January 3, 2003

New Research Casts Doubt on Value of High-Stakes Testing to Improve School Performance

TEMPE, Ariz. — The high-stakes testing policies adopted by many states and the new annual student testing required by the federal government in the 2001 No Child Left Behind legislation may be counter productive, according to two studies conducted by the Education Policy Studies Laboratory at Arizona State University for the Great Lakes Center for Education Research and Practice, a Michigan-based think tank.

The two reports, “The Impact of High-Stakes Tests on Student Academic Performance: An Analysis of NAEP Results in States With High-Stakes Tests and ACT, SAT, and AP Test Results in States With High School Graduation Exams” and “An Analysis of Some Unintended and Negative Consequences of High-Stakes Testing” are the first in what will be a series of annual reports on the impact of high-stakes tests.

Arizona State University researchers Audrey L. Amrein and David C. Berliner conducted both studies.

“The impact high-stakes tests and high school graduation exams have on academic achievement is, at best, ambiguous,” according to Amrein. She adds: “Contrary to popular thought, high-stakes tests do not increase academic achievement. Instead, after states implement high-stakes tests, academic achievement continues to look much like it did before high-stakes tests were im-plemented. In addition, negative or unintended consequences emerge as students, teachers, and schools attempt to reconcile learning and the attachment of serious consequences to test performance.”

According to co-author David Berliner, “The relative failure of high-stakes tests to achieve their intended purpose and their numerous negative consequences must be considered as America prepares to launch a massive testing program in the effort to improve our schools.”

Impact on Student Academic Performance

“The Impact of High-Stakes Tests on Student Academic Performance” looked at data from 28 states where high-stakes testing programs are already in place. These programs include tests that students must pass in order to advance to the next grade, and graduation tests that students must pass in order to receive a high school diploma, regardless of their performance in the classroom.

Results from these states’ high-stakes tests were compared with the performance of students from those states on other widely recognized measures of student achievement: the National Assessment of Educational Progress (NAEP), American College Test (ACT), Scholastic Aptitude Test (SAT), and Advanced Placement (AP) assessments. The study found that these measures showed no systemic evidence of improved achievement after states implemented high-stakes testing programs. States that implemented other high-stakes performed “much like the rest of the nation” on the ACT, SAT, AP, and NAEP tests, Amrein and Berliner found.

The Amrein-Berliner data also suggest that in states that implemented high-stakes graduation exams, academic achievement appeared to decline. According to Amrein and Berliner, ACT, SAT, and AP scores fell in states that implemented high-stakes graduation exams.

The Amrein-Berliner analyses suggest that, as indicated by student performance on independent measures of achievement, high-stakes tests may inhibit the academic achievement of students, not foster their academic growth.

The researchers found that when state high-stakes test scores rise, it is likely as a result of student training that focuses on taking the tests. They conclude “such training does not appear to have any meaningful carryover effect when assessment of student learning is made on the independent measures of achievement that we used.”

Unintended and Negative Consequences

“An Analysis of Some Unintended and Negative Consequences of High-Stakes Testing” examined the unintended consequences of high-stakes tests in 16 states that have implemented high-stakes graduation exams. In those states, Amrein and Berliner found increased dropout rates, decreased graduation rates, and higher rates of younger people taking the GED equivalency exams.

Examining news reports and qualitative data, the pair also found associations between high-stakes testing and:

 Higher rates of retention of low-performing students in grade in years before high-stakes tests were administered, possibly to better prepare students to take high-stakes tests;

 Higher numbers of low-performing students being suspended before testing days, expelled from school before tests, or being reclassified as exempt from testing because they are determined to be either Special Education or Limited English Proficient (LEP)—“all strategies to prevent low-scoring students from taking high-stakes tests,” the authors noted;

 Reduced offerings in art, music, science, social studies, and physical education—all subjects that are less often tested;

 Higher numbers of urban school teachers, in particular, “teaching to the test,’ limiting instruction to only those things that are sure to be tested, requiring students to spend hours memorizing facts, and drilling students on test taking strategies;

 Increased flight from public schools by teachers who sought to escape state testing mandates, “because state rules make them feel compromised as professionals,” according to Amrein and Berliner;

 Instances of cheating by teachers and other school employees under pressure by high-stakes testing programs.

Amrein and Berliner conclude that there are enough negative unintended consequences to call into question the value of high-stakes high school graduation exams. Their data indicate that it is quite possible that the adverse consequences of high-stakes tests outweigh the benefits that advocates claim for them.

Return to the Frontpage