Montana students will soon be taking a standardized test that is half performance assessment:
In the usual fill-in-the-bubble standardized tests, students don't have to draw a graph tracking global warming, or write an essay about how a hurricane is formed. When a new kind of standardized test hits the desks of Havre fourth-, eighth- and 10th-graders this spring, nearly half the possible points will come from questions like those, to be answered in the students' own handwriting.
Will students be exposed to material suggesting that the global warming controversy is overhyped? Can they write essays that cite Bjorn Lomborg and still get high scores? And wouldn't learning about tornados be more useful for Montana residents than hurricanes? Could an acceptable essay about hurricanes begin with, "Well, we don't get many hurricanes in Montana, but..."
Okay, I'll be serious now.
The new tests require a specific level of mastery in each subject area. The standardized tests students take now, the Iowa Test of Basic Skills and the Iowa Test of Education Development, score students by comparing their scores with those of students across the country in a given year.
That is, they're moving from a norm-referenced test to a criterion-referenced test. And with the addition of open-ended questions, there's certainly more face validity, but the issues of reliability and scoring get trickier (and more expensive).
In constructed-response reading questions, students read short passages and write short essays on them. A sample fourth-grade question asks students to read a two-page passage about a debate over school dress codes and uniforms. They are then asked to "describe how school clothes can create problems for students. Use information from the article to support your answer."
In a sample math test, fourth-graders were asked to make two different patterns of numbers based on rules they make up, and to tell how many 1-inch cubes would fit inside a box with given dimensions.
The reading section sounds fine, but I'm leery of mathematics assessments that involved a lot of writing, simply because writing ability becomes confounded with math ability on these types of tests. When the math assessment is primarily "story questions" and the results depend on writing and labeling as much as on addition and subtraction, kids who are good at math, but not at English, are going to be at a disadvantage.
Posted by kswygert at February 2, 2004 10:06 PM