Education researcher David Berliner has just written an excellent guest post in The Answer Sheet, a Washington Post education blog. It’s called Why Rising Test Scores May Not Mean Increased Learning.
I’ve posted in the past about Berliner’s exceptional work.
In his guest post, he makes six key points, and he elaborates on each one. I’d strongly recommend you read his entire post. I’m just going to briefly quote each of the six:
1). Virtually all states have changed the passing score on tests so that more children are classified proficient.
2). School districts across the nation engage in excessive, perhaps unethical, and, in some cases, illegal test preparation. This results in higher test scores, but not necessarily greater learning.
3) Familiarity with the objectives and the items on a test invariably results in increased test scores.
4) The test items we use do not tap the knowledge we really want to assess.
5) Afraid they could be fired or their schools closed because of NCLB test scores, district and school administrators invent ways to prevent the poorest performing students from taking tests.
6) It is common for scores to go up because of cheating. For example, there are companies that look for anomalies in test scoring. They often find incidents such as a low-scoring student suddenly getting seven items right in a row, or a class in a low-performing school suddenly outperforming classes in a neighboring high-performing school. These may or may not be instances of cheating, but several hundred of these anomalies are associated with NCLB tests in many states.
How timely! We just spent a half-day in PD discussing data, specifically the test results on the English Language Arts state assessments. I was frankly stunned at the participants all agreeing what a wonderful job the district had done to raise our test scores from year 1 to year 2. HELLO!?!?!?!
I asked that they consider the following possible scenarios:
1. Possible that year 2 assessments were easier.
2. Possible that students just became better ‘test takers’ as they progressed from grades 3 through 8.
3. Possible that the questions asked were better aligned with what was taught in the classrooms.
What I wanted to see was not a general break-down by grades, but rather a break-down by student, and preferably over at least a three year time frame. What types of questions is the student consistently missing? Are the other students in that particular class missing the same types of questions? Are there types of questions that our district are missing that other districts are getting?