to find out what
they want,” he says.

STATE DATA CONNECTIONS
Swiggum is also working to connect data
between states. The Southeast Education
Data Exchange, or seed, relies on ceds—the
common data standards—to share information in member states.
Georgia, Kentucky, and North Carolina are

Districts share records in real
time as students and teachers move
across district lines to ensure seamless
enrollment, reduced paperwork, and
more robust academic records.

Data move efficiently across
education sectors to better track
individual students’ progress over
time while providing feedback
throughout the system to ensure that
all students are on track to succeed.

When students, families, and
educators move across state
lines, their data move with them to
ensure that stakeholders have better
information regarding outcomes.
SOURCE:

piloting the project, which matches 85 different data fields between states to exchange
student information, allowing teachers to retrieve information and records quickly for a
student who transfers from a district in one
state to a district in another.
Because there are no easy ways to share
data across states, educators are not getting
a full picture of student performance, says
Paige Kowalski, the director of state policy
initiatives for the Washington-based Data
Quality Campaign, which works to increase
the availability of high-quality education
data to improve student achievement.
For instance, in many states, a large
percentage of students go to college out of
state, which means that without linking
data between states, the states cannot link
K-12 data to those students’ postsecondary
performance, Kowalski says.
“It’s really hard to get 50 states to come to
some sort of agreement on their own,” she
says. “There’s going to need to be funding or
some tie-in to something that’s happening
federally” to drive the creation of a crossstate solution. n

Technology Counts - March 14, 2013

Table of Contents for the Digital Edition of Technology Counts - March 14, 2013