The difference between being anti-testing and being anti-today’s testing regime can sometimes get glossed over. But the wide space between the two positions was demonstrated damningly in a paper published this spring.
Written by two vehement advocates for the national tests now under construction, the paper is mainly a blueprint for what a re-imagined national testing system could look like. But it begins with a succinct, damning description of what its authors call our current “testing bind”:
Though no one intended to do so, we have created a testing bind that, as it tightens, drives attention away from the intended standards. The effects are greatest in the poorest schools. The nation’s current approach to raising achievement and increasing equity in the education system is having an effect opposite from the intended one. It is trapping poor children in a basic‐skills teaching program that gives them little chance to acquire the deeper knowledge and abilities we seek for everyone. And it may be lowering the learning opportunities even for many more privileged children as schools turn their energies to the test‐based basic skills program.
The paper, “An American Examination System,” is written by two people who may very well have a hand in shaping the new testing regime: University of Pittsburgh professor Lauren B. Resnick, who helped draft the “common core” standards endorsed by President Obama and many states, and Wireless Generation CEO Larry Berger, whose company is likely to make a bid to build the technological pieces of the national tests that will be tied to those standards.
Among the litany of problems Resnick and Berger describe with today’s tests are the many best practices they don’t meet. Among these is something called “instructional validity.” Here’s their definition:
Assessments are considered instructionally valid when student performance improves after quality instruction on the content of the assessment.
In other words, imagine that a good teacher gives a quality lesson on a certain set of content, leading her students to master it. Then, her students then take a test about that same content. If the test results are good, then that test is considered “instructionally valid.” If the students score poorly, then it isn’t.
It’s no wonder that “instructional validity” is “part of the gold standard for educational testing.” And yet, Resnick and Berger write:
“it is almost never established in current assessment practice.”
Berger and Resnick’s full paper is below.