You may have heard this one-- just teach your students really well and they will do well on the test, just like a person who can drive really well can do well on the driver's test.
Well, no. This is not a great analogy.
Let's say instead that it would be like saying that since you are a great driver of your family automobile, you should be able to pass the driving test on a motorcycle or an eighteen-wheeler.
For our purposes today, I'm just going to talk about the reading test. There are several reasons that just Being a Good Reader does not mean that things will automatically be hunky dory on the Big Standardized Test.
First-- the reading task is completely inauthentic.
In plain speech, the reading assigned on a BS Test is, by design, unlike the reading that good readers do.
The selection is short and disjointed, and would qualify as authentic if real readers routinely grabbed a book from the shelf and said, "Well, this looks interesting. I'll just read one page of it." And the disjointed part is on purpose. Test designers try to make selections that take prior knowledge off the table-- but that is also an inauthentic task. All reading that we do, we do because we are connecting something to prior knowledge. We pick a book because it's about something we're already interested in. We read informational texts in order to extend the knowledge we already have.
We do not, in the real world, say, "Boy, I'd like to read about something I know nothing about-- but I'd only like to read just a little bit, and with no context, either, please."
Second-- the reading task is not the only reading task.
We don't talk about this enough-- the students don't just have to read and understand the reading selection-in-a-bubble. They also have to read an interpret the questions.
In our driving test analogy, this is like saying that since you are a good driver, you should have no trouble taking a driving test while balancing a stack of plates on your head. We have added a complete extra task and tied to slough it off as something that , of course, everyone can do.
Test manufacturers and promoters try to hide this by talking about test questions as if they descended from God on a cloud powered by a burning bush. The implication is that these questions are completely objective. But they can't be. Yes, "what color is the ball in the picture" is pretty cut and dried, but the higher order we get to in thinking skills, the more subjective the questions and answers must be, until we arrive at questions along the lines of "which picture of the ball is best?"
Test prep for SAT and ACT and other BS Tests is largely centered on How To Read The Questions, or How To Figure Out What The Guy Who Wrote the Question Wants You To Say.
This is where the much-noted bias creeps in. If it's easy for you to get in the head of the test writers, to paraphrase John Oliver, congratulations on your rich, white penis.
More to my point-- this pick-out-a-correct-answer-from-the-choices-we-gave-you is an inauthentic activity that does not resemble anything we ever do anywhere ever in the real world. It is actually worse than the inauthentic reading around which the test is centered, because that at least bears a passing resemblance to real reading. Driving the eighteen-wheeler has some passing resemblance to driving my car, but balancing the plates on my head has nothing to do with anything!
To suggest that just making students into good readers will automatically insure good test results is baloney, and it's not like the relationship between driving well and passing the driver's test at all. The BS Test is a hugely inauthentic task, and as such has its own set of skills and behaviors that it rewards. And I have no desire to practice balancing plates on my head just so I can get a drivers license.