Statewide test scores that will be released soon are just one measure of how well schools are doing and should be considered in that context, educational experts said Monday.
The Utah Education Association sponsored a news conference to stress that school-to-school or district-to-district comparisons of the test data without considering many factors could be damaging."We are concerned that the public, legislators and teachers do not fully understand the uses and limitations of testing. We hope to offset some of the confusion," said UEA President Lily Eskelsen.
The scores will be released by individual districts beginning in mid-December, with a statewide report due in mid-January, said David Nelson, state testing specialist. The statewide testing was mandated by the 1990 Legislature and involved children in grades five, eight and 11, who took the tests in September and October. It is the first time all the children in certain grades in the state have taken the same test. The test selected was the Stanford Achievement Test, Eighth Edition. The test is the most recent in the Stanford series and was given to thousands of children in 1988 to establish norms, Nelson said.
Because the test is new, scores can be expected to be somewhat lower than those received on older tests because there is a consistent slight increase in basic skills across the country, he said.
Over time, Utah schools will be able to compare test scores from year to year to see if they are improving.
Several education experts agreed that trying to compare schools without knowing all the factors can be misleading. At the UEA offices Monday, along with Nelson, were James Cangelosi, professor of education, Utah State University; Robert V. Bullough Jr., professor of education studies, University of Utah; Robert Ellison, director of the Institution for Behavioral Research; and John Ross, specialist in measurement and assessment at the State Office of Education.
Achievement tests such as the SAT allow for a loose measurement, in essence, of what a child has had the opportunity to learn. It is a narrow range of information at best, said Bullough. They do not tell whether a child has learned to think or to write, for instance. They tell nothing about attitudes and values, which are important elements of education.
The United States has "a fetish with testing," he said, that makes high test scores an end in themselves without contributing anything substantially worthwhile to education.
Cangelosi said the achievement tests used in schools throughout the country are skewed in favor of children from higher socioeconomic circumstances and don't take into account the fact that poor children tend to be more creative.
While poor children are not inherently less intelligent than rich children, there are a number of factors that mitigate against their success in school, he said.
Utah's test scores are expected to show differences related to socioeconomic and other factors. In anticipation of that, the State Office of Education has prepared a range of "expected scores" that take some of these factors into account. A school whose scores fall within the expected range can be considered to be doing well, even if they are below national norms.
Test scores do not necessarily identify good or bad teachers but may create a tendency to teach the test materials to the exclusion of more important things.
Ellison said the tests can be a "powerful tool for improving the performance of children, the schools, school districts and the state" if they are used to identify strengths and weaknesses and to design more effective curriculum.