As a 5th grader, Sally Brooks has already tackled several standardized exams and has the test-taking routine down pat. After crawling into bed at a decent hour and eating a hearty breakfast, she arrives at Patton Elementary School in Arlington Heights, braced for hundreds of multiple-choice questions.
But while Brooks and several thousand other pupils in Arlington Heights District 25 are currently laboring over both the Stanford Achievement Test and the Otis-Lennon School Abilities Test, a change in routine is on the horizon.
Next year, District 25 pupils will take Achievement Level Tests, multiple-choice exams that school officials say can measure the wide range of skill levels within any classroom in a way the current tests can’t. Parts of the popular Stanford test and OLSAT might be retained for certain subjects like science and social studies.
“A test like Stanford gives a broad picture of kids as a group. What it doesn’t do is give us very valuable information as to how an individual is progressing over time because it isn’t constructed to do that,” said Janice Miller, District 25 superintendent for special services and assessment. Achievement Level Tests “would be like a yardstick, which measures physical growth in feet and inches. Some kids start out lower, and others are higher.”
While many groups are proposing major reforms in individual state assessment systems, debate grows over President Clinton’s proposal for standardized national exams in math and reading.
According to a recent national study of testing programs by FairTest, the National Center for Fair and Open Testing, one-third of state K-12 student assessment systems need a complete overhaul, and another third need major improvements.
“A large majority of states continue to rely on outmoded, off-the-shelf multiple-choice tests that are poorly aligned with their own standards,” said report author Monty Neill, FairTest associate director.
On a five-point scale, Illinois earned three points and was classified as needing “some significant improvements” in the study, supported by grants from the Joyce and Ford Foundations. Only Vermont’s was declared close to a model system.
“Those tests (like Stanford) tend to measure low-level skills such as recall and proofreading, rather than what many states are moving to, which is performance assessment,” said Bob Schaeffer, public education director of FairTest. Performance assessment means “actually doing things and rating how well kids can master skills and knowledge set forth in clear standards,” he said.
“What you get on Stanford is a ranking on a kid for a school district versus an average, so you can determine whether the student is above or below average,” Schaeffer said. “But that doesn’t tell you whether it was good or not.”
This year, the district will spend $25,000 for the Stanford tests to be scored and reported. A detailed report for each pupil will be provided to parents and teachers within five to six weeks.
“We initially began giving the tests because we felt we needed some yearly evaluation of how our students were performing in relation to students nationwide,” Miller said. “It helps us look at our programs, and in the process re-evaluate the best way to accomplish our goals.”
In addition, Miller said the OLSAT will attempt to assess a pupil’s ability to succeed in the classroom by measuring cognitive skills such as perception, recognition and logic.
“It’s not an intelligence test, but it does help the parents and teacher take a look at where a student is at,” Miller said. “Of course, we certainly look at all of this in the light that a test is only one snapshot in time. Testing is difficult for some kids, and it may not always be a good experience.”
