This self-described "research study" and following press release are intended to influence the debate over the direction of the reauthorization of NCLB, offering a defense of the current test-based accountability system against the inclusion of "multiple measures." The report presents a review of the research on portfolios in large-scale school accountability systems, concludes that portfolio assessment is severely flawed, and then characterizes portfolios as a proxy for all non-test-based measures of student performance. The report has several glaring weaknesses, however. The literature review cherry-picks two studies, both conducted 13 years ago and, on the basis of those studies, concludes that portfolios are not reliable and are too expensive for large scale accountability systems. Yet other large-scale studies of portfolios – some of which are discussed in one of the two studies that the report itself relies on – come to different conclusions but are not examined or even mentioned. An even bigger problem with this new report (which is repeated in the press release), however, is the author’s decision to present portfolios as somehow representative of all non-test-based measures of student performance – which they clearly are not. This results in a document that is of little value for research or policy development.
Note that an "Issue Brief" (press release), published by the Lexington Institute after the main report, is available at:
Mathis, W. (2007). Review of "Portfolios: A Backward Step in School Accountability." Boulder and Tempe: Education and the Public Interest Center & Education Policy Research Unit. Retrieved [date] from http://epicpolicy.org/thinktank/review-portfolios-a-backward-step-school-accountability