The Math Forum

Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Math Forum » Discussions » Education » math-teach

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Test Scores Are No Sure Guide to What Students Know
Replies: 0  

Advanced Search

Back to Topic List Back to Topic List  
Jerry P. Becker

Posts: 16,576
Registered: 12/3/04
Test Scores Are No Sure Guide to What Students Know
Posted: Jul 12, 2014 9:15 PM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply
att1.html (6.5 K)

From The Wall Street Journal, Friday, July 11, 2014. See
Test Scores Are No Sure Guide to What Students Know

Results Say More About the Way Test Makers Decide to Measure
Children's Knowledge

By Jo Craven McGinty

When New York and Kentucky rolled out the first tests aligned with
the Common Core State Standards, the results were dismal: Most
students failed the new standardized tests, in stark contrast to the
old assessments, which the vast majority passed.

The results alarmed parents, but the scores on these new tests-just
like those on earlier forms of assessment-reveal less about what
children know than about the way the test makers decide to measure
that knowledge.

The National Governors Association and Council of Chief State School
Officers unveiled the Common Core standards in 2010, saying they were
intended to raise academic standards, and the test scores so far
appear to reflect the increased expectations.
SIDEBAR GRAPHICS: View Graphics at
When New York first administered Common Core tests in 2013, less than
a third of students demonstrated "proficiency"-considered a pass-in
math and English. Kentucky recorded similar scores when it launched
its assessments in 2012. And results are expected to be comparable in
other states as they roll out new tests next year.

Experts warned, however, against reading too much into the
proficiency labels or pass-fail numbers in assessing how much
students were actually learning.

"It's the same set of kids," said Jonah Rockoff, an economist at
Columbia Business School who studies school accountability. "It's the
same set of teachers. Their school didn't change dramatically

In devising the grading scale for the new tests, New York used a
"bookmark" method to identify four levels of achievement from "well
below proficient" to "excels," according to Ken Wagner, deputy
commissioner for assessment and curriculum at the New York Department
of Education.

"We brought together 95 teachers from across the state," Mr. Wagner
said. "We gave them the test the students took in order of difficulty
from easiest to most difficult. You keep going until you say, gosh, I
don't think a student at level one would get this correct, but
someone at level two would."

At that point, each teacher dropped a "bookmark" and continued until
the threshold for each performance level was identified. The panel,
divided into math and English groups, repeated the process four times
before arriving at final cut scores.

"It's a deep concept," Mr. Rockoff said. "How do you send a message
to kids about what is good enough?"

Setting cut scores can indeed be somewhat arbitrary, and New York
provides another example of how manipulating the scale changes the
perception of how well children are performing.

In 2009, before the Common Core was introduced, 86% of students
taking New York's standardized tests scored proficient or better in
mathematics and 77% in English. Those impressive results led the
state and outside observers to conclude that the tests were too easy.

To address the criticism, New York simply raised the cutoffs, so that
passing required a higher score. With the new scale, 61% of students
ranked proficient in math and 53% in English.

"We use words like 'proficient' that carry a lot of baggage," said
Daniel Koretz, a Harvard professor who is an expert on educational
assessment and testing policy. "People believe they know what these
labels mean. It has nothing to do with how well kids are doing. It is
a way of making a judgment about how performance is going to be

Various groups object to the Common Core standards, which were
drafted with the aim of better preparing students in all states for
college and career. Some regard the uniform approach as an intrusion
on states' rights. Others object to the frequent testing. Teachers
dislike being evaluated based on student test scores.

But such debates over Common Core have diverted attention from
legitimate questions about whether U.S. schools adequately educate
children, and there is compelling evidence to the contrary.

The federal government's National Center for Education Statistics
compared individual states' assessments with the "nation's report
card"-an assessment mandated by Congress to provide data on student
achievement-and found that, by its standards, most students weren't
proficient in reading or math. A respected international assessment
of 65 countries shows U.S. students scoring below average in math and
average in reading and science. And a study by the National Center
for Education Statistics found that 20% of college freshmen in the
U.S. need remedial classes.

"We have 210,000 students," Mr. Wagner said of New York. "Only 35%
graduate-college- and career-ready. That means almost 140,000
students every year, year after year, are leaving the fourth year of
high school not ready for what's next. That's a real stark reality."

Still, experts caution that concerned parents who want to understand
more about their children's education should look beyond testing

"The deeper question parents ought to be asking themselves," Mr.
Rockoff said, "is 'Did I know what my kid was learning last year, and
if I compare it to the new Common Core curriculum, am I happy or
Write to Jo Craven McGinty at

Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© The Math Forum at NCTM 1994-2018. All Rights Reserved.