Educational Leadership - November 2009

Perspectives
The Tests That Won’t Go Away

How many hours of classroom time do you typically spend adminis- tering standardized tests to students each school year? In my search for that statistic, I found one high school teacher estimating he spent 40 school days each year administering and prepping students for “bubble tests.”

Perhaps an even more important question is, How many hours does a teacher spend preparing students for “multiple assessments”?

That answer depends on the interpretation of the term assessment—are you counting pop quizzes and spelling bees, essays and multimedia projects, teacher-made and standardized tests, entrance and exit tests, pre-tests and post-tests, interim and benchmark assessments, statewide and national tests, and preparation for the AP exam, SAT, and ACT? Are you adding in daily, minute-by-minute checks for understanding? If all answers apply, many teachers might answer that they spend all their time teaching, if not to the tests, then with the tests in mind.

There is no doubt that in the past 10 years, school culture has become a testing culture. As assessment experts Stephen Chappuis, Jan Chappuis, and Rick Stiggins write (p. 15), “NCLB has exposed students to an unprecedented overflow of testing.”

But do all these “multiple measures” really lead us to achieve the three most often cited goals of testing: building proficiency in basic skills, closing achievement gaps, and fostering the top-notch knowledge and skills that students will need in a competitive global society? No, according to the testing experts. Neither using single tests nor incorporating multiple measures necessarily leads in these three directions. Our authors describe what makes it more likely that the instructional hours sacri-

ficed to testing will return dividends in the form of more learning for students and better instructional decisions for teachers.

Here, in brief, is what they tell us:

Become assessment literate. If educators understand the different ways to define multiple measures, the various ways to combine measures, and when to use which (Susan Brookhart’s chart on p. 10 certainly helps), they will be one step closer to identifying and choosing appropriate measures for various educational purposes. As assessment-literate professionals, we are in a much better position to educate those in the general public, media, and policymaking positions who blindly accept the validity of any test for any purpose.

Our authors also issue cautionary warnings about specific tests—from the NAEP to TIMSS (p. 32) to value-added assessments (p. 38). It is essential that more people understand the aims of these tests, whom they test and how, and their strengths and limitations in providing useful and valid information about students and schools. Although assessment literacy is no magic bullet—Jim Popham calls it a magic BB— it has a power of its own to transform assessment into a form of teaching.

Keep your eyes on the prize. If the goal is not just achieving higher scores, but furthering students’ learning and understanding, assessment must be for learning, not just of learning. Kari Smith (p. 26) speaks from experience about how to make assessment into a teaching tool. Her students not only take tests but also make tests and learn from all the processes

involved, from wording the questions to grading the answers. Douglas Fisher and Nancy Frey (p. 20) suggest a systematic approach that incorporates three components: feed-up, feed-back, and feed-forward. Together these three steps—not bubble tests—make up the strongest intervention available to increase student achievement.

Asked to name the biggest obstacle
their school is facing this year, 41 percent
of responding educators in
a recent informal ASCD
SmartBrief poll picked
“pressure on students and
teachers to improve test
results.” The general public
has a different slant,
however. In the recent
annual PDK/Gallup poll,
Americans by a two-to-one
majority supported annual
testing of students in
grades 3 through 8.

“We will not soon be

doing away with standardized tests, nor should we really want to,” Kari Smith writes. “Tests illuminate an important aspect of students’ learning, namely the ability to present factual knowledge within a given time limit.” But that knowledge goes only so far. Standardized tests— even under a system of “multiple measures”—do not guarantee better teaching and learning and can have the opposite effect when used wrongly and excessively. It is time to shine a bright light on multiple measures and use them in a more sophisticated way.