Friday, October 16, 2015

PARCC Expectations

As states continue to brace themselves for the release of crappy PARCC scores, now is a good time to look, again, at the PARCC Levels of Student Awesomeness:

Level 1: Student did not meet expectations.
Level 2: Student partially met expectations.
Level 3: Student approached expectations.
Level 4: Student met expectations.
Level 5: Student exceeded expectations

All levels share a critical term. Expectations.

It's a well chosen word from a PR perspective. Well-chosen, but not correct. Even, kind of, a lie.

After all-- what are expectations? They are an idea you set about before the fact. I have expectations about how my food will taste, and then I taste it. I have expectations about how good a movie will be, and then I watch it.

I don't listen to a new music release and then, after I've heard it, develop some expectations about whether it will be any good.

And in Teacher 101, we all learn that our expectations of our students will shape their performance-- what we expect them to accomplish will affect what they actually accomplish. Expectations are the horse, and performance is the cart.

So if we talk about expectations on a test, that means that before students take the test, we say, "I expect that students who really know this stuff will get at least nine out of ten items correct." In fact, if we're good teachers, we share the expectations with the students so that they know where the bar is set. That way they can also set some expectations.

By talking about "expectations," test manufacturers give the impression that their tests follow a similar chronological procession. They design the test. The set expectations of the "top students will get nine out of ten correct" sort. The students take the test. We score them and see how well they met the expectations.

That, of course, is not how it works at all. The test is designed. Students take the test. We score the test. And then, we set "expectations."

And that can only possibly be true if PARCC headquarters house a time machine.

You cannot set expectations after an event has already occurred.

We need a new word, a different word, for what test manufacturers and bureaucrats do when they set cut scores and decide who does well and who does not, because they are not setting expectations. Words have meaning, and that is not what "expectations" means. They might just as easily says "Student exceeds badgers" or "Student is taller than blue." But to say "student exceeded expectations" when you had no idea what the expectations were before you handed out the test-- that's simply a lie. The use of "expectations" is a way to hide the truth of the process from parents, teachers, students and politicians.


  1. Perceptive and insightful, as always.

  2. The PARCC/SBAC setting of cut scores after the fact PROVES that they really have NO IDEA what it means to be "college or career ready".

  3. The more fundamental problem with the word "expectations" is, whose? It's one thing to talk about a teacher's expectations for his or her own kids or parents expectations for what their kid is going to learn, or even (better yet) a kid's expectations for what s/he wants to get out of the experience. But a standardized test by definition is a standard expectation for all students who may or, more likely, may not have learned the same things.

  4. BTW, aren't levels two and three mixed up? If you've "partially met" expectations, doesn't that mean you've met some but not others? And if you're "approaching" expectations, doesn't that mean you haven't met any?

  5. Postspectations. "Having massaged the grading to our satisfaction, then having clustered the scores into politically expedient ranks, we can now tell you how students met our postspectations last year and what those postspectations would have been."

  6. Another angle to pursue here are the different levels. What, exactly, is the distinction supposed to be between level 2 and level 3? Are there different expectations such that a student could meet some but not others? Is that partially meeting expectations? As for "approaching expectations" does that mean coming close to some of the expectations? How many? How many expectations are there in the first place? How are they measured? If measured separately, why not report them separately?

    The fact that these questions don't seem to have answers indicates that this is all pseudo-science. These tests are not objective. They don't really measure anything definitive.

  7. This comment has been removed by the author.

  8. Postgnosticate. Postficiencies. The teaching postfession. Get with the postgram!