Study Questions Fate of NAEP in Common-Test Era
Common Standards' Math Expectations Are Found to Be Far More Rigorous
An analysis released last week raises anew questions about what the advent of common standards—and the development of common assessments to complement them—means for the future of the National Assessment of Educational Progress, often called “the nation’s report card.”
The Brookings Institution report, which seeks to match up publicly released NAEP items in mathematics with the standards for that subject from the Common Core State Standards Initiative, suggests that the coming common exams signal that “a new era is dawning for NAEP,” though what that future will look like remains murky.
The analysis found that the NAEP items examined were, on average, two to three years below the 8th grade math recommended by the common standards, which have been fully or provisionally adopted by 43 states and the District of Columbia.
“A state might have really high proficiency on one of these tests and not ... on the other,” said Tom Loveless, the report’s author and a senior fellow at Brookings.“There’s the potential that we’re going to confuse a lot of people.”
Issued by the Washington think tank’s Brown Center on Education Policy, the report notes that the discrepancy between NAEP and the type of testing expected for the common standards has to do, at least in part, with varying definitions of what a test at a particular grade level means.
The main NAEP is directed at grades 4, 8, and 12. The 8th grade math test, which the Brookings report focuses on, assesses all the math that young people have learned through the 8th grade, the report says; that is, a lot of the content comes from material presumably learned in earlier grades. By contrast, the tests devised to match the common standards are expected to gauge the knowledge and skills learned specifically by the end of each grade level.
In the end, the report suggests that its findings serve as a sign of the challenges ahead for NAEP.
“Now, common-core assessments are on the way,” it says. “Whether the new assessments push NAEP aside, succeed in augmenting the information provided by NAEP, or force a redefinition of NAEP’s role in monitoring student learning will be at the top of the NAEP policy agenda in the years ahead.”
David P. Driscoll, the chairman of the National Assessment Governing Board, which sets NAEP policy, argues that the federally sponsored assessment will remain vital.
“Even if the common assessments ... are wildly successful and highly acclaimed, I still think you’re going to need NAEP as an objective verifier,” he said.
Mr. Driscoll added that NAEP also holds great value for other reasons. For one, it tests students in a variety of subjects, including the arts, science, U.S. history, and economics, while the common assessments are expected to focus on reading and math. Further, NAEP provides historical achievement data back to the early 1970s.
The Brookings study examines NAEP math items for 8th graders, coding all publicly released items from the algebra and number strands based on the grade at which the common standards recommend teaching the math assessed by the item. In all, 171 items were available.
In algebra, the items on average were at about the 6th grade level, and questions in the number strand were at about the 5th grade level.
“As currently planned, the two programs will assess different mathematics and might report different results,” the report concludes. “Even if they report similar results, each score will reveal something different about American students’ math skills.”
The report suggests that one way to help minimize conflicting signals from the separate assessment systems would be to increase the difficulty of NAEP items, bringing the test into closer harmony with the common standards. Another idea, it says, would be to incorporate adaptive testing, which has flexibility in supplying some questions that are sensitive to individual students’ achievement levels, to help bridge the gap.
Currently, two separate consortia of states are crafting assessments aligned with the common standards. Leading participants in the consortia say their goal is to design the tests in a way that will facilitate cross-state comparisons, both among the states in a particular consortium and even across them.
“The common-core [testing system] is going to be way more difficult than NAEP, and I think that’s going to be a really interesting discussion,” said Mark S. Schneider, a former commissioner of the National Center for Education Statistics, which administers NAEP.
Mr. Driscoll, though, cautions that it remains to be seen what those assessments will look like.
“Until the standards are turned into assessments,” he said, “it’s pretty hard in my judgment to just look at standards and how they [match up with NAEP].”
On the broader question of NAEP’s future, Mr. Schneider, now a vice president at the American Institutes for Research, in Washington, agrees that the implementation of common assessments will raise some tough questions. He also points to the growing prominence of international exams, especially the Program for International Student Assessment, or PISA, as further complicating NAEP’s future.
“Ten years from now, I’m not sure where NAEP sits, if even we’ll need it,” he said. “Again, some of its most basic functions are going to be taken over by the common-core [assessments].”
Bigger Role for Trend Data?
Mr. Schneider suggests that, ultimately, the main NAEP could be pushed aside, while the separate long-term trend assessment may gain greater prominence. That program provides only national results, unlike the main NAEP’s state-by-state data. It remains relatively unchanged since its inception, and tests only in reading and math, once every four years.
“In some ways,” Mr. Schneider said, “the less-visible and less-cited report may turn out to be the key product, because it will give us something that nobody else can.”
Vol. 30, Issue 17, Page 6