Early-Years Educators Out Front in Ditching Traditional Tests
This is the third story in an occasional series that will examine trends in assessment and new ways of measuring what students know and are able to do.
Joyce Love was a tough sell on the changes that swept through Garrison Elementary School here about three years ago.
The new "developmentally appropriate" curriculum was one thing.
But then the school scrapped report cards and traditional tests in favor of a multifaceted assessment system using grade-level guidelines, checklists, portfolios, and summary reports to measure children's progress.
"It was a big departure," the veteran teacher said of the "Work Sampling System"--a method of tracking student progress that is being used in more than 3,000 classrooms across the country. It is widely considered the front-runner among early-childhood assessment models.
Ms. Love soon realized, however, that the new method was spurring other positive changes in the classroom: It challenged her, boosted her children's confidence, and increased parent involvement.
"It's given me an opportunity to look at the whole child, and to document achievement in more varied ways so we can see all the multiple intelligences that a child has," she said.
Over the past decade, educators across all grade levels have criticized the use of traditional multiple-choice tests to measure what a student knows.
But the outcry has been particularly strong among early-childhood experts, who argue that all children deserve a fair start in education, unimpeded by such tests.
Many kindergartners fare poorly on traditional tests because of an inadequate preschool experience, lack of early care, or cultural or language differences, such experts maintain.
In the past several years, many policymakers have denounced the use of pencil-and-paper tests to gauge the abilities of young children, especially when the results are used to place students in remedial classes or force them to repeat a grade.
"Everyone is trying to move to more developmentally appropriate measures," said Lorrie A. Shepard, a professor of education at the University of Colorado at Boulder.
Just how schools are changing their ways, however, is still largely undocumented even by experts in the field.
Ms. Shepard is the co-chairwoman of a task force, created by the National Education Goals Panel under the federal Goals 2000 act, that has been charged with drawing up guidelines for early-childhood assessment.
The 16-member group met in early March to begin researching different models, and it hopes to have a more comprehensive picture of the national scene by fall.
A number of efforts are now under way.
The Ypsilanti, Mich.-based High/Scope Educational Research Foundation, for example, began implementing a system three years ago that uses anecdotal record-keeping to evaluate children's progress.
The Council of Chief State School Officers, meanwhile, has created prototypes for early-childhood assessments in mathematics and language arts and is piloting the portfolio-based guidelines in half a dozen states.
Work Sampling System
But clearly the most comprehensive performance assessment for young children to date is the five-year-old Work Sampling System.
Developed by Samuel J. Meisels, a professor of education and an associate dean for research at the University of Michigan, the system's principal function is to improve instruction.
Assessment is ongoing and integrated into daily classroom life, rather than sporadic and intrusive. "This is a terribly important distinction," Mr. Meisels said.
Most tests--even those that are performance based--require students to demonstrate their knowledge "on demand." To show their knowledge of electricity, for example, students might be asked to draw pictures of circuits.
By contrast, Mr. Meisels said, teachers using the Work Sampling System are less concerned about electricity per se and more focused on how children come to understand scientific principles.
"We'll ask teachers to examine how children pose scientific questions and offer explanations of physical phenomena," he said.
Geared toward children in preschool through 5th grade, the Work Sampling System covers seven curriculum areas: personal and social development, language and literacy, mathematical thinking, scientific thinking, social studies, the arts, and physical development.
Evaluation across these domains consists of three interrelated elements: developmental guidelines and checklists, portfolios of children's work, and summary reports.
The guidelines provide teachers with a realistic picture of what students can be expected to know and be able to do at each grade level. The purpose is to enhance teachers' observational skills and enable them to make judgments based on a set of professional criteria.
The guidelines for literature and reading, for example, suggest that 1st graders should listen with interest to stories and other texts read aloud. (See related story.)
Three times a year--in the fall, winter, and spring--teachers fill out checklists that track a child's progress across the different domains.
The 1st-grade checklist for social studies, for example, asks teachers whether students can recognize that they and others have shared and different characteristics, and if students show a beginning understanding of why rules exist.
Children then play an active role in choosing work to include in the portfolios that document their progress.
Teachers complete the assessment package by writing a summary report, which essentially is a detailed portrait of a student's strengths and weaknesses.
Not Just a Fad
Schools in Boston, Cleveland, Phoenix, Pittsburgh, Washington, and a host of other districts have already implemented Work Sampling. The Philadelphia district expects Work Sampling to become a focal point of its $50 million Annenberg Challenge Grant.
Pending legislative appropriations, the state of South Carolina has agreed to pilot the system in more than(See education officials in Illinois, Vermont, and Oklahoma, to name a few, are considering large-scale implementation in their states.
This summer, about 700 teachers employed by the federal Bureau of Indian Affairs will be trained in Work Sampling tenets.
The growing use of the Work Sampling System is not just a fad, its authors say. They emphasize that it is a method based on "best practices" and years of research.
Usage varies slightly from place to place, as districts tend to adopt the changes gradually and adapt them to their own needs.
At Washington's Garrison Elementary School, Work Sampling--along with a home-brewed version of the summary report--is the assessment tool in classrooms from Head Start through grade 3.
The key to the system is that it "doesn't compare children against other children or against some national standard, but against their own growth over some period of time," explained Austine Fowler, a consulting teacher at Garrison. The school is one of 52 schools in the city that are in various stages of adopting the program.
Schools that sign on to the method are encouraged to prohibit letter grades, standardized tests, and retention.
What does a teacher do, then, with a child who is clearly behind his or her peers?
"What we tell people is that it shouldn't happen like that," said Mr. Meisels. Teachers using the approach should be acutely aware of a child's weaknesses throughout the year, and the task would be to work on those areas, he said.
In Work Sampling, "there are no surprises," he added. It is "a method for staying in touch."
Parents appear to applaud the system. In an informal survey of parents in several Pittsburgh schools, 81 percent said they felt positive about the assessment. Only 8 percent did not like it, while the rest were undecided or gave no response.
"We had a lot of parents who wanted grades come hell or high water," said Ms. Fowler, who is also one of 16 specialists serving on the Work Sampling System's "national faculty."
But, she added, many parents have since said, "'Wow, the teachers really know my child,'" or "'This is the first time I've received anything positive about my child.'"
Most of the concerns that parents raised in the informal survey in Pittsburgh involved the workload for teachers and issues of fairness.
But most teachers seem to find the additional work insignificant once they get accustomed to the new system.
"You become more effective," said Stephanie Early Abney, a teacher of multi-age classes at Garrison Elementary School.
"It's also incumbent upon administrators to make time for teachers," Ms. Fowler added. "You just have to be creative."
Research conducted on the Work Sampling System has shown that it is highly reliable as a measure of children's achievement.
In a test involving 100 kindergartners, the system proved to be an accurate predictor of performance on norm-referenced tests, even when researchers controlled for the potential effects of gender, age, and initial ability.
But perhaps the best indicator of success is in the behavior of children.
"They're involved," said Ms. Love, who has seen children come and go from her classroom for 27 years. "They're much more confident about discussing their work and defending it, and have a much better sense of self-evaluation."
The "Review Session" series is made possible by a grant from the John D. and Catherine T. MacArthur Foundation.