High-Stakes Testing

Testing in Accountability Systems

The high-stakes testing that is used for educational accountability dominates many of today's policy discussions about testing. Large-scale achievement tests constitute a central component of what is often called a test-based accountability system. Test-based accountability systems typically include four key components: (1) standards that describe what students are expected to learn (often called content standards or academic standards) and the level of proficiency they must demonstrate (often called performance standards or achievement standards); (2) tests that are intended to measure attainment of the standards;(3) targets for performance on those tests (which may be simple cut scores or more complex combinations of information from multiple tests, such as the Adequate Yearly Progress [AYP] measures required under NCLB); and (4) a series of consequences that could [Page 466]include a variety of rewards, sanctions, and interventions. The tests used in these systems typically have several purposes, including motivating educators and students to work harder, shaping instruction to make it resemble the standards, and providing both school personnel and the general public with information about student performance to help promote better decision making. Advocates of test-based accountability argue that attaching high stakes to test scores is essential for ensuring they serve these purposes adequately, whereas detractors argue that stakes distort the quality of information and create incentives for student and teachers to engage in counterproductive behaviors. Some of the research findings bearing on this debate are summarized later in this entry. Accountability testing is distinguished from many other high-stakes testing contexts by the fact that the stakes do not necessarily affect the test-takers but instead are often focused on the educators whose performance is being judged on the basis of students' test scores.

...