The use of state-mandated, high-stakes testing recently has expanded and led to debates about the utility of such testing.
On one hand, many policymakers think schools and educators must be held accountable through the use of test scores. On the other hand, some educators have advocated for the discontinuation of all state testing.
We believe there is a middle ground.
State test scores can provide useful information by shedding light on student performance, identifying achievement gaps and indicating shortcomings in curriculum and instruction.
Test scores are useful, however, only under certain conditions. But even under the best conditions, test scores provide only a small slice of information about what students know and can do and should never be the main focus of educators or policymakers. Rather, such scores should be used for diagnostic purposes.
Under what conditions, then, are state test scores useful?
First, test scores are useful only to the extent that teachers do not “teach to the test.”
Specifically, they are useful if teachers do not focus instruction on specific items thought to be on the test; and if teachers do not teach students how to find the correct answer using test-taking strategies. Both of these approaches defeat the goal of students being able to transfer their learning to future experiences.
Unfortunately, the use of test scores to identify schools — and now teachers and principals — as effective or ineffective creates a strong incentive for teachers to teach to the test.
Although the State College Area School District is working to eliminate this practice, we have no control over other districts. Thus, state test scores cannot be used to accurately compare SCASD to other districts.
Second, tests results must be provided to educators in a timely fashion to be useful to educators.
In Pennsylvania, scores are not returned until the fall after testing, thus are of limited use.
Third, test scores should be used appropriately by the state and districts. Although SCASD endeavors to use the scores appropriately, researchers have concerns about how the state reports and uses the test scores, particularly with respect to identifying educators and schools as effective or ineffective.
Unfortunately, for these and other reasons we will discuss at the forum, we think state test scores currently provide little useful information to educators and the public. Indeed, the only useful information they provide is limited evidence about strengths and weakness of the curriculum.
This is how SCASD currently uses the scores.
Changes to the state’s approach to testing could make the state test scores a useful complement to the rich information SCASD currently gathers about students and teachers.
Most importantly, the state could remove the high stakes from the testing so that scores could be used for diagnostic and informational purposes.
Finally, we think all schools should appropriately assess all types of student learning to support teachers’ efforts to develop more responsive and respectful learning experiences.
We hope everyone will join us in this very important discussion.
Ed Fuller is an associate professor in education policy studies at Penn State. Bob O’Donnell is the superintendent of the State College Area School District.