Publications & Resources

A First Look: Are Claims for Alternative Assessment Holding Up?

Dec 1994

Joan L. Herman, Davina C. D. Klein, Tamela M. Heath, and Sara T. Wakai

Drawing on data from student surveys, demographic data, interviews with students and teachers, and structured classroom observations of students, CRESST researchers studied teachers and students who participated in the 1993 California Learning Assessment System (CLAS) test in mathematics. Among the key findings—alternative assessments stimulate student thinking and problem-solving. Students understand that something different and more rigorous is required in open-ended vs. multiple-choice questions. “This is not to say,” write CRESST researchers Joan Herman, Davina Klein, Tamela Heath, and Sara Wakai, “that students `like’ open-ended items more than multiple choice ones.” In fact, add the authors, students prefer multiple-choice problems, perhaps because they are familiar with these types of problems and because they think they perform better on standardized tests. But the results tend to support the idea that students learn from performance assessments. One of the major research questions was whether or not students in different types of schools have equal opportunities-to-learn (OTL) the material being assessed. Researchers surveyed students on a variety of OTL indicators including if they had adequate access to calculators, opportunities to work on problems that can be solved in more than one way or problems that required them to explain their thinking. Surprisingly, the researchers found that urban school students had equal access to many OTL resources such as calculators and a curriculum that went beyond standard “drill and kill” instruction. More problematic was the finding that urban students tended to have more questions about key concepts in mathematical thinking and less access to current textbooks than their suburban counterparts. Finally, interviews and surveys indicated that suburban students clearly felt better prepared than either urban or rural students for the CLAS assessments. The authors note that their results are preliminary and say that the next part of this study will include actual student performance on the CLAS tests. “We will be looking more closely,” say the researchers, “at the interrelationships among and between student demographics, instructional practices, attitudes, and performance.”

Herman, J. L., Klein, D. C. D., Heath, T. M., & Wakai, S. T. (1994). A first look: Are claims for alternative assessment holding up? (CSE Report 391). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).