Near the end of a public meeting held here last week, the director of the Race to the Top Fund competition at the U.S. Department of Education, Joanne Weiss, asked a group of assessment experts to summarize their thoughts about how the federal agency could work to improve the country’s assessment systems.
“Good luck,” deadpanned Lauress Wise, a scientist for the Alexandria, Va.-based Human Resources Research Organization, a nonprofit evaluation group.
The remark drew laughter from the researchers, federal officials, state assessment directors, and test vendors in attendance. But it also underscored the challenges the department faces in spending $350 million in economic-stimulus money to aid consortia of states in developing common assessments in reading and mathematics.
Three common messages emerged from the testing experts convened for the first of three meetings being held to advise the federal officials on how to design the competition for those funds:
• State consortia should consider devising assessments to aid instructional practices, in addition to the annual accountability tests now required by the federal No Child Left Behind Act.
• Teachers must be much more involved in the development, use, and possibly even the scoring of assessments.
• The Education Department should seek to structure state consortia in such a way that the one-time infusion of cash leverages sustained work.
In an interview, Ms. Weiss said that the early feedback had already started to shape possible approaches to the competition.
“I think there’s actually a path through,” she said.
The meeting was part of a series intended to help officials gain input into the design of the competition, one part of the Race to the Top program established under the American Recovery and Reinvestment Act. The Education Department plans to issue a final competition notice and application guidance for the assessment funding in March. (“Starting Gun Sounds for ‘Race to the Top’,” this issue.)
The ambitious goals for the assessments include providing teachers with more-useful instructional information, measuring student growth, and gauging teacher and principal effectiveness.
Early in the Nov. 12 hearing in Boston, Ms. Weiss sought to reassure panelists that the department wants states to build multiple ways of testing students, not just one assessment instrument that attempts to perform all those tasks.
To do so, the panelists said, state consortia should consider providing a comprehensive measurement system, including support for formative, real-time classroom instruction; benchmark tests to provide a sense of student progress over the course of the year; and higher-quality accountability tests. Each should be aligned to support an emphasis on college and career readiness, they said.
As part of that work, the experts said, funded consortia must pursue test-item formats capable of measuring higher-order critical-thinking skills, including performance-based tasks.
Such tasks—which might require students to engage in the process of scientific inquiry, write a research paper, or give an oral presentation—are typically embedded in curricula, reflective of real-world scenarios and able to provide richer information about a standard.
But the scores on such tasks tend to have lower mathematical reliability than those for standardized-test items, and the experts disagreed about whether such performance tasks could or should be integrated into school accountability scores in a fair, reliable way.
Raising an issue that has proved controversial in the past, several of the panelists invited by the Education Department to testify said that teachers should be intimately involved in the design of those tasks, even to the point of contributing to the scoring process.
“What we have found in the use of our testing program is that people become very familiar with what the standards are for their particular area for the curriculum,” said Jim Dueck, the assistant deputy minister of accountability and reporting in the Canadian province of Alberta, which relies on panels of teachers to score parts of its grade 12 tests.
Greater teacher involvement also creates better opportunities for professional development and teacher buy-in, said Jeff Nellhaus, the commissioner of education for Massachusetts.
“Teachers end up being the best ambassadors of your [testing] program when it’s being criticized,” Mr. Nellhaus said. “Having teachers involved in item development, [and] scoring of performance tasks, to the extent feasible, is critical to ensuring the quality, transparency, and integrity of the system.”
Assessment experts also encouraged Education Department officials to think carefully about how to encourage consortia that will work together effectively.
The consortia must be able to deal not only with interstate disagreements, but also with any political fallout from testing, added Henry Braun, who formerly worked for the Educational Testing Service.
“Just think about a consortium of eight states that each has a contract with a different vendor on a different schedule,” said Mr. Braun, now a professor of education and public policy at Boston College. “I think you have to recognize that there are going to be complaints. And we’re going to get a complaint that the assessment tail is not only wagging the instructional dog, but waving it around and sending it into orbit.”
The next hearings to gather input on the assessment competition will take place in Atlanta and Denver.
Coverage of the American Recovery and Reinvestment Act is supported in part by grants from the William and Flora Hewlett Foundation, at www.hewlett.org, and the Charles Stewart Mott Foundation, at www.mott.org.
A version of this article appeared in the November 18, 2009 edition of Education Week as In Funding Common Assessments, Tough Challenges