Bucking a Trend, Md. AssessmentsEmerge as Model School-Reform Tool
This is the sixth and final story in an occasional series examining trends in assessment.
In this new era of testing, while other states' assessment programs have run into brick walls--occasionally with fatal results--Maryland has bounced over the bumps, doctored the bruises, and emerged with a system that appears to be strong enough to lead its public schools into the next century.
The Maryland School Performance Assessment Program in many ways resembles the pioneering California Learning Assessment System. But while CLAS was abandoned last year under intense political pressure, the Maryland assessment has progressed.
"They have a good, open-ended assessment system," said Edward Roeber, the director of the state collaborative on assessment and student standards for the Council of Chief State School Officers. "They have good people there, they have taken time to do it, and they haven't oversold it."
And, Mr. Roeber added, "They have gone out and talked to people about why it was important."
Launched in 1991, the school-performance assessment program is the keystone of Maryland's broader plan to improve student and school performance and make the schools more accountable.
The assessment grew out of a system of academic outcomes in which Maryland spells out what it expects of its students. By basing the assessment on high achievement levels, Maryland officials believed the M.S.P.A.P. would elevate curriculum and instruction.
"There are a lot of educators out there who think that is a bad idea, but we have demonstrated that you can model good instruction through assessment," Steven Ferrara, the director of state assessment, said in a recent interview at the Maryland education department here.
"But," he added, "the assessment alone cannot do the job."
The other central elements are teacher participation and professional development. Teams of teachers create the assessments, devise the scoring guidelines, and score the tests in consultation with education department personnel.
"The process is labor intensive, but the labor is all Maryland teacher labor," said Mark Moody, the assistant state superintendent for planning, results, and information management. "What that does for us is get them to buy into it and to think about things differently," he said.
The Maryland program is part of the new breed of measures known as performance assessments. Although this type of test varies considerably, students typically are asked to execute tasks and answer questions, generally in writing. There are no bubbles to be filled in on multiple-choice sheets. Nor is there one correct answer to most of the questions. Sometimes the tasks are carried out in groups and sometimes individually.
Half the states had performance assessments as of the 1993-94 school year, though such tests have faced increasing criticism in recent months. (See related story.)
Repeating a memorized historical fact or mathematical formula is insufficient. Maryland students are expected to be able to explain events, analyze situations, and solve problems.
The tests are administered for a week each spring to 3rd, 5th, and 8th graders who are assigned randomly to groups in their schools and tested in reading, writing, language usage, mathematics, science, and social studies. Some tasks are confined to a single subject; others integrate disciplines.
The assessment is designed to measure school performance rather than student performance, although the state is conducting a pilot to determine the best and most accurate way to present parents with meaningful data about their children.
However, another, perhaps tougher, challenge may lie ahead. The state is trying to develop high-stakes performance assessments for high school students, whose diplomas are likely to hinge on how they do. Last week, a task force recommended to the state school board that the tests be given at the end of classes in such core courses as English, math, science, and social studies. The assessments, which would be mandated beginning with the Class of 2002, might also include student portfolios.
Now, the major consequence of poor performance for schools "is the threat of state intervention," said Ronald A. Peiffer, the assistant state superintendent for school and community outreach. That fate has befallen five Baltimore schools.
Many problems surfaced in the first year or two of the Maryland assessment. Teachers were inadequately trained to administer the tests. Because examination packets were withheld until immediately before testing began, teachers were unfamiliar with the materials, and experiments went awry. Some test questions proved faulty. And the children were unaccustomed to taking such tests.
Joan L. Goodwin, an elementary-reading specialist in the Title I program in Allegany County in largely rural western Maryland, said 3rd graders there "literally cried because they couldn't express what they were thinking."
The problems all showed up in the district report cards that the state issues each December. Even wealthy districts with nationwide reputations for excellence, such as the systems in Montgomery and Howard counties in the Washington suburbs, could not, as a whole, meet a single standard.
James R. McGowan, the associate superintendent for instruction and school administration in Howard County, said his district is making steady gains. "There are a few [schools] that have met them and exceeded them."
Many of the problems have been corrected. Training has been intensified, teachers get the exams earlier, and assessment tasks are screened more thoroughly. And, after five years, students are more comfortable with the process: None of the students Ms. Goodwin encountered this year cried because of the tests.
Test scores have risen, but there still is not a district in the state that has met a standard, although more schools have improved their scores. Had many of the schools already met the standards, said Mr. Ferrara, "we would have failed our mission."
Complaints from the public have decreased, also, Mr. Peiffer said, because people now have a better understanding of the assessment and more teachers and administrators have been drawn into the process.
Mr. Ferrara said: "One of the things that we have managed to do is avoid some of the controversies that other states have gotten caught up in. We [bypass] topics you can deal with in a classroom that you can't deal with in a state assessment."
Maryland officials learned their lesson after a 5th-grade task about constitutional rights incorporated a political cartoon depicting a movie marquee promoting an all-nude dance review.
What remains to be seen is whether the M.S.P.A.P. is doing what it is intended to do--improve student achievement through curriculum and instruction reform.
The federally financed National Reading Research Center, a consortium of university researchers, has found evidence suggesting that the program has had consistently positive effects on classroom practice. Yet, researchers from the University of Maryland and the State University of New York at Buffalo tempered their remarks because they studied schools that were considered exemplars of instructional change.
In another study, the researchers found a number of barriers to implementation: lack of time and resources, unclear communication and test materials, and an inadequate fit between the model of good instruction and assessment and teachers' beliefs and practices.
The study was conducted in schools that had shown relative success and a positive reaction to the state's mandate.
"If this is the stuff that is being encountered in those schools, it might be a little more difficult in districts that don't feel they have enough resources or teacher cooperation," said Peter P. Afflerbach, a principal investigator at the research center and an associate professor of curriculum and instruction at the University of Maryland.
Fewer Blanks in Booklets
Those involved in the program acknowledge mixed results.
Last spring, Gail Lynn Goldberg noticed a test about Martin Luther King Jr. on a teacher's desk. It posed such questions as, In what year did Dr. King deliver his "I Have a Dream" speech? Ms. Goldberg, the state's test-development specialist, would have preferred a question that asked students how their dreams were similar or dissimilar to the civil-rights leader's.
Over all, though, Ms. Goldberg said there has been tremendous progress that does not always show up in the published data--an observation made repeatedly by teachers scoring tests last month at Randallstown High School in Baltimore County.
Numerous teachers reported such encouraging signs as fewer blanks showing up in test booklets and more reasoning in the responses. And the teachers said that they had changed their practices.
Others are less optimistic. Though a booster of the program, Doralynnsic Jacobs, an elementary school librarian, said many teachers in her school have been resistant to changing the way they teach.
In a recent phone interview, Ms. Goodwin of Allegany County said that many of the teachers in her district are veterans who were not eager to change. Now that they realize that the state assessment is not a flash in the pan, however, they have acquiesced.
"I don't know of many people who do like the test," Ms. Goodwin said. In part, she said, that sentiment is due to the realization that unlike multiple-choice examinations, "you can't help the children pass the test." But, she continued, teachers can help them prepare for the test, beginning in the earliest grades.
"We have to teach children to think, which is a lifelong skill."
The "Review Session" series is made possible by a grant from the John D. and Catherine T. MacArthur Foundation.