Performance-based or alternative forms of assessment are being hailed by some educators as the solution to problems inherent in traditional, standardized educational testing. Others, however, caution against adopting alternative forms of assessment too rapidly. Gene I. Maeroff, senior fellow at the Carnegie Foundation for the Advancement of Teaching, writes that although alternative assessments may appear fairer and more authentic than standardized testing (the assessment itself is generally thought to measure more closely what is taught and valued in schools), they are fraught with difficulties.
If alternative assessments are to replace or supplement standardized tests, they must be able to be administered more efficiently and less expensively than now seems possible. For one thing, scoring will need to be standardized in order to minimize subjectivity. (Scoring criteria are currently being developed and tested in many states and school districts.)
Maeroff also expresses concern that, under pressure from policy makers to design and use alternative assessment measures, there will not be enough time to establish the validity and reliability of these new assessments or to train teachers in alternative assessment procedures.
Not useful for student comparisons
Maeroff advises educators to consider more carefully exactly how alternative assessments will provide the information they need. Alternative assessment, he suggests, may be a good indicator of an individual child’s prgress, but it may not be useful for comparisons between children. Maeroff describes the merits of alternative assessment in helping teachers design and evaluate instructional activities and in helping students evaluate and improve their own performance. However, he warns that children who score poorly on standardized tests will not necessarily perform better on alternative assessments.
Fairness across cultural and economic groups
In fact, despite what many believe, alternative assessment may not always be the most fair measure of student ability and achievement across cultural and economic groups. Maeroff writes that the purposes of testing need to be established before assessment tools are designed so that the type of assessment can be judged on the basis of its ability to fulfill these specific purposes.
Similar concerns were expressed in a recent article by Robert L. Linn, Co-Director, Center for Research on Evaluation, Standards and Student Testing, University of Colorado at Boulder; Eva L. Baker, Co-Director, Center for Research and Evaluation, Standards and Student Testing, University of California at Los Angeles, and Stephen B. Dunbar, Associate Professor, University of Iowa. They write that information on the intended and unintended consequences of alternative tests and the degree to which performance in non-testing situations must be addressed. The fairness of the assessments for different groups of children must also be established and a basis for judging both the quality and the comprehensiveness of the content coverage needs to be provided.
Linn et al. caution that a test that is highly valid for one use may be totally invalid for another; careful analysis of alternative forms of assessment is necessary if we are to make the right choice. Linn et al. conclude that the alternative assessment movement has succeeded in challenging the educational community to reconsider the standards it holds regarding valid interpretations of any kind of assessment information. They hope this challenge will stimulate discussion of traditional and alternative assessments and the way each might contribute to educational reform.
“Assessing Alternative Assessment” Phi Delta Kappan Volume 73, Number 4, pp. 272-289 “Complex, Performance-Based Assessment: Expectations and Validation Criteria” Educational Researcher November 1991 Volume 20, Number 8 pp. 15-21.
Published in ERN March/April 1992 Volume 5 Number 2