EDFacts – Education Datapalooza

My job is to lead something in the department
called EDFacts and many of you sitting in the
room, you heard Arne say the fuel to a lot of these great ideas is the data. And the
US Department of Education source of K-12 data
is EDFacts. But EDFacts itself isn’t going to solve
a lot of the ideas or fuel a lot of the ideas you have in your head, when you think about
the entire data chain of education. So I want to take
just a couple minutes to get us rolling this morning
and ground us and a quick understanding of where do the data about individual K-12 students
start, where do they go, how did they move. And that all starts with schools and school
districts. We can all think about that educational moment between a teacher and a set of students.
That’s where the data begins. That’s where everything starts. Districts are the ones collecting the data.
The schools, the teachers, they’re managing where they
can, but it’s really districts who are thinking about the systems, what can they provide teachers
for grade books, for attendance systems, et cetera. And why are they doing it? They have
operational needs to go beyond just the teacher. They’re federal and state programs, funding
needs they’ve got for state legislation, there is payroll, busing, daily attendance, they
have to provide transcripts. There are operational
needs that they as a school district have to manage.
For that reason, they’re building data systems. I’ve been in school districts where the data
system – I’ve had the secretary say you mean this
thing? This file cabinet where the end of the year i pick
up a fifth grade label and I make it the sixth grade label and I take out the sixth grade
label and I make it the seventh grade label? So there’s that version of a data system but
more and more districts have moved away from that,
almost all of them at this point to something that gets the data in that machine readable
format. How do we keep it moving from there and not
having to go back in the something static? So
most of what’s in those district systems, they’re going to have information in their
systems about certification of their teachers, hiring data,
compensation, length of service, evaluation of their
teachers, it’s all living in the system because they need it at that local operational level
for their school boards, for their superintendents.
On the individual students, who are they, how will they
get classified in various subgroups and other populations, attendance, grades, etc. It’s
all at the individual level and being managed as I said.
At some smaller districts, very low-tech formats, but more and more in more open systems. Then those school districts, whether they’re
one or two schools districts or whether they’re Los
Angeles Unified or New York City Public Schools, they’re all working to send data up to the
states. In the past ten years, we’ve seen states develop significant information system
capacity they did not have ten years ago. We’ve invested
a lot of money as the US Department of Education since 2005 in grants for statewide
longitudinal data systems. Their operational needs
as a state education agency in K-12 are very different then the operational needs a local
school district, but they’re still basically needing
a student level set of data, that’s what they found. They
used to just gather up some statistics from their districts, but they’re really realizing
to meet state legislative needs, to meet their own policy
concerns, to understand how to keep their state of
viable economic market where the kids coming out of K-12 are hirable, or at least trainable
or ready for second for secondary and postsecondary
education in the twenty-first century. They need to look at individual student data to
do those large-scale analyses. So most of what they’re collecting, while
it is for program reporting, compliance, legislative analysis, operational improvement, or providing
technical assistance, most of it is a subset of
what’s in those local districts systems. They’re not polling daily attendance, most of them,
they’re not pulling individual quarterly grades.-
Excuse me. – But they are pulling information on what
accountability termination was made for that school, when do kids move in and out of schools.
That was the first thing most state agencies set up a district system to do. Buying an
established common statewide ID so that when a student
moves, a new district receiving that student had a
place they could search find out, Oh! This kid joined us from this other school district.
Let me contact them to see if I can get the records.
Sometimes parents bring all that information, sometimes they don’t. So the states are building
systems, and then they started getting together thinking about how do we connect across states. So it’s a question of how frequently do these
systems refresh from their local school districts. Sometimes it’s monthly, sometimes its quarterly,
for some statistics it’s just annual, it depends. But they’re loading it into longitudinal student
level systems that they can look at over time to
answer those questions like Arne was just mentioning. How do we get more kids ready
and graduating and staying with high school? Who
were the ones most at risk of dropping out and
we identify them not a twelfth grade but in the eighth grade, in seventh grade, and can
we help them through that transition out of middle
school into high school to the point where they really
understand why. It’s important to keep pushing through school, what the applicability is,
where they can take it, what they should be focused
on. And most states are really asking this question,
how can they turn that statewide data back to the districts? How could something that
is statewide be applicable back down? What feedback
loops can they put in place? Can they connect at that state level to higher ed databases
or work force databases and return some sort of
a report to their districts that say your kids who graduated, here where’s where they
are four years later ,six years later, things along
those lines. And that – those are questions we haven’t had
the data maturity to do with education in most states; some states have been wrestling
with this, a lot of states are just putting the systems
in place. But how do we get larger than one school
district learn from the large scope of that data and feed some knowledge in value back
for the operations, to the policy decisions, etc. Which brings us to the final step. –Someone, where am I supposed to point this?
There it goes. – Which is EDFacts and if you think about
each state, each state is setting up their own systems.
They are unique to state legislative mandates, they are unique to the considerations of a
state superintendent or governor. There’s a lot
of consistency, they’re collecting a lot of the same
information, and we’ve been working for the past three years, to put in place common education
data standards across the country to help the states map their systems to something
common and consistent, so that they can pass information
around the region and not just keep it within a state.
But each state has some aspect of uniqueness to it. EDFacts doesn’t. EDFacts is a consistent
– consistently structured national model. It’s also
aggregate. Where all those systems in the states are student level and teacher level,
we are collecting aggregations. There’s nothing at
the individual unit record level inside EDFacts as a
federal K-12 system. It’s the data needed for Title I, for individuals with disabilities,
for Title III, for NCES, National Center for Education Statistics,
and its statistical programs. So we collect all
that information, how many kids did wind of graduating in a certain year, why did that
actually turn out to as a graduation rate, what percent
of kids – what number of kids are proficient, how
about kids who met a state definition of truant. There’s a lot of counts in there, and some
of them depend upon state policy and state definition.
Others are very consistently defined across the
entire country. We use that for a lot of different purposes,
our improved monitoring, target and technical systems, our policy analysis, and all those
membership counts, the assessment results, those
program participation counts. We also want to do with those is we know they’re not going
to solve a lot of the problems people are thinking
about solving that do need to be solved. The My
Data button may not apply to EDFacts, but there’s a lot of other context that can be
gained from EDFacts from looking at the entire scope of
schools across the number of different programs and
how to pull it all together. So as states and districts are thinking how do we use our
student level data and get it into more settings, we think
there’s a lot we can provide in terms of context from
EDFacts and we can do that by getting more aggregate data files out which is something
we will be working on over the next few months.

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *