Feature

This year, the No Child Left Behind Act of 2001 moves closer to realization as schools nationwide gear up for full compliance by 2005. Likely of particular interest to psychologists and other educators will be schools' handling of the legislation's mandate that each state administer standardized tests to third- and eighth-graders in language arts and mathematics; science knowledge must be added to the mix in 2007.

While some psychologists oppose such mandated tests outright, others believe they could be beneficial if well and fairly implemented.

The fact is, even as opponents continue to warn about the tests' potential for limiting creativity, teaching to the test and other potential pitfalls, schools are faced with the challenge of instituting them. That's where psychologists and other testing experts can step in to help--advising schools on ways to minimize such dangers and to test students fairly and with minimal negative repercussions, says school psychologist Ron Palomares, PhD, APA's assistant executive director for policy and advocacy in the schools.

"The ultimate goal has always been the same--make sure kids are learning," Palomares says. "But it becomes very complex with high-stakes tests; many different factors need to be considered."

Getting the testing combination right requires careful attention to which tests are used and in what context the results are interpreted, explains psychologist Eva Baker, EdD, director of the Center for the Study of Evaluation at the University of California, Los Angeles. "I agree with the goals of No Child Left Behind as they are broadly defined," Baker says. "But I've been concerned by the degree to which the tests that are actually used have the psychometric properties that allow them to be sensitive to the contributions of instruction. We need to investigate the validity of the measures put in place so that they are really reflecting the things that schools can legitimately do."

Baker also points out that if a test is inadequately sensitive to the improvements possible through instruction, it can understate the effect of even the best teaching techniques. This is especially important, she says, when, through accountability measures, test results have a direct effect on teachers' salaries, funding and even on keeping their jobs. "Tests should be chosen and designed so that if you teach well, it should show up on the test without the curriculum being completely focused on the test itself," Baker says.

Challenges and pitfalls

Baker's fear that testing won't fairly identify good teaching is especially real given the fact that school faculty will be evaluated on the basis of the test results. States will be able to force staff changes if they see inadequate improvements, and parents will have the right to choose schools based on test scores.

A number of psychologists, including Howard Gardner, PhD, warn that the testing legislation, however well-intentioned, will ultimately do more harm than good. Gardner says it will drive high-quality teachers out of public schools and won't miraculously raise test scores or narrow the achievement gap. Also problematic is a lack of federal and state funding to fully implement curricula that should accompany the tests, he says.

"Since the mandates for equipping teachers to prepare students for the tests are underfunded, the whole impetus for the legislation is being undermined," Gardner says.

Aiming for fairness

Testing should serve a number of purposes, says Janet Wall, EdD, co-chair of the Joint Committee on Testing Practice (JCTP), which includes APA and six other professional and scientific associations. For example, testing provides a snapshot of student progress that can identify problem areas in an individual's learning and encourage specific work toward improvements. It also shows instructors and administrators where students' strengths and weaknesses lie, allowing them to make teaching adjustments that build on those strengths and address the weaknesses.

What's critical to making test legislation achieve its goal of improving education is assuring a high degree of alignment among the state standards to be reached, the material that is taught and what is tested, Wall says. With those pieces coordinated, she says, accountability measures are not more fair, but will improve the overall educational landscape.

"The positive aspects of assessment far outweigh any negatives if technically strong tests are selected wisely and the results are used properly by competent professionals," Wall says.

To tackle the challenges of the No Child Left Behind legislation's assessment provisions, the JCTP has revised its Code of Fair Testing Practices in Education, which was first published in 1988--APA is reviewing the revised code as part of its endorsement process. "The code was developed to be a technically comprehensive document, designed to be a straight-forward and easy-to-use resource for professionals involved in all levels of educational test development and administration," explains Marianne Ernesto, PhD, APA's director of testing and assessment and staff liaison to the JCTP.

Geared for the general audience of teachers, administrators and test developers, the new version of the code was designed to reflect the revised Standards for Education and Psychological Testing, which was published in 1999 by the National Council on Measurement in Education, the American Educational Research Association and APA, Wall says. Those standards spell out for psychologists the best practices for giving and developing tests.

The new code offers expanded consumer information in the areas of validity, accommodations for persons with disabilities and rights and responsibilities of test takers. Also included are suggestions on how to best manage test selection, test administration, test scoring, test result interpretation and the rights of test takers. Although it has been approved by some JCTP member organizations, the revised Code of Fair Testing Practices in Education is currently undergoing APA's governance approval process for possible endorsement. The revised code, under consideration by APA, can be found at revised code.