Standardized testing is truly absurd.
It’s both a prediction and an assessment.
You take a test to determine what you’ve learned and that will in turn predict what you will be able to learn in the future.
We hardly ever do this anywhere else in life.
We don’t measure babies’ leg muscles to predict whether they’re ready to walk. We let them do what they do, possibly with some encouragement and positive models of locomotion, and they do it.
There are cognitive and developmental benchmarks we look for, and if children don’t hit them, we provide help. But no further prediction is necessary – certainly not based on artificial markers put together by corporate interests.
In most situations, predictions are superfluous. We just assume that everyone can learn if they so desire – unless something happens to make us think otherwise. And whether someone actually learns something is demonstrated by doing the thing, itself.
The only time we link prediction and assessment so closely is when the consequences of failure are irreversible – like when you’re going solo skydiving for the first time. If you jump out of an airplane and don’t know how to pull the ripcord to get your parachute to work, you probably won’t get a second chance to try again.
But most things in life aren’t so dire.
The world of standardized testing is very different. The high stakes nature of the assessments are what ramp out the consequences and thus the severity.
Testing looks at learning like two points on a map and sets up a gate between points A and B.
In order to cross, you have to determine if you’ve passed through the previous gate. And only then can you be allowed to progress on to point C.
But this is wrong on so many levels.
First, you don’t need a test to determine which point you’re at. If Point B is the ability to add, you can simply add. If it’s the ability to write a complete sentence, you can simply write a sentence.
There is no need to fill out a formal multiple-choice assessment that – depending on the complexity of the task being considered – is completely inadequate to capture the subtleties involved. The task, itself, is enough.
Imagine if you were testing whether someone had learned how to drink a glass of water. You could just give them a cup filled with H2O and see if they can gulp it down. Or you could have them sharpen their number 2 pencil and answer questions about how their throat works, their digestive and excretory systems and the chemical composition of agua – all answers predetermined to A, B, C or D.
Observation of a skill, we are told, is not enough to determine success because it relies on the judgment of an observer. A standardized test replaces the observer with an impersonal, distant testing corporation which then assesses only predetermined markers and makes decisions devoid of any situational context.
This is done to remove observational bias but it doesn’t avoid bias altogether. In setting up the markers and deciding which elements of the task are to be assessed (or in fact can be assessed in such a distant manner), the testing corporation is inserting its own biases into the process. In fact, in any assessment conducted by human beings, this would be inevitable. So going through this maze of perceived objectivity is really just a matter of subterfuge meant to disguise the biases of the corporation.
Second, assessing people in this way is extremely unnatural because very few fields of knowledge can be divided and subdivided into two or more discrete points.
When writing a complex sentence, for example, you need to know not just spelling and grammar but logic, handwriting, subject matter, colloquialisms, literary devices, and a plethora of other cultural and linguistic artifacts.
Moreover, there is not always a natural progression from Point A to B to C. Sometimes A jumps directly to C. Sometimes B leads directly to A. Sometimes A leads to Z.
Knowledge, skills and human cognition are far too complex a web to ever hope to be captured by such a reductive enterprise. But by insisting that we make this complexity fit into such a small box, we end up depriving people of the right to move on. We say predictive models show they aren’t ready to move forward and so we bury them in remediation. Or we deny them access to important opportunities like advanced classes, electives, field trips, extracurricular clubs or even post-secondary education.
Third, this emphasis on knowledge as discrete bits of information or skills (often called standards) leads to bad teaching.
Assessment expert W. James Popham provides a helpful distinction: “curriculum teaching” vs. “item teaching.” Curriculum teachers focus on the full body of knowledge and skills represented by test questions. For instance, if the test is expected to include questions about decimals, the teacher will cover the full range of knowledge and skills related to decimals so students understand what they are, know how to manipulate them, understand how to use them to solve more complex problems, and are able to communicate about them.
By contrast, item teaching involves narrowing instruction, organizing lessons around look-a-like questions that are taken directly from the test or represent the kinds of questions most likely to be found on the test. In this way, the teacher only provides the chunks of knowledge students are most likely to encounter on exams. For instance, item teachers might drill students on a certain set of vocabulary words that are expected to be assessed rather than employing instructional strategies that help students build a rich vocabulary that best contributes to strong reading comprehension.
A focus on standardized testing or even trying to educate in a system where these tests are attached to high stakes, results in an increase in item teaching. We often call it teaching to the test.
I’m not saying that item teaching is always bad. But curriculum teaching is to be much preferred. It is a best practice. The problem is when we resort to endless drills and give students innumerable questions of the exact type we expect to be on the test.
So when we find students who have made dramatic improvements on standardized tests, we often don’t find equal improvements in their over all knowledge or ability.
Test scores are often a false positive. They show students have mastered the art of taking the test but not necessarily the knowledge or skills it was meant to assess.
They are more like trained circus animals who can jump through flaming hoops but would be lost in the wild.
That’s why certain computer modeled artificial intelligences are able to pass standardized tests but would fail preschool.
These reflections have troubling implications for our system of standardized testing.
The false curtain of objectivity we’ve set up in our assessments may also be hiding from us what authentic learning is taking place and it may even hinder such learning from taking place at all.
Any sane society would halt such a system with these drawbacks. It would stop, regroup and devise a better alternative.
To continue with such a pedagogical framework truly would be the most absurd thing of all!
Like this post? I’ve written a book, “Gadfly on the Wall: A Public School Teacher Speaks Out on Racism and Reform,” now available from Garn Press. Ten percent of the proceeds go to the Badass Teachers Association. Check it out!