A New Kind of Testing
Andy Dousis made headlines earlier this year when, as a member of the East Lyme (Conn.) Board of Education, he took his third-grade daughter out of elementary school during a two-week standardized testing period as a protest. However, this senior consulting teacher for the Northeast Foundation for Children, former elementary teacher and co-author of The Research-Ready Classroom (Heinemann, 2006) gives a thumb's up to daily or formative assessment.
"That's the kind that master teachers do in conjunction with the learning that goes on in the classroom," he explains. "The good ones know how to take that information and put together a plan for not only the class but each student."
Thousands of researchers, administrators and teachers agree with Dousis because the practice has little to do with traditional tests-and the evidence is mounting that this approach reaps serious results. To date, Vanderbilt University researchers Lynn S. and Douglas Fuchs cite more than 200 empirical studies published in peer-review journals as evidence of this approach's reliability and validity.
By the Numbers
America has had a national love affair with standardized testing during the last 60 years, with congressmen passing laws to ensure we keep collecting data. Rick Stiggins, executive director of the ETS Assessment Training Institute in Portland, Ore., refuses to badmouth the trend, even if he was at one point "a loser in the system," as he refers to his inability to read and make the grades by those standards. Test scores at the national level help politicians with policymaking, resource allocation and other 10,000-foot-level needs.
Principals, curriculum directors and teacher leaders need to know not so much how each student is progressing, but who is and isn't meeting standards. Every nine weeks or so, they need data to assess whether their programs are working or require adjustment. And that kind of information comes from tests.
Teachers at the classroom level need continuous information on every student-feedback to help them determine what comes next so that each individual progresses toward the relevant achievement standards. This, says Stiggins, is formative assessment, which he likes to call "assessment for learning." Others know it as progress monitoring or curriculum-based measurement. By any name, "it's absolutely essential if we are going to make sure that students are learning exactly what we want to teach them," says G. Reid Lyon, executive director and executive vice president for research and evaluation at Higher Ed Holdings and Whitney International University Systems in Dallas.
Perhaps more important, Lyon was previously chief of the child development and behavior branch within the National Institute of Child Health and Human Development at the NIH and, as an advisor to President George W. Bush, helped develop the No Child Left Behind standards. When he talks about testing, folks pay attention.
Going Against the Grain
Neither Lyon nor Stiggins advocate tearing down standardized testing as a healthy direction; they both say the answer to testing effectively lies in changing the classroom attention. "We need balance; the system must work on all three levels," says Stiggins. "Historically, we've struggled because of our obsession with the once-a-year test, which isn't very helpful to teachers making decisions every three to four minutes. We need to make sure they have access to that dependable information every three to four minutes."
Even Dousis will agree there's a place in education for standardized testing, but says it should be done through a sampling. "If you want to know how many folks support the war in Iraq, you don't have to ask every person in the country," he says. "You take a sample. This is what good science does." Formative assessment, in his eyes, means the kids not participating in the sample test one year wouldn't necessarily fall through the cracks.
On the other hand, formative assessment isn't a magic pill to cure the pressure on teachers to make sure their students perform well on standardized tests. "But we do have compelling evidence that when that classroom-level assessment process is managed productively, student achievement skyrockets on the annual tests," says Stiggins. "Good decisions made continuously during the learning sets kids up for success."
A Closer Look
Quality formative assessment begins with teachers mastering the state standards students are expected to hit. Next, the teacher deconstructs those standards into what Stiggins calls scaffolding: learning nuggets the kids will climb on as they work their way up to each standard. In math, for example, children gain control of numbers, numeration systems and specific reasoning patterns before learning to solve problems.
The teacher's responsibility is to transform each achievement target into a high-quality assessment, such as a multiple-choice quiz, an essay assignment or a performance project. The paradox is that, historically, teachers have not been trained to create dependable assessments.
Finally, the student needs to understand her goal and where she currently stands in regard to it. In essence, teaching becomes all about helping students close the gap between the two. Many educators hand out examples of good and poor writing, for instance, to demonstrate what their students will be capable of. "We're using the assessment process to get kids on winning streaks so they remain confident they'll get there if they keep trying," says Stiggins. "We're trying to avoid the counterproductive response of, 'I can't learn this any way. I quit.' We can't have kids giving up in hopelessness, not in a standards-driven environment."
Feedback, which includes helping students generate their own critiques, is another important component. The idea is to make each child a collaborator (with her teacher) in managing her education, and the possibilities are endless. Stiggins has seen teachers enlist class help in developing comparative questions for homework assignments, write multiple-choice quizzes or sit side-by-side with individual students to solicit their input in grading essays. With this mindset, homework is akin to practicing the piano.
Where It's Working
Four years ago, when Steve Price, the superintendent at Middletown City Schools in Ohio, introduced teachers to formative assessment, many of them shot back, "I already know how to test." His first assignment was to show them the difference. "It's not just evaluation," says Price. "It's understanding how to use this to motivate kids.
"And it's not a blame game," he adds. "These are things they didn't teach in college. You're left to get this stuff in your professional development when you get out in the field."
Teachers in Bloomington Public School District #87 in Illinois, too, showed initial reluctance to adopting the strategy eight years ago. "They thought this was just one more thing we were asking them to do," says Robert Nielsen, the superintendent who made this a priority his first summer on the job. Few would blame those teachers, since studies indicate that a typical teacher spends one-half to one-third of his professional time under the traditional system.
But that foot-dragging quickly disappeared. "Teachers now say this is amazing," says Nielsen. "These kids are doing phenomenal work and teachers are very proud, so it becomes an escalating good news: Let me show you what my class can do." Proponents say formative assessment makes grading easier, therefore saving time instead of eating it.
Taking the Next Step
Whitney International University embraces formative assessment as part of its educational curriculum, and companies like ATI have produced materials to instruct teachers in this approach. But on a practical level, the bucks needed to train today's teaching ranks in formative assessment are stuck in another era. ATI devotes a chunk of its time lobbying policymakers to include continuous assessment on the same platform with standardized testing. Still, where the spirit is willing, the pocketbook remains empty, which is why innovative leaders like Nielsen are finding corporate partners to get the job done. Today, a vast majority of his 440 teachers in Bloomington have received training thanks to $50,000 that State Farm Insurance has contributed for the past three years to pay the bill.
Bruce Herzog, a fifth-grade teacher in the NookSack Valley District in Everson, Wash., did his formative assessment coursework at ATI thanks to grant money from the Bill and Melinda Gates Foundation. Meanwhile, Price squeezed what he needed out of his Title I funds.
Here Come the Products
Where there's even a handful of change, savvy entrepreneurs seek to turn the situation into a business. Take, for instance, Wireless Generation's mClass software and handheld computers designed to help teachers track skills like reading rates and word accuracy. Co-founder and chief executive officer Larry Berger loves to post success stories like Fogarty Elementary in Oklahoma's Guthrie Public School System, which reduced the time needed to assess each student's progress from 15 to 20 minutes per child down to 10 to 12 minutes per child. Students who were reading 50 words per minute in January 2004 had rocketed up to 100 words per minute 10 weeks after officials there handed over the handhelds loaded with software to the second- and third-grade classrooms.
New Mexico and Ohio put Berger's product in every Reading First school in their states in 2003. The Department of Education cited New Mexico's progress in its 2005 National Educational Technology Plan. Teachers give it rave reviews. So does Lyon, and not just because it saves time and paperwork. "When the data comes directly to the teacher, it groups students in terms of the types of difficulties they're having," he says. Teachers can then spend the rest of that class time on those areas.
That's why Lyon sees more and improved technology coming down the pike. "But frankly, we never would get anywhere unless we learn whether kids are learning. Progress monitoring is where the rubber meets the road," he says.
The Northwest Evaluation Association, a national non-profit organization for educators in Portland, Ore., calls its computerized assessment program for reading, math, language and science in this arena Measures of Academic Progress, a feedback system that weighs academic growth across years of time. NWEA's 2,100 school districts across the country are licensed to give the computerized test that provides the child with results immediately; the teacher receives classroom data by 6 a.m. the next morning up to four times a year. The group estimates several million students have used its testing.
"I am often asked by directors of curriculum, 'If we use your test, will student learning improve?' " says Allan Olson, president and executive director of NWEA. "And I say no. If you use our tests and teachers and administrators and counselors use the data to change what they do, then learning will improve."
Even ATI itself has upped the ante on the business front. Educational Testing Service purchased the company earlier this year to publish and deliver its teaching products. But such partnerships don't make friends with Andy Dousis. "When I hear there is a company putting something together around formative assessment, I get a little leery," he admits. "You don't need a program to do formative assessment. You need experience, you need coaching, you need a good mentor."
Nielsen judges the growing number of players in this niche by their associations, such as if the group has leading international scholars on board.
Dousis recommends administrators put approaches to work in a classroom before leaping into a wholesale purchase. The catch: Forget measuring what it does for student test scores. How does it change the teacher? After all, the goal is to change learning, not merely test scores.
Julie Sturgeon is a contributing editor.