Special Consideration

##AUTHORSPLIT##<--->

Year-end testing can produce a false picture of studentswith unique challenges. They need individualized, frequentassessments—thanks to technology, they're getting them.

Special ConsiderationSEVERAL YEARS AGO, Sandy Futrell had a fourth-grade specialeducation student who, according to all the tests, couldn't recognize lettersor read math problems. "Like other teachers, I assumed that because studentsare in special ed, they have trouble in every single thing," she says.

After a few weeks, however, Futrell, the media specialist at Oklahoma City's Westwood Elementary School, discovered that the boy was quite intelligent, though not always willing to talk. After some prolonged effort, she would read to him and ask him questions, and he'd answer everything. By the end of the year, after using a phonics program, he could read at a firstgrade level. When he read a first-grade book to her, Futrell broke down in tears. "He'd been there five years," she says, "and no one had figured it out."

After the boy's reading improved, so did his math skills. The episode spurred Westwood Elementary to look differently at the way it handled assessment, especially for special populations. "Is it a math problem? A reading problem? A comprehension problem?" Futrell says. "We [began to look] at each individual skill with each individual child."

And that is precisely what yearly state assessments fail to do. Futrell's fourth-grader and other kids with similarly unique circumstances— whether they have emotional or learning problems, English language deficiencies, or simply bad dispositions—are poorly served by one-size-fits-all state tests, which don't give special needs students the attention they require throughout the year and may not deliver an accurate view of their abilities. Accordingly, they need individualized assessments, and software makers are rising up to provide them.

Electronic individualized assessments do double duty: They determine what a student can and can't do, and then point the way to the appropriate instruction. But most critically, they are given regularly, so the students who need special instruction can get it throughout the year—before they're faced with a oneshot end-of-year test, which, if they score poorly on it, can brand students undeservedly and stunt their progress through the school system—not to mention the damage low test scores can do to a district's Adequate Yearly Progress numbers.

Testing, Testing

Westwood had been on Oklahoma's "low-performing" list for five straight years. Nearly all its students are in the free- or reduced-lunch program, and most of them speak English as a second language. "Some kids can't read anything," says Futrell, who has been at the school for 13 years, "but they understand it if you read it to them. Other kids don't recognize the letters and sounds, and they still don't understand it if you read it to them."

The school's principal, Jan Borelli, came to Westwood three years ago. Ever since, the school has made striking progress. It's no longer low-performing; in fact, last year Westwood students scored in the top 10 percent on the Oklahoma Core Curriculum Tests. Borelli's strategy is to use a variety of assessments:

  • Dynamic Indicators of Basic Early Literacy Skills (DIBELS) to assess students in kindergarten through grade 3 at the beginning, middle, and end of the year
  • Riverside Publishing's Gates-MacGinitie Reading Tests at the beginning and end of the year
  • district benchmark exams aligned to state standards three times a year
  • mini-assessments with the district's Edusoft software
  • Learning Letter Sounds for students in kindergarten and first grade
  • Renaissance Learning's Accelerated Math and Accelerated Reader for students in grades 1 to 5

"There's not a silver bullet that's going to work for every kid," says Borelli. "You have to teach them wherever they are." She meets with teachers by grade once a week to review student progress based on assessment data, and then together they decide which interventions to make.

Borelli is sensitive to the charge of over-testing, but she says, "If you're not assessing all the time, how do you know if they're getting it or not? We need to know what they know so we can teach them what they don't know." Her goal is to have 90 percent of Westwood's students scoring at or above standard in all subjects; she estimates they'll get there in the next year or two.

About 10 miles east of Westwood Elementary School, in neighboring Midwest City, June Weston coordinates a specialneeds alternative program at West Side School. She has found a lot of success using technology to engineer assessments. She says her students "learn better and show greater progress when computer programs are used than when direct instruction is applied." Weston works mostly with male teenagers who have behavior disorders, emotional disturbances, learning disabilities— and a history of trouble with law enforcement. How tough are these kids? One day, a superintendent visited Weston's classroom and saw the following objectives written on her board: 1) Stay out of jail. 2) If the police say ‘Stop,' then stop; don't run faster. That's tough.

"Their behavior overpowers their education," Weston says. "Finding out exactly how smart they are, or how smart they're capable of being, is a real challenge."

For the past five years, she's been using the American Education Corp.'s A+nyWhere Learning System. The online system assesses students against specific standards and then prescribes individual lesson plans based on the students' needs. The courseware itself contains more than 5,000 lessons, 200,000 pages of content, and 126,000 assessment items. Weston, who started teaching in 1975 with boys who had recently been released from a reform school in Helena, found that she seemed "to be able to reach this population when others could not or would not work with them."

Weston started using the program thanks to a West Side student in the 11th grade with oppositional defiant disorder, indicated by a pattern of defiance and hostile behavior toward any type of authority. The student responded to it—and to Weston— very slowly. Early on, though, Weston was able to determine that he read at a fifth- to sixth-grade reading level; soon after that, he was able to improve to an 11th- to 12th-grade reading level. She remembers thinking, If it will help this kid, what will it do for the others? She brought in her director of special education, and they ordered the program for the school.

These days, Weston begins assessing her students by telling them, "I don't care what you don't know. I want to know how much you know." And then the computers come out. "Once I put computers in front of these kids, their ability just shot through the roof," she says. "They weren't scared anymore. A computer they can handle."

"If you're not assessing all the time, how do you know if [students] aregetting it or not? We need to know what they know so we can teachthem what they don't know."
—Jan Borelli,Westwood Elementary School

Raising Expectations

Along the southern edge of Missouri, among the Ozark Mountains, school districts are seeing big jumps in student achievement, following the introduction of assessment tools. Nick Nichols, principal of Noel Elementary School in Noel, MO, in the southwest corner of the state, also uses the A+nyWhere Learning System. A full 40 percent of his 560 K-8 students are English language learners, most of them Hispanic. They've been struggling with statewide tests and state AYP standards. But A+nyWhere includes a language arts and a mathematics series, both in Spanish, with assessment modules. Nichols makes the program available after school, for whoever wants it. He reports outstanding gains, such as two-year increases in reading levels, citing two factors behind the improvements: immediate feedback and total accountability—plus, of course, a language his students can understand.

Meanwhile, in rural West Plains, MO, regular assessments have helped change a culture of low expectations—and a pattern of low outcomes—at Glenwood School District R8 (which consists of a single school, Glenwood Elementary School). "It's a very laid-back area," says Leigh Ann Morrison, Glenwood's counselor/assessment coordinator. "We weren't expecting enough out of our students, compared to what was expected statewide and nationally." And because they weren't expecting much, they weren't getting much. State test scores floundered.

Morrison began to use CTB/McGraw-Hill's Acuity, an online and paper-and-pencil assessment that aligns to state standards, in the district's reading and math classes. Student test scores surged; one third-grade class went from the 35th percentile to the 70th percentile on the Acuity tests. Like Nichols, Morrison cites immediate feedback as a big motivator for students.

"As soon as they do the test, they can click on results," she says. "And the computer's not giving them attitude. The computer's not saying, ‘You should have known this.' They're not worried about what their neighbors did. This program has really been an awesome help for us."

Another state-aligned computerized test that personalizes assessment is the nonprofit Northwest Evaluation Association's (NWEA) Measures of Academic Progress, which tests K-10 students in math, reading, language usage, and science. MAP is designed to accommodate students with special needs: It isn't timed, and the math tests can be read aloud to students who can't read them independently. In fact, the program can read questions aloud in Spanish as well. Recently, the NWEA announced new capabilities for its K-2 assessments, including tools to translate the assessment data into instructional strategies, visual displays of student performance over time, and reports that provide teachers with insights into overall class ability.

Julie Evans, the assistant superintendent at Michigan's Sturgis Public Schools, has been using MAP for three years. About 20 percent of the students in Sturgis' eight schools are English language learners; 35 percent of the district's elementary school kids are ELLs. The district administers tests three times annually in grades 2 to 8. "I think the key is that we have to have specific information about kids' strengths and weaknesses," she says, "and we have to be sure that our instruction is in line with those strengths and weaknesses. Frequent assessment is critical, and it's especially critical for those populations that struggle."

ONE FOR ALL, ALL FOR ONE

CAN ASSESSMENTS BE DESIGNED TO WORKFOR EVERY STUDENT? ONE SOFTWAREDEVELOPER IS AIMING TO DO JUST THAT.

"The overall paradigm has changed over the last decade," declaresRichard Patz, CTB/McGraw-Hill's vice president ofresearch. Patz, a former high school math teacher with a doctorate instatistics, is working on developing assessments that serve all studentsand don't ignore those with special needs. "The definition of the normused to exclude people who couldn't take a test exactly the same way,"he says. "Now the definition is more inclusive."

Patz describes several different levels of accommodation that testscan make for students of differing abilities. The descriptions can alsobe found in a CTB/McGraw-Hill 2005 publication that Patz had anintegral role in writing, "Guidelines for Inclusive Test Administration".

  • Category 1 accommodations don't influence the students' performance"in a way that alters the standard interpretation of eithercriterion- or norm-referenced test scores." An example is providingstudents with a separate environment in which to take a test.
  • Category 2 accommodations might affect the students' performancebut not what's being measured. An example is givingstudents extra time to take a timed test.
  • Category 3 accommodations are likely to change what's beingmeasured because the accommodations are related to theknowledge, skills, or ability of the student and are specific to thetest content. An example is letting students use calculators formathematics computation tests, which would change what'sbeing measured from the ability to subtract to the ability toperform subtraction on a calculator.

Accessible Assessment

Barbara Shoap, a special ed math teacher at Philadelphia's George Washington High School, says that the level of her students' ability ranges from third to eighth grade. Last February, Shoap began using Apangea Learning's SmartHelp, an online math-tutoring program that can double as an assessment tool. The school as a whole wanted to know which students were in danger of dropping out; assessment was critical. For Shoap's students, it was critical and difficult. But Shoap gave her kids as much control as she could.

SmartHelp has some interesting features, including a real-time tutor that students can instant-message; a voiceactivated function that reads problems aloud, a motivation strategy that rewards student progress with points redeemable for prizes; and content that appeals to students' interests in sports, fashion, and entertainment.

Within a month, Shoap's students were enjoying the program. She noticed that it worked especially well with students who were typically poorly organized. With the online program, they signed in, got a computer, sat where they wanted to, and proceeded at their own pace. "They could open up the laptop and everything they needed was there," Shoap says.

Shoap maintains that the SmartHelp program kept her students "attending, engaged, and progressing, commensurate with the effort that they made to apply themselves within the program. I think that is a realistic definition of success. I was able to be more available to their individual needs, and my relationship to them was both teacher and facilitator."

For many of these students, it's a matter of whether the assessment is accessible: Does it speak their language? Does it allow for their special needs? Does it engage them? "Some of us remain committed to giving these kids some muchneeded skills and some direction," Shoap says, speaking for other beleaguered teachers of special populations. "It's an uphill battle."

Yet the response to these challenges is not totally technological. One of June Weston's experiences demonstrates that. Weston had been given a 17-year-old student who was the most notorious drug dealer in town. The boy, known as "The Mailman" because that's how he delivered his drugs, was assessed at about a third-grade reading level.

At one point, early on in Weston's relationship with him, he physically threatened her. But Weston, usually accompanied by what is best called a bodyguard (such is the volatility of her student population) promptly set the teenager straight. Soon, The Mailman began to trust her. And within nine months—with the technology, with the programs, and with Weston's help—his reading level rose from grade 3 to grade 8. That was two years ago. Today, he's attending classes at a community college and staying out of trouble.

"You can give them all the tests in the world," Weston says, "and it won't do anything unless they believe they can [achieve]. The assessment tells me where you are today. Where you are tomorrow depends on you."

Neal Starkman is a freelance writerbased in Seattle.

This article originally appeared in the 11/01/2007 issue of THE Journal.

Whitepapers