Ever since two big groups of states won federal grants to design tests for the common standards, lots of folks have been piling on to offer advice about how the tests should be designed. Some has been solicited directly by the U.S. Department of Education (see here and here). But much of it has flowed into the dialogue by way of white papers and such.
The latest comes from the Center on Education Policy, which has long monitored the implementation of No Child Left Behind and its related policies. In a letter to the assessment consortia, posted today on its website, the group suggests that common-test designers take key steps to maximize their usefulness.
The letter speaks for itself, so take a look. But in essence, the Center urges the consortia to design a system which will expand the kinds of score reports available, and enable comparisons of data across all states. (The consortia have already said that comparability is a high priority.)
For example, the CEP encourages the consortia to make sure that their systems produce average assessment scores for the entire group of tested students, as well as for student subgroups at the state and local level. Since NCLB requires states to report what portion of students reach key achievement levels, many states don’t report mean scores that could offer insight, the CEP says.
A consortium-related side note: the PARCC group has a newly redesigned website. PARCC also has a page on the website of the Florida Department of Education, since Florida is its “procuring” state, and a page on the site of its project management partner, Achieve. The other consortium, SMARTER Balanced, plans to revamp its website as well, but currently it’s got space on the website of its “procuring” state, Washington.