In order to use assessment properly, to increase student learning, here are five things we need to keep in mind:
- Use the right tool for the right job. Often we are told as teachers to "use assessment data to drive instruction." The problem is that by "assessment data", those making this demand are talking about state assessments, benchmarks, or diagnostics. You can't make day to day changes that benefit students based on this data. Learning that one of my students scored low in the "geometry category" five months ago on a state assessment is worthless to me compared with the exit card that showed me that he/she didn't understand that area was a two-dimensional measurement. The latter allows me to correct the misunderstanding immediately, thus leading to greater learning.
- I've heard Chris Lehman say before that educational technology should be like oxygen - imperceptible, ubiquitous, and necessary. The same can be said for assessment. We need it and should be using it all the time as a way to guide our students, but if our students are stressed about how they are being graded, ranked, sorted, or judged, they aren't focused on learning. And learning should be our goal.
- "Assessment" and "Grading" are not interchangeable terms. Often they are used that way because we tend to want to make everything measureable. Data doesn't have to be numbers to be useful. Again, learning should be our focus, not ranking or judging students. Tests and quizzes will, for better or worse, always likely have a place in schools. What is more beneficial for students, though: giving them a 30 on a quiz in which they got 7 out of 10 questions incorrect, or sitting down with that student to discuss their confusion and helping them identify ways to learn what they haven't yet? "Grading" is something that is done for the benefit of teachers, parents, colleges, and others. Good "assessment" is done for students.
- Standardized tests, benchmarks, and diagnostic tests are not bad assessments unless we use them in ways for which they were not designed. When we start using data from a benchmark or diagnostic tests to determine a student's placement in basic or advanced math classes or data from student standardized test scores to judge teacher efficacy and school quality we fail our students. Arguments that my car got great gas mileage because it goes from 0 to 60 in 2.5 seconds or that my brother is a great basketball player because he has can punt a football 60 yards would be dismissed as absurd because those aren't valid metrics to use to judge such things. Why aren't the conclusions we are erroneously drawing from bad metrics in education being dismissed as absurd? I believe, as Joe Bower put it so well, we can't measure what's important, so we are putting importance on what we can measure. It needs to stop for the sake of our children. They deserve better.
- We need to do a serious cost/benefit analysis of how we assess students. The assessments that are given the most importance in schools right now are also the most costly in terms of time and money that have ever been given in schools before. We spend billions of dollars as a country each year on the tests themselves, test prep materials, and resources to meet the logistics of administering the tests. We spend weeks of time that could be spent on learning critical thinking and innovation demanding that kids learn test taking skills and low-level thinking facts so that they can pass the tests. And what do we get? Lousy data. Data that is far, far inferior to the formative assessment data I could have collected in much less time and that could have been used immediately to teach students.