Comparisons among Individual and Cooperative Performance Assessments and Other Measures of Mathematics Competence

The purposes of this study were to examine how well 3 measures, representing 3 points on a traditional-alternative mathematics assessment continuum, interrelated and discriminated students achieving above, at, and below grade level and to explore effects of cooperative testing for the most innovative measure (performance assessment). Fourth graders (n = 131) with prior training and experience in cooperative mathematics activities individually completed 2 more traditional assessments; half completed performance assessments individually, the other half in dyads. The 2 more traditional measures, but not the performance assessments, demonstrated strong discriminative validity. Among individually administered measures, intercorrelations were moderate and significant; correlations were stronger for performance assessments that were individually rather than cooperatively completed, and exploratory analyses suggested that cooperative performance assessment scores corresponded better with other measures for above-grade-level students than for below-grade-level students. Findings are discussed for statewide assessment programs and instructional decision making with respect to the misalignment between performance assessment forms of mathematical knowledge and typical classroom curricula and the need for preparation in collaborative group processes on problem-solving activities.