Are tests and rubrics the enemy?
One of the challenges we face when trying to improve education is that opinions often greatly diverge as to the best course of action. This disagreement is evident in both informal discussions among colleagues as well as conflicting scientific studies on the topic. Alfie Kohn decries the culture of testing in schools in “The Case Against Grades.” According to Kohn, “frequent temperature-taking” in the form of tests is unnecessary and, furthermore, inadequate to evaluate student learning and progress. Kohn goes on to argue that grades produce anxiety among students that detract from learning and decrease creativity. I can identify with the feeling that tests sometimes do a poor job of asking students to show what they know. I have led a few lectures for my advisor in his undergraduate hydrology class, and he asked me afterwards to write a few exam questions on the material I covered. His tests are a combination of multiple choice, short answer, discussion, and calculation problems. I always found the short answer, discussion, and calculation problems fairly easy to write, and I think they can be crafted in a way that tests the knowledge of the student pretty well. However, I had a lot more trouble with the multiple choice questions. Maybe creating multiple choice problems gets easier with practice, or it might be somewhat of an art, but I remember thinking that no matter how I phrase the question or what answer options I provide, the questions just seem inadequate and either really easy or sneakily obscure. Kohn insists that tests should be a rarity, and Marilyn Lombardi talks about other options for demonstrating learning, such as portfolios.
To complicate matters, other pedagogical studies talk about how tests are one of the most effective learning tools and that we should test more, not less, often. Preposterous, you say? Perhaps. What I am referring to is called “the testing effect” and is discussed in Make it Stick: The Science of Successful Learning. Apparently copious research shows that, if you want your students to remember something, you should test them on it. A test does not necessarily have to take the form of a high-stakes, anxiety producing, multiple choice final exam. The authors include any form of information recall that students do without looking at their notes, such as using flash cards or quizzing each other. Any time that you have to work to remember something, your brain makes a stronger connection to find that information, so it is easier to do so the next time around. The authors also warn readers up front, “your students won’t like this.” However, they also give advice on how to incorporate the testing effect without terrorizing your students. Namely, giving frequent, low-stakes quizzes that do not really impact the grade that much, which also helps to decrease the negative connotation of tests. I was a big convert to the testing effect after reading this book, but I do have reservations about the frequent quizzing, which would become a form of taking attendance. I think Kohn is pretty extreme in his arguments, but I do not think that traditional tests are the best method of student evaluation in many circumstances. Portfolios, papers, and projects are often far superior options, but I think that tests do also have their place. For example, I tagged along during a dendrology field lab last week to observe the professor, and dendrology is definitely a class that requires substantial memorization. The professor did a great job of interweaving stories and context to the different trees and also gave students tips about how to organize their tree descriptions to see connections among species. He also quizzed the students four or five times during the class on trees they learned the previous weeks. I think this sort of class (anatomy would be another one) is a good candidate for frequent testing, which the dendrology professor is already doing. I guess I would caution that tests do serve a purpose in some cases, so do not completely overlook their potential
Similarly to tests, scholars disagree on the value of rubrics. Kohn thinks that rubrics discourage creativity by telling students what to expect and delimiting boundaries on the project. On the other hand, Lombardi promotes rubrics. The rubrics I have seen as a student are usually pretty general and do not seem to greatly constrain the project, especially if the professor includes something along the lines of “other project formats are acceptable but must be cleared by the professor to make sure it is appropriate.” I honestly think rubrics are kind of annoying, but I also believe they can be good to guide the assignment with a general set of expectations. In another book I read, How Learning Works: Seven Research-based Principles for Smart Teaching, the authors describe and then problem-shoot a common complaint of professors that students come into a class unable to carry over previous knowledge from former classes. The authors attribute this inability to a lack of “deep learning,” which may be the issue more often than not, but I also feel that sometimes students simply suffer from tunnel vision and do not think to apply knowledge they already possess in a new environment. Small prompts on assignment instructions or rubrics might go a long way in helping students tap into these other resources they possess. Thus, though counterintuitive, maybe such guidance can actually increase creativity? Rubrics are also good for transparency in grading to decrease resentment among students and help them to understand what they did and did not do well. I had a T.A. last semester who deducted points for nit picky and really just random and unfair reasons that made no sense or were flat-out wrong: we could do no right on our assignments, according to him. In the words of my friend in the class with me, “I have never felt personally attacked by a graded assignment in my entire life until now.” We never debated the grades with him to avoid being “those people” that quibble over points, but he would have avoided considerable resentment if there was a rubric at least suggesting some of the logic behind the strange deductions. It’s like, “if you wanted it that way, why didn’t you just say so?”.