School Chiefs Endorse State-By-State Comparisons
Wilmington, Del.--In a sharp shift of policy, the Council of Chief State School Officers last week called for regular comparisons of the educational performance of the states and urged that national achievement tests and other measurements on which to base such comparisons be developed.
After two days of unusually heated debate, the council also voted to take an active role in deciding what indicators should be used in comparing the states and agreed to establish a center to coordinate nationwide assessment activities.
See related story on page 5.
These were the key recommendations of a report approved by the council by a vote of 27 to 12 at its annual meeting here.
Before adopting the report, the council defeated, by a vote of 20 to 19, an amendment by Robert D. Benton, superintendent of public instruction in Iowa, that would have tabled for a year the controversial section of the report calling for the creation of a national system of evaluation and state-by-state comparisons of the results of such assessments.
Instead, the chiefs voted to make the implementation of such testing and reporting contingent upon a plan to be considered by the council a year from now. Under the plan approved by the council, state participation would be voluntary.
"This is a very, very different direction for the chiefs," said Gordon M. Ambach, commissioner of education in New York and president of the council. "They have endorsed the idea of joining together to decide on and then design ways to evaluate the nation's schools. It's never happened before."
"Two years ago, anyone making such a proposition would have been run out of the room," said Verne A. Duncan, superintendent of public instruction in Oregon.
The report was developed in recentmonths by a committee of the council in cooperation with a number of testing experts, in particular Michael Smith, a professor of education at the University of Wisconsin at Madison.
It notes that the creation of standardized national achievement tests--which would assess a core of knowledge in science, English, social studies, mathematics, reading, and writing and would be developed by the council in cooperation with testing organizations--would take at least five years to put into place.
Among the reasons cited by council members for the organization's change in policy were the need for a wider and more dependable range of factors with which to judge the benefits of the flurry of reforms recently enacted in many states and the need to suggest to the supporters of the reform movement the value of their investment.
"You can't take a negative stance on accountability and assessment or you may sour support amoung those who are providing us with resources, and we cannot afford to do that," Bill Honig, superintendent of public instruction in California, told the council.
"Everyone is looking at [state-by-state comparisons] as a competition," he added, "but we can use it to set national goals--in such things as enrollment in particular subjects and dropout rates--and that is a very powerful, legitimate function."
'Data Serve To Exhort'
Others viewed the council's actions as a way of fostering a healthy competition in the name of improvement.
"By comparing the performance of a school, district, or state with itself over time, or with other schools, districts, or states, data serve to exhort, motivate, or reward," wrote the authors of the report. Secretary of Education Terrel H. Bell made a similar point in releasing his so-called "wall chart" of education indicators last January.
The wall chart, which focused on a comparison of state college-admission-test scores and dropout rates, was repeatedly attacked here for making comparisons among states without taking into consideration demographic and economic factors that influence their educational performance. It was described by many participants as a major reason for the council's actions on the assessment issue, even though the chart was released two months after the council had decided pursue the issue.
"The publication of the 'wall chart' brought to the forefront the issue of state-to-state comparisons," wrote the report's authors. "On a political level, the attention given to the Secretary's wall chart makes inevitable future state-to-state comparisons on outcome measures."
"The wall chart was just the beginning of what's to come," Gerald N. Tirozzi, commissioner of education in Connecticut, told the council in support of the report's recommendations. "And I would rather have accurate, appropriate, and fair measures of comparison than biased, distorted, and inaccurate ones."
Sources within the Education Department say that a second chart is expected to be released in January.
Opponents of a national system of state-by-state measurements, many of them from the South and from states in other areas where students traditionally fare poorly on standardized tests, charged in sometimes impassioned statements that comparative information will only lead to unfair conclusions and inappropriate judgments.
"This is going to be a ranking of where Georgia is on a scale of 1 to 50, and nothing more," said Charles McDaniel, superintendent of schools in Georgia. "The media will ignore the details and I can see the headlines in the Atlanta papers now: 'Georgia: 40th.' Education in the state is going to be hurt by this monstrosity, not helped by it."
"I don't think there is any such thing as an accurate state profile," said John H. Lawson, commissioner of education in Massachusetts. "We have 436 very different school districts and to say that there is a profile of a typical Massachusetts student is just inaccurate."
High Costs Foreseen
Others suggested that the cost and logistics of gathering comparable information on students in 50 states would be overwhelming.
Archie LaPointe, executive director of the National Assessment of Educational Progress (naep), a federally funded program that regularly measures the achievement of nationwide samples of students, said last week that the assessment's $4-million annual budget would have to be six or seven times larger to support testing groups large enough to permit state-by-state comparisons of naep results.
Mr. LaPointe said discussions with Secretary Bell about such a project have led "nowhere," largely because of the cost involved. He added that three Southern states--Florida, Louisiana, and Virginia--have agreed to pilot a program in which naep will test 11th-grade students in reading in each state so scores can be compared. The other Southern states declined to participate in the project, sponsored by the Southern Regional Education Board. (See Education Week, Aug. 22, 1984.)
No funds were allocated by the council at the meeting for the creation of the assessment center or the development of measurement tools.
Other opponents of the report's recommendations warned that states that compare unfavorably on the various measures could suffer economic hardship if potential investors are troubled by the results and take their business elsewhere. The national achievement tests were also challenged on the ground that they would threaten state and local control over education by helping to promote a national curriculum.
"State comparisons of student achievement tests invoke the specter of a single national exam," wrote the authors of the report.
Fears Called 'Unfounded'
Stephen Kaagan, commissioner of education in Vermont and the principal author of the report, said that "fears of a national curriculum are unfounded. Many things we do are prevalent nationwide--such as language development. ... You can't have it both ways. If you say states are responsible for education, then they have also got to do the reporting."
Mr. Kaagan said the key to implementing the report's recommendations is to make fair comparisons among the states. "It is a challenge, but it can be done," he said.
Many council members voted to endorse state-by-state comparisons only if efforts are made to ensure that the system is fair.
"I'm going to vote to shoot myself in the foot and I don't look forward to it," said Carolyn Warner, superintendent of public instruction in Arizona. "I support it only if weighting factors for such things as the number of non-English-speaking students, the percentage of kids drawn off to private schools, and the number of kids below the poverty line, are figured into the comparisons."
The report calls for such factors to be considered when comparisons are made.
Kenneth Hilton, a member of the Delaware State Board of Education and president-elect of the National School Boards Association, during the meeting said of the council's decision to endorse state-by-state comparisons of achievement scores: "Unless we are comparing apples to apples, I don't want to see it."
A representative of a major testing company who attended the council meeting said that "if a national test evolves [from the council's actions], it will destroy the testing industry," because "it would be the only test in town."