Standardized tests have been a staple of American education for generations. They play a role in determining everything from student retention and acceleration to assignment for remediation services and eligibility for gifted classes. They influence academic awards and post-secondary scholarships. Standardized test scores can even determine state-assigned grades for school rankings and make or break school reputations.
Given the high-stakes consequences associated with standardized test scores, we might assume that these assessments provide an accurate and comprehensive measure of learning performance and potential, offer predictive power for future success, and reflect equitable opportunities for learners regardless of background, economic status, and culture. We might think that schools with high test scores must be particularly effective, while schools with lower scores must be deficient in some important ways and thus deserve shame, blame, and intervention.
Yet, when we consider how standardized tests are constructed and administered, what they are designed to measure, and what their results can offer, a very different picture emerges. Let’s explore five ways in which learners, our communities, and we ourselves can be misled by over-reliance on test scores as measures of learning, performance, and potential.
First, standardized tests are designed to capture narrow information about certain academic knowledge and skills. While they can provide some information on student knowledge in subjects such as math, science, and English, standardized tests do not measure other important areas of knowledge and skill; they typically ignore student abilities to solve complex problems, think creatively and critically, or show artistic ability. Yet these areas of knowledge, skill, and potential are highly valued and necessary for success in future careers—and life itself.
Second, year-to-year comparisons of test scores typically do not reflect learning growth, as they do not compare the same group of students from one year to the next. Consequently, shifts upward or downward in grade-level scores are comparisons to the performance of the previous year’s group of test takers, so while the test questions may reflect consistent grade-level learning expectations, different students are responding to them. Year-to-year comparisons alone tell us little about how the learning of any group of students is growing, static, or falling behind.
Third, standardized test score comparisons can mask the work of some effective schools and overstate the performance of other, less effective schools. Typically, schools in more affluent areas attain higher test scores. Not surprisingly, the students they serve typically come from families that have a history of success in formal education and are able to equip their students with robust background knowledge and experience. Test scores in these schools can be as heavily influenced by what students bring with them to school as by their school experience. Meanwhile, schools serving less advantaged students may be generating higher rates of learning but are still achieving lower test scores. In fact, a recent study conducted at Stanford University found that when test scores were compared with a focus on the amount of student learning growth generated, lower-scoring schools in large metropolitan school districts frequently demonstrated more learning growth than schools serving more affluent and advantaged students in suburbs and elsewhere.
Fourth, efforts to build students’ learning engagement, ownership, independence, and efficacy are rarely measured by standardized tests. Yet, these skills can have a determinative influence on students’ ability and commitment to be lifelong learners in addition to heavily influencing their success beyond formal education. As students are focused on developing these key learning and life skills, their standardized test scores often do not grow as rapidly as those of students in learning environments with a heavy focus on test scores, even though over time their learning may serve them far better.
Fifth, standardized test scores are not highly predictive of future college success. In fact, grade point average (GPA) is a far better predictor of the likelihood that students will succeed in higher education than the most popular college entrance exams. A recent study involving Chicago Public School students found that GPAs are five times stronger predictors of college success than the test scores on which colleges have relied for decades.
Standardized tests can play a role in collecting and reporting information about student academic performance, but they are far from infallible, and they rarely offer a comprehensive picture of what and how well students are learning. Obviously, there is a need to reduce our reliance on scores associated with standardized tests. We also need to broaden our perspective and begin collecting more learning-related information to gain a more complete—and more accurate—picture of how our students are learning and growing.