Acquiring knowledge is like building a desk. I teach in a world where, unfortunately, the goal is for my students to build an Ikea desk, instead of a real one, from scratch.
You know the Ikea type: it is sturdy and functional. It looks nice, but its aesthetic is merely the result of composite wood and easy-to-follow directions.
We've all built an Ikea desk (or bedframe or end table). When you finally finish pounding that last nail into your Ikea desk, you feel like a million bucks. You crack a beer and admire your work, like a carpenter. Except for one thing: you are not a carpenter. You are simply very good at following directions.
Meredith Broussard is a data-journalism professor at Temple University and a mother of a first-grader. As part of her job, Broussard hangs out with civic-hackers, the folks who crunch government data in their free time. She tried to game Pennsylvania’s standardized test, in an effort to help her kid with his homework.
Broussard soon found the secret to standardized testing. It wasn’t game theory, but capitalism. To beat standardized tests, the boogeymen of educational reform, you must merely read the textbook.
Well, to be more specific, you must read the textbook published by the same company that writes the standardized test in your state. In some states its Pearson and in others its Houghton Mifflin or McGraw Hill, a different Big Three.
After starting down the rabbit hole of education policy, especially the decisions made by low-income districts and schools, Broussard quickly found a correlation between the material tested at the end of year and the material covered in certain textbooks. “Standardized tests are not based on general knowledge,” observes Broussard, “they are based on specific knowledge contained in specific sets of books: the textbooks created by the test makers.
There are the horror stories: Houghton Mifflin has a 38 percent market share, according to its press materials. In 2013, the company brought in $1.38 billion in revenue. Its customers are schools and its target audience is children.
Pearson used a reading passage in a textbook. They used it again on a standardized test. Talk about double dipping.
Now, Broussard misses a few small points and mistates the emphasis on a few others, but that doesn’t matter much in the end. She gets it. Hopefully she brings others with her. It is refreshing to see an outsider discover for herself how a few decision makers have rigged the system against children and against teachers.
Now allow me to attempt to add to the discussion that Broussard has so eloquently started: she makes the point that this is a problem primarily for low-income students and the schools that serve them. Since textbook companies write standardized tests, students and teachers must have the correct brand of textbooks in order for scores to rise and students to succeed. And, it is exponentially more difficult for low-income schools to obtain textbooks, let alone the “right” ones.
Without the textbooks, asks Broussard, how can students who are already behind hope to catch up? Without the directions, how are students supposed to assemble the desk? How are teachers supposed to help?
Again, for Broussard, this is a problem for low-income school districts, for "poor schools." Superintendents in rich areas understand what’s happening, so they buy the books. Decision makers in similar positions for low-income districts are stuck participating in “triage” after years of budget cuts and other crises created by the poverty in which their students live. These decision makers are either too under-funded or too over-worked to prioritize books purchases.
Broussard refers to a “textbook problem” of selecting the right books and the “data problem” of keeping track of textbooks and curriculums across massive urban school districts, like Philadelphia. She refers to the budget problems - - last year Philadelphia schools were allotted $0 per student for textbook. This year’s budget allocates the same amount for books.
These are real problems that face schools and principals and teachers and students.
But, the (singular) problem facing low-income schools is broader. The problem is the current state of standardized testing, and the current companies who write them.
One solution to the problem is buying the right textbooks. Another solution is eliminating the tests altogether, both of which are suggested by Broussard.
But, you see, the problem is not simply a dollars and cents one. It is not simply that low-income school districts cannot buy answers (textbooks). The problem, also, cannot simply be eliminated, like a redundant variable in an equation. Standardized tests have been codified in law, both state and federal, and tethered to funding and salaries.
The real problem is that the standardized tests are not good tests. And, by valuing (and over-valuing) bad standardized tests, districts place unrealistic pressure on principals, who place unrealistic pressure on teachers, who place unrealistic pressure on the little people sitting in classrooms. The results are not usually good.
Today, teachers do not stand and deliver, they stand and produce or they stand and fail, alongside their students.
You see, I do not have a problem with standardized testing, in the abstract "let's sit in my professor's office or the local coffee shop and tease out the complexities of this week's topic of choice" sense.
Having an end-of-year test is a good idea. And, when teachers teach to that test it is also fine, as long as two things hold true:
1. The test must be a good one and
2. Teachers must write, or at least understand the contents of the test.
My students, however, take a test that is not a good test. It is a bad test. And, my students are usually taught by teachers, great teachers, who are giving and kind and reflective, but usually do not understand the test as well as they would like (in part because previous years' tests are not released, at least in Georgia). I can answer every Jeopardy question asked by Alex Trebek and every LSAT question the Law School Admissions Council going back to the 1990s. But the Georgia DOE and McGraw-Hill won't let me see the questions on the Sixth Grade Science CRCT from last year or any other year?
The Georgia CRCT that is administered to my sixth graders is not a good test. And, the Georgia MAP test that will replace the CRCT is unlikely to be any better, as it is written by the same people and they will be paid the same amount of money for their services.
The Georgia DOE produces a very basic CRCT study guide. In the Earth Science section, it includes a few sample questions which show the deficiencies in the test:
Which of these features is MOST LIKELY to affect a planet’s ability to support life?
A. the presence of water on the planet
B. the distance of the planet from the sun
C. the number of moons that orbit the planet
D. the gases that make up the planet’s atmosphere
Now, I went back and checked my Astronomy PowerPoint that explained why there is life on Earth, but not on any other planet. I taught my students that three things were needed for life: water, oxygen, and a moderate climate. I taught my students that this was a checklist. The absence of any of the three would deny the possibility for life. Specifically, I emphasized how Earth was a “goldilocks” planet - - neither too hot nor too cold. I asked my students a question about why two of the conditions (water and climate) were related. They understood: if the climate is too hot (because the planet was too close to a sun), the water would evaporate into space. We read a current event article about the discovery of goldilocks planet, Kepler 186-f, in a distant galaxy. Yet, this test question asks my students to decide which of the three conditions is MOST important. Scientists are allowed to debate the relative importance of water, oxygen, and climate. My students aren’t. At least not according to the test.
If half of the problem is the test itself (not only the textbooks aligned to the test), the other half is the inadequate training teachers receive in preparation for the test.
I have not received any training on what topics will be covered on the annual CRCT. I now know what will be on the test, but that is a result of close-reading densely worded standards and treating professional development sessions like flea markets - - hawking benchmarks from neighboring counties and charter schools. Neither of these activities is in my job description, or even encouraged by my superiors.
There are many Earth Science topics, but only some have a chance to appear on the test. For example, consider clouds. Most earth science teachers in Atlanta teach clouds (they are in the textbook after-all). But they are absent from the Georgia Performance Standards, which are written to guide my instruction. S6E3.b. tells me that my students should be able to "relate the various atmospheric conditions to stages of the water cycle." S6E4c tells me that my students should be able to "relate how moisture evaporating from the oceans affects the weather patters and weather events such as hurricanes." Can you tell how difficult these standards are to read, and how difficult it can be to determine what topics fall under their umbrellas?
These standards are more William Faulkner than Malcolm Gladwell. But assuming that I get through the syntax, I will learn that my students must know something about clouds, but not a lot. Certainly my students need to know that clouds exist, and that they are formed by the evaporation of liquid water. They need to know that heat from the sun causes water to evaporate, but too much heat will cause the clouds to dissipate. They will need to know a lot about hurricanes - - which get a personal shout-out from the Georgia DOE. But will stratus and cumulus clouds be on the test? Nope. So, why teach it?
Now, the right textbooks might help alleviate these problems, but they will not solve them.
Broussard wishes that education budgets would match curriculum requirements. If that proves impossible, she suggests eliminating standardized tests or at least stop aligning them to specific sets of books. Hopefully these things happen.
But in the meantime, forget the textbooks. We either don’t have them or our students don’t read them, due to poor reading skills.
Instead, write better tests. And, ask teachers to write the tests, not companies. We are cheaper, and more qualified. What a deal.
And, in the meantime, give teachers some training about what will be on these tests. We are teaching to the tests anyway, we might as well have a better idea of what will be on them. Do this and see what happens: teachers will teach more efficiently and purposefully, freeing up extra days for lessons and extra weeks for activities that matter to children and the communities in which they live. Give teachers a clearer picture of what is on the test, and they will give you a curriculum that raises scores, and one that even stands a chance of closing the achievement gap.
Give teachers a better test and a better idea what will be on it, we may just teach our kids to be critical carpenters, capable of building a mahogany desk of an idea from scratch and polishing it to perfection. Such critical carpenters will be able to do much more than assemble an Ikea desk