The idea of a set of interoperability technology standards that permit states to share data and test items across states and that enables the tagging of test content in a way to make test questions more accessible to different kinds of English-language learners and students with disabilities got a lot of play yesterday in a public meeting about the tests being developed for the states' common-core academic standards.
Michael Russell, the vice president of innovation for Measured Progress, a nonprofit test developer in Dover, N.H., described at the meeting how something called the Accessible Portable Item Profile Standards, or APIP for short, can help to tailor tests to different kinds of students. The standards permit the tests, for example, to turn off or on particular accommodations such as language translation, or audio of test items being read aloud, by tagging content according to student profiles that are part of the system. For example, if a student profile signals that a student is blind, the test may provide what Russell called "refreshable electronic Braille" to that student.
Russell was one of six testing experts invited by the U.S. Department of Education to attend the meeting here with the purpose of advising the consortia of states creating assessments for the common-core standards how best to include ELLs and students with disabilities. Representatives of the four consortia creating such tests, both the regular tests for all students and the alternate assessments for students with disabilities, were at the meeting.
All of the consortia appear to be seriously considering APIP for accessibility. Russell told me that the Dynamic Learning Maps consortia has already adopted it. He noted that APIP is an open source system, not a commercial one. Representatives of the Smarter Balanced Assessment Consortium and the Partnership for Assessment of Readiness for College and Career both told me they are leaning toward using APIP.
Some people at the meeting also pointed out potential problems with such computerized systems for improving accessibility.
Wendy Carver, an assessment specialist at Utah's department of education, said, for example, that her state has moved toward a computer-based assessment system and run into problems with giving students "too much information," with all the different accommodations for accessibility. She also wondered whether states will be able to agree on policies for embedding accommodations into tests.
"Once you start moving into using visuals and animation, you're interacting with stuff, and you can very fast get to overload," said Rebecca Kopriva, a senior scientist at the University of Wisconsin, and one of the six experts invited to advise the consortia.
The experts at the meeting stressed that the creation of the new tests is an opportunity not to repeat mistakes of the past, when test developers looked to the needs of English-language learners and students with disabilities after tests were already created.
They emphasized that special populations need to be addressed at every stage of test development and that the consortia need to create a document with specifications for creating test items that will address accessibility.
Meeting now to discuss how best to make the new generation of tests accessible to special populations of students is "a little later than is optimal, but it's not too late," commented Martha Thurlow, the director of the National Center on Educational Outcomes, at the meeting.