Dealer, August 24, 2008
Tests measure what students know. Like a Polaroid, they give a
snapshot of knowledge frozen at one moment in time.
But what if you could measure how much a child learns over the
course of a school year? What if you could gauge what a school
actually adds to a child's learning experience?
In Ohio, you can.
This year's district and school report cards, which will be
released Tuesday by the Ohio Department of Education, for the
first time will include a measurement known as value-added.
The revolutionary formula, designed more than two decades ago by
a homespun statistical guru from the rolling hills of eastern
Tennessee, has rocked the education world.
Put simply, value-added tracks whether a year's worth of
learning is actually happening in the course of a school year --
regardless of whether a child passes a test at the end of that
Ohio schools and districts that exceed expected growth will be
rewarded with a plus sign on their report cards and could move
up a rung on the ladder of academic ranking. Those who make
expected growth will get a check mark. Those who miss their
target will get a minus sign.
More important, the formula will show parents which schools are
really teaching children something and which schools are
It will reveal that in some affluent districts that easily hit
state testing targets, students aren't being challenged enough.
And it will disclose that students in some high-poverty
districts are making amazing gains, even if they aren't yet
passing state tests.
Think of it this way: Many a parent has used pencil marks on a
bedroom wall to plot a child's growth. Value-added does the same
thing, measuring academic growth rather than penalizing a child
for not making a predetermined height like, say, 6-foot-2.
"It is, at this point in time, the most robust methodology that
has been developed," said William Sanders, who began working on
the concept 25 years ago as a University of Tennessee
researcher. "It's fair to hold adults accountable for the
progress rate of children."
Ohio and just three other states -- Tennessee, Pennsylvania and
North Carolina -- use the value-added measurement statewide.
Ohio's valued-added data is based on the reading and math scores
of public-school students in grades 4-8.
A Columbus-based nonprofit, Battelle for Kids, began the
value-added experiment in Ohio in 2002 with a pilot project
using the concept in 42 volunteer school districts. By last
year, the experiment had evolved into the largest value-added
pilot program in the United States.
State officials hired Sanders -- an affable professor with an
easy drawl and a keen mind for numbers -- as a consultant.
Together they fashioned a formula that fit Ohio's system.
What makes Sanders' formula compelling -- and controversial --
is its ability to predict what a student should achieve. When
actuaries compute life expectancy, they take into account
factors such as a person's family history, or habits such as
smoking or hobbies such as skydiving. Likewise, value-added
takes into account a student's academic background and comes up
with a predicted score.
"This is the most important thing we do, in my view," Sanders
said. "I'm not going to get them all right, but if you bet
against me, you're going to lose."
It also represents a seismic shift in education research. For
decades, educators embraced the thinking of the late sociologist
James Coleman, who argued that a child's socioeconomic status
largely determined academic achievement and that what schools do
doesn't matter much.
Sanders turned that thinking on its head. Instead of adjusting
for the income level of a child's family, Sanders compares the
child against his or her own past academic performance.
Because all children don't achieve at the same level, schools
are evaluated on what "value" they add to a student's
experience. Schools that don't add anything need to ask why.
"If a district is below the expected level of growth, it needs
to take account of what's going on," said Matt Cohen, executive
director for policy and accountability for the Ohio Department
Not everyone is completely enamored of Sanders' work. Some, such
as Virginia-based education researcher Gerald Bracey, contend
value-added is flawed because it is still based on test scores.
And the Education Trust's Kati Haycock warns that while growth
is important, it's also crucial that all children -- poor, rich,
black, white -- eventually master academic skills.
"It's not just growth," said Eric Gordon, chief academic officer
of the Cleveland public schools. "It's growth and standards."
But the most potentially explosive thing about value-added is
its ability to determine which individual teachers are effective
and which are not. Some school districts, such as Houston, are
already using Sanders' work as the foundation for a performance
pay system for teachers.
In Ohio, Battelle for Kids is using valued-added in a volunteer
pilot program involving about 1,800 teachers in about 240
schools across the state. What it has found: effective teaching
is not limited to the state's affluent districts with the
highest test scores.
The Ohio pilot is not being used to evaluate or pay teachers but
rather to learn what works in the classroom and what doesn't,
said Jim Mahoney, executive director of Battelle for Kids.
"The worst thing that can happen is that you're given this tool
and you make it into a weapon," Mahoney said. "The fact that a
tool might be misused is not a good reason to do away with it."
Questions and answers on value-added measurement of students
These are some frequently asked questions about the value-added
Q. How is value-added different from traditional measures of
Student performance can be measured in two very different ways,
both of which are important. Achievement describes the absolute
levels attained by students in their end-of-year tests. Growth,
in contrast, describes the progress in test scores made over the
Q/ Why is measuring both achievement and progress important?
In the past, students and schools have been ranked solely on
achievement. The problem with this method is that achievement is
highly linked to the socioeconomic status of a student's family.
In contrast, value-added assessment measures growth and answers
the question: How much value did the school staff add for
Q. Is it possible to show progress with all students -- special-ed,
gifted and low-performing
Yes. The value-added methodology used is sensitive to individual
students' achievement levels. It measures growth from the end of
one year to the end of the next year, regardless of whether a
student performs below or above grade level.
Q.Does value-added analysis require additional testing?
No. Value-added analysis uses existing standardized test data
and can be done only where annual testing is provided.
Q. Do socioeconomic factors affect valued-added measurements?
No. Leading experts have shown that those factors -- such as
household income or parents' education level -- have no
significant relationship with student progress measures.
SOURCE: Battelle for Kids; University of Pennsylvania.
back to the top ~
back to Breaking News
~ back to