I am glad President Obama has admitted that there is too much standardized testing in our public schools. He has produced a “Testing Action Plan” to address this issue. He wants our testing policies to be smarter, and he and his advisors have come up with a set of guiding principles to that end. Every one of the principles is both something I agree with and something the administration’s policies directly contradict, but it would take hundreds of thousands of words to address all that, so I will focus on just the first principle, that tests must be “Worth Taking”.
This, in fact, is the essential issue. The fact that there are too many tests is just a symptom. One sentence in the definition of this principle gets to the heart of the matter: “And assessments should provide timely, actionable feedback to students, parents, and educators that can be used to guide instruction and additional supports for students.” Indeed.
A problem with any large scale standardized test is that the results are never immediately available. The further the test departs from a pure multiple choice model, the longer it takes to score, and the greater the time gap becomes. The results of the standardized tests currently in use are generally not available for months. Even someone who has never taught should be able to understand that “feedback” that is not received until the next semester is by definition neither “timely” nor “actionable”.
On top of that, for most of these tests, students and teachers are not allowed to see which questions were answered incorrectly. There is no way to know whether the student actually wasn’t able to answer the question, or just misread it, or filled in the wrong bubble on the answer sheet, or was too tired to concentrate by that point. A test score is just a number. It is of no instructional use without detail and context.
Let us drop the facade that these tests are intended to be of any use to the students who take them or their teachers. If they are of use to anyone, it is policymakers and high level administrators who want some way to compare the performance of students, teachers, and schools. That is a perfectly understandable goal, and I actually support it. I just don’t think these tests are the best tool for achieving it.
We already have a more reasonable test for this purpose, the NAEP, which uses sampling and has no stakes automatically attached to it, and for which no one does any test prep. It provides information that is as reliable as most of the best standardized tests, which is to say, fairly reliable. I am not aware of any standardized test that has ever been proven to provide truly reliable information about anything other than socioeconomic status.
I have purposely used the word “test” and not “assessment”, because these words are too often used interchangeably. There are many kinds of assessment other than tests. Tests and reports and anything else that is essentially paperwork will never give the full picture of what is happening at a school. You have to actually go there and observe and talk to people, and you have to do it for a significant period of time, and more than once, and sometimes without warning. It does not have to be more complicated or more expensive than the thousands of hours and billions of dollars that have been spent on testing companies and consultants in recent years. Similar things have long been successfully done in countries such as the UK.
Mr. President, the problem is not that there are too many tests. It is that the tests, by your own definition, are not worth taking. Why is it so difficult for otherwise intelligent, thoughtful people like you and your advisors to understand that?