It’s time we hold tests and grades accountable

This is a classic cookie-cutter response that I regularly receive when discussing the abolishment of grading and testing, and I would like to respond with a few points.

Testing and grading have not existed forever; they are modern day technologies that educators and education systems adopted relatively recently. With every technology there are sources of error. Remember that in order to reduce something as messy as learning to a number requires some underlying algorithms which are fallible. Your doctor will tell you that your cholesterol test has 20% error associated with it. Political polls like to believe they are gaining valuable insight into who will win the next election, but they advertise their error rate. In other words, many professionals attempt to make inferences based on samples from an entire domain. But do teachers or the education system acknowledge the error associated with grading and testing? Every time teachers, administrators and policy makers sell these grades as an act of precision, they are engaging in assessment malpractice.

There are only two ways you can test people. You can ask them to supply a response/performance or you ask them to pick an answer. The latter includes things like multiple choice exams which only came about around 1910. The former has been around forever. To claim the latter as a requirement for teaching in public education seems ignorant.

Any accountability system that shrugs at the harms done to student learning in the name of reporting must be held to account. Anyone who believes our mania for reducing everything to numbers is reporting real learning accurately has lost the plot. Grading and testing as a means of holding schools accountable is not like the weather – it’s not just this thing we have to get used to; rather it is a morally objectionable and intellectually indefensible political movement that must be opposed.

“Worse than giving a grade is giving one that isn’t based on fact” is a very powerful comment, but if Linda McNeil from Rice University is correct in saying that measurable outcomes may be the least significant results of learning than the traditional methods for counting grades may be based on “facts” or criteria that simply don’t matter all that much. If you talk to parents and ask them what they want for their children from a good education, they don’t say “well, I want them to understand standard deviation and coordinating conjunctions, and they better help their school score well on their accountability pillars.” I’m not sure how we can properly grade and test creativity, perseverance, initiative, intrinsic motivation, democratic citizenry, social justice, patience, thoughtfulness and diligence, but I do know our curriculums, programs of studies and state standards tend to ignore all these “cool” things. Some might speculate that giving a grade not based on ‘fact’ is worse than giving any grade, while others will say that there is a big difference between valuing what you can measure and measuring what you value.

The perceived need for grades is an argument built on the need for ranking and sorting children which have nothing to do with learning, and it’s time to hold such an argument accountable.