The arguments we hear against annual standardized exams typically revolve around students being tested too much and testing harming education. But...
The arguments we hear against annual standardized exams typically revolve around students being tested too much and testing harming education. But a look at evidence suggests something different: Annual state tests do not use much time. That’s especially true when you distinguish them from benchmark tests, which districts often give on their own to periodically assess student performance.
The evidence tells us state tests, which must be administered annually under federal law, don’t take much time. Data from 12 urban districts show that state tests take between 4 to 15 hours a year for third graders and between 4 to 12 hours a year for seventh graders. These figures are for math and English/language arts tests, the only ones required under federal law.
The same study of 12 urban districts looked at benchmark tests, which are not part of any federal requirement. The study found that districts spend between 3 to 15 hours on district tests of third graders and 2 to 14 hours for seventh graders. Another study reported that in one Kentucky district, students were tested 20 times in a year. The district required 16 of those tests.
When time for the two kinds of tests is combined, districts range from about 8 to 25 hours of testing a year. Those numbers are about the same for third graders and seventh graders. This is not a lot of time compared to a year of instruction, which is about 1,000 hours.
The study also reported time teachers spent preparing for and administering tests. On average, teachers estimated spending about the same amount of time preparing for and administering the test as the test time itself.
Add all this up, and teachers spend about 50 hours annually on state and district tests. They spend about 16 hours at the low end. Against the backdrop of a 1,000-hour school year, this is not much time.
A word here about benchmark tests: Part of their popularity can be traced to the notion that they help districts do better on state tests. Districts believe that if they orient teaching and learning around benchmark tests, they will do better on state tests.
But there is not much evidence that benchmark tests help districts perform better on state tests. A study of four commonly used benchmark tests found there was no evidence that a student’s score on one could predict the student’s score on a state test. Similarly, a study in Massachusetts compared how schools performed on the state’s math tests when schools did or did not use benchmark testing. Scores on the Massachusetts state assessment were nearly identical.
Aside from using benchmarks to improve state test scores, districts also may use them to guide teaching and instruction. The tests give teachers data about which students are progressing and in which areas they might need support. Some curricula and textbooks that districts can purchase include these tests as part of a package.
In short, benchmark tests are the work of educators, seen in some ways as a support for teaching and learning. But annual state tests are part of accountability systems, which are how taxpayers, parents and schools can determine the performance of each campus and district. When we hear someone decrying tests, it is really important that we know which tests they are talking about.
We all recognize that education has many facets and not all of them are amenable to scoring on a test. But we need to know something about how schools are performing. And information from state test scores is objective, in the sense that it does not depend on teacher grading policies like class grades do. The information also is transparent—it’s easily posted and comparable between schools and districts because all of them use the same test.
Reducing annual state testing will significantly impair our ability to gather all that important information. In return, we will save only a small amount of classroom time. In fact, it’s hard to argue that teachers are depriving students of the very best learning activities because of these tests. Annual state testing is going on for at most 25 hours a year.
Keeping annual tests is worth it. Asking whether district benchmark tests are doing what they are supposed to be doing is worth it, too.
Mark Dynarski is a Bush Institute education consultant and president of Pemberton Research in New Jersey.
