ABSTRACT
Promoters of federal education policy assert that high-stakes testing leads to increased achievement (Ravitch, 2002a) and that “No Child Left Behind Is Working” (U.S. Department of Education, 2006, original emphasis). Th ese promoters argue that student achievement is improving generally and that achievement gaps between Whites and other students are closing because of high-stakes testing used in policies like NCLB. However, such arguments do not necessarily match the data. Analyses of NCLB have found that the high-stakes testing has not improved reading and math achievement across states and has not signifi cantly narrowed national and state level achievement gaps between White students and students of color nor gaps between rich and poor students (J. Lee, 2006; Mathis, 2006). Additionally, studies which compare test score results of the Texas high-stakes testing system (the blueprint for NCLB) with four other standardized tests commonly taken by students in Texas and nationally-the American College Test (ACT), the Scholastic Achievement Test (SAT), the National Assessment of Educational Progress (NAEP), and the Advanced Placement
(AP) test-fi nds that the Texas high-stakes testing in early grades either was indeterminate, had no impact, or had a negative impact on student achievement in relation to the other tests. Th us, the implementation of high-stakes testing in Texas could not be shown to improve learning (Amrein & Berliner, 2002b; Nichols, Glass, & Berliner, 2006).1 Th ese fi ndings correlate with other research nationally (Harris & Herrington, 2004) and raise the question of the overall eff ectiveness of high-stakes tests in improving achievement generally.
