Tuesday, August 1, 2017

On Track

The reformsters at The74 are providing a platform for the reformsters at Learning Heroes to pitch the same old dreck (I provide the link not because I think you should follow it, but to keep myself honest and allow anyone to check my work who wants to).

Learning Heroes is one more Gates-linked group (via funding and founder Bibb Hubbard, who's a  Gatesian herself). They are particularly devoted to the Cult of the Test, and in fact we did pretty much this exact same song and dance about a year ago.

This time around writer Kate Springer frames the issue as On Track-- 90% of parents say their children are on track in math and reading, but "The real number? Just 1 in 3"

On track? On track??

On track to what? As determined by whom? Based on what evidence?

And really, "track" is problematic as a metaphor, because a track runs from one place to another. The people who assemble these slabs of data-like numbers never look at where the students are coming from, yet that's a piece of information with which parents are intimately familiar. And this "track" data is not a track at all, but a point, a single slice of data from one day of one year. So in a real sense, the conversation between reformy tracking groups and parents is something like this:

Reformsters: Your child is not on track because he's only 5' 4" at age 15 and he is supposed to be 6' when he's 18.

Parents: But he was 4' 10" just last year. He's growing like a weed!

Reformsters: Nonsense. He's stagnant and doomed to failure.

The underlying reformster assumptions here are all false. They assume that there is one single track that all students must travel on, and therefor one track on which their position can be measured. One track for everyone, a track that starts at the same place and ends at the same place. They also assume that they know where that track and they know what the endpoint should be. They talk as if we can know with great certainty that if you are in the Cleveland train station today then you ar right on track to arrive in Boston, even as they assume that they know you should WANT to be in Boston tomorrow. And who gets to do that? Who can best decide when a child is "doing fine" and who has the right to decide what "fine" means? Who are these people so wise that they know the one destination that all students should pursue, and the one track that leads there?

In fact, every single student-- every single live human being-- is on an individual track, a track that starts from where they began their journey, and which ends at the destination of their choosing (including, as well, the many times that they will change their minds about the destination).

Finally, above all the rest of this, this kind of language is an attempt to obscure and give undeserved weight to what we're really talking about-- the student score on a single narrow standardized test.

"Student achievement," "on track," even, as in Springer's piece "how they are doing in school" are all rhetorical smokescreens for "score on a standardized test."

Why can't we talk about what we're really talking about? Because then this conversation would be transparently foolish. "90% of parents thought their children will score above the arbitrary cut score on a single narrow standardized test, but they turned out to be wrong" is not much of a grabber. Because so what? Test scores are a reliable proxy for what, exactly? There's no reason to believe that the answer is anything except "nothing." And if you don't believe me, consider the writing by Jay Greene, an unrelated much more reformy Greene, who has written repeatedly about the disconnect between test scores and life outcomes.

But this misuse of tests as proxy is everywhere in reformdom. Here's DC claiming that their teacher evaluation system weeds out bad apples, when all they've done is prove that if you focus on keeping teachers who are good at test prep, your students will get better test scores.

If reformsters want to talk about test scores, then let's talk about test scores and stop pretending we're talking about bigger, loftier matters of actual substance. But pretending test scores are really indicators of a student's future or measures of student aspirations or the fur depth on an average yeti-- well, that kind of pretending will not help get anyone on track.


  1. Shades of Arne Duncan and his soccer Mom rant. If Kate Springer only knew how little credibility that the Common Core test scores have she would give up her day job. When your foundation is made up of *false assumptions you can only wonder WHEN the collapse will occur.

    *The BS tests are valid
    *The BS tests are reliable
    *The tests are scored accurately
    *The tests are predictive
    *The tests are well written
    *The test are non-political
    *The tests have no hidden agendas
    *The BS tests are instructionally sensitive
    *The tests are taken by highly motivated students

    Kate if you are reading this, we (educators) really don’t give a shit what you think. What goes around comes around.

  2. 1. "Let's get those babies walking!"

    2. Thank you, Peter, for your willingness to do battle on a daily basis. You do make a difference.

  3. A friend with a PhD in psychometetry explained Common Core tests and validity. They are valid insofar as they test what they say they do - Common Core - as opposed to, let's say, horseshoeing.

    But even that's debatable. One source found they strayed into other content 46% of the time. And multiple sources say they inadvertently test computer skills many students don't possess.

    Next, they have no external validity because they don't stick to nationally or internationally recognized formats. For instance, the PISA never phrases a question in a way that is more difficult than the task itself. Yet Common Core tests ask 3rd graders, "Which of the following sentences contains an error in grammar usage?", instead of "Which sentence had a mistake?"

    These tests don't even target grade-level material, but instead whether students fall within grade bands of 3-5, 5-8, and 11. This leaves teachers to scratch their heads as to what the results even mean.

    Finally, even enthusiasts admit most questions are above grade-level. Smarter Balanced ranks most questions at 2-5 years too high. 11th graders are expected to know words like "estuarine", 6th graders "philanthropy", and 3rd graders "infer". That is simply not age appropriate.

    In short, these tests are unfair and useless. They are designed to disrupt the system, not help teaching and learning.