There are essentially two types of testing talked about these days: testing to measure and testing to learn. It’s probably a false dichotomy, but let’s assume it’s valid for the moment. Testing to measure is the standard testing process: students receive information, study the information, take a test on the information. Of course, there’s a lot more to the process. A teacher often gives small quizzes along the way in preparation for the test, but the quizzes and the test function in the same way. Students perform, teacher grades, class moves on.
Teaching to learn is different. Students test themselves until they know the information. They can receive the information along the way, sometimes before they begin testing, sometimes after they have tested themselves, but the point is that they are constantly taking tests that provide immediate feedback, usually earning higher and higher scores along the way. You’ve probably heard it called the Testing Effect. Think of it like a computer game with an infinite number of tries. If you don’t pass the level, you play until you beat the level. You don’t have all the information before you start trying to beat the game, but you learn how to beat it by making mistakes, learning from those mistakes, and trying (read “testing”) again.
Both types of testing have their place. Testing to measure is the traditional classroom form, and it isn’t going away any time soon. And that’s okay. Students ultimately need a grade that assesses their performance. Testing to learn may seem new to the classroom, but it’s not at all new to learning. Think of learning to tie shoes. We tried until we got it. The assessment came in seeing how long our laces remained tied. When they came undone, we learned to tie tighter or added a double-knot. Or think of learning to swim. I can clearly remember learning to swim from one corner of the pool to the other side by floundering to the nearest side and then gradually (and more gracefully) floundering further. The reward came in finding myself swimming.
So, if testing to learn feels new, it’s only new to the classroom experience, and it can be very useful. Here’s how I used it in a Latin class several years ago. Using the Quizzes tool in OAKS, I set up quizzes that students could take as often as they wanted. The only limitations were that they had to complete the quiz in a short amount of time (normally within two minutes) and before a certain time, at which point they got to keep the highest grade they earned, which was automatically updated in the OAKS gradebook. These quizzes started out fairly easy to master, but they quickly became much more difficult. Why? Because I set up question pools that the Quizzes tool pulled from. Here’s an example:
- I set up a folder of fifteen vocab words that I thought would help students read Passage 1 of an ancient Latin test. The quiz tool created a quiz on ten of the fifteen vocab words, pulling randomly from this pool of fifteen and creating a unique quiz each time the student attempts the quiz. Students had up until class time (or an hour before) to earn the highest grade they can. Many reached a 100%.
- For class two, I created another folder of 15-20 vocab words, but this time set up the quiz to pull seven words randomly from the second folder and three words randomly from the first folder.
- By the third class, students were taking quizzes that pulled five words from folder three, three words from folder two, and two words from folder one.
- For the fourth class, . . . you get the picture.
The point was simple: students needed to know the vocabulary to read the passages and compose Latin prose. The quiz tool allowed them to test themselves as often as they needed, but it also forced them to recall vocabulary from previous passages. They tested themselves to increase their vocabulary capacity so that they could read longer passages more quickly. The analytics showed that the class of 14 students tested themselves over eighty times on some weeks.
One last thing to point out. I’m promoting testing to learn to acquire quick access to basic information so that students can more readily understand, apply, analyze, evaluate and create. As with the vocabulary quizzes from the Latin course, this basic information needs be tested periodically, mixed up and interleaved with previously studied basic information. Remembering is the lowest rung on the Revised Bloom’s Taxonomy, but the lowest rung doesn’t imply unimportant. Rather, it means basic. Remembering is a basic skill that allows us to target those coveted higher order thinking skills of analysis and evaluation that will ultimately enable students to create/adapt/build/compile/compose/etc.