Using College Admissions Tests in State Accountability Systems is a Mistake

 

Assessments can carry significant consequences for students (e.g., graduation exams), educators (e.g., teacher evaluations), and schools (e.g., school accountability), and whether one supports large-scale student testing or not, the research is clear that high stakes assessments impact the content educators cover and the instructional materials they use in their classrooms. Research on standards-based initiatives found that teachers tend to focus more on tested material and spend less time on the untested standards. 

n a 2012 survey of teachers, almost two thirds of those surveyed said that they skipped important topics to cover curriculum covered by assessments.2More recently, in a study of teachers in states that adopted both college- and career-ready standards and one of the consortia assessments designed to measure those standards, researchers found that 82 percent of mathematics teachers and 72 percent of English teachers changed more than half of their instructional materials in response to changes in their state standards, and “85 percent of teachers increased the number of writing assignments covering specific standards and the content included on the consortia assessments.” 

In an environment where assessments influence what is taught in the classroom, it is important that states and districts select high-quality, well-aligned assessments. States and districts must often make choices about assessments based on resources, time constraints, and other outside pressures. There has been a significant movement in the past several years towards states opting for college admissions tests, specifically the ACT or SAT, in place of a traditional summative assessment developed by the state. 

Fourteen states will administer either the ACT or SAT in 2017-18 as the statewide summative assessment for high school. Even more states are currently considering a change. While many states have historically administered the ACT or SAT to all students and included these assessments as part of a college readiness indicator (and many continue to do so), the adoption of a college admissions test as the statewide summative assessment for accountability purposes is a new development for most states. 

Using a college admissions test as the statewide summative assessment is an attractive but risky option for some policymakers and parents. These assessments are used for admissions by nearly all higher education institutions, are shorter in length than most state-designed and consortia assessments, have brand name recognition, and are known for predicting first-year college performance. 

However, notwithstanding their appeal and instrumental value for college admissions, neither the ACT nor College Board, the developer of the SAT, developed these tests as measures of how well students are meeting state mathematics and English language arts (ELA) standards, which is the primary purpose of state accountability tests. When they are used as a state’s mathematics and ELA tests — when they “count” for schools, educators, and students — there is the greatest likelihood that they will drive classroom instruction more than state standards do.

This report looks across the current evidence available on the two primary college admissions tests in order to provide state leaders and policymakers with the information they must consider in selecting high school assessments.