Friday, April 15, 2011

Standardization & Test-Based Accountability Makes Kids Hate School

Dana Goldstein recently published a piece of long-form journalism in the The American Prospect called "The Test Generation." It's about education reforms that have taken hold in Colorado Springs, Colorado, and it has elicited praise from most with some criticisms from others.

I know that Matt Yglesias is smart and clever and very influential, but every time I read something written by him on education, I want to break something. (I know, I know, everyone loves the guy. I'm like Elaine in this episode of Seinfeld about "The English Patient.")

In response to Goldstein's piece he says:
"The exposition is very clear and informative, and though it’s obvious Goldstein is skeptical (for basically Campbell’s Law reasons) that this is going to work out, those of us who are less worried about the Campbell’s Law phenomenon see here a very strong portrait of a school district going all-in on measuring and rewarding quality."

There's a huge assumption being made here: That standardized tests measure quality of learning and teaching. They mostly don't do that. Yet he regularly makes this assumption in his columns.

Standardized tests have important uses such as for diagnostic purposes (for example, for gauging reading proficiency levels so teachers know which books students can handle or for diagnosing disabilities) and for estimating at the school and district level where students are, but I reject the notion, especially if you look at the low quality of many of the tests themselves, that they measure quality of teaching or learning.

As a teacher and now as a parent I don't want just a measurement; I want an evaluation. I want teachers and tools that evaluate my children's achievements, what they've learned, what they know and what they can do. I want to read their writing, to listen to a podcast they've made about the life of a famous American, to view their art work, to see science experiments they've done. And I want to evaluate my children's teachers based on the climate in their classrooms, on the relationships they have with my children, and on the quality of knowledge gained and work done under their guidance. Standardized tests tell me almost nothing about this.

In the same post Yglesias says:
"The one real bone I’d pick with her characterization of what’s happening in Colorado is that I’m not sure how much sense it makes to complain that 'high-stakes testing' is being administered too often. If you have someone sit for a test 25 separate times per year then the stakes for any one test can’t be all that high. That would seem to me to be part of the point of testing frequently."
Okay. Yes, the stakes will be diluted by giving twenty-five tests. But that's twenty-five freaking tests! Students will be bored to frustration and mostly not engaging in meaningful learning during those twenty-five days. I'm okay with some testing, especially if it's teacher-generated and meant to help students remember and practice what they've learned. But I am not okay with twenty-five days of standardized testing and if that comes to my district, I will complain vigorously about it whenever I have the chance.

Alexander Russo made a similar criticism in this post about Goldstein's article, that he wasn't sure how "stressful" mid-year assessments were. Okay, granted--some kids get stressed about the tests, some don't. My sons don't, but they get really, really tired of preparing for them, of engaging in learning through the lens of a multiple choice test. They're only in second grade, mind you. In Virginia, the high-stakes standardized tests don't start until third grade, at which point I've heard that the testing mania gets much, much worse.

For now, they are constantly taking tests: multiple choice tests, benchmark assessments, passage-based reading tests. They are bored by taking so many tests. They tell me this all of the time. (For a truly disturbing tale from a teacher at a school with an abhorrent culture of standardization and testing, read this moving account from Mary Ann Reilly's blog.) Now, some things in school, and life, are boring, and there are good reasons behind some testing, and I tell my kids this, but do we have to bore them and waste their time so often and so insidiously, just for the sake of a measurement that a bunch of think tankers, oligarchs, politicians, legislators, and education reform industrialists want? Do we have to ruin the learning experiences of our children to get these people their probably-not-even-valid measurement?

When I was a teacher, I didn't get stressed about the tests themselves. As Teacher Sabrina describes here, the testing days were easy. There was nothing to plan, the kids behaved, every one knew what to do. The part that's stressful is having to constantly prepare for the tests, to give practice tests, knowing in some cases that the students must pass them to graduate. It was stressful knowing that the quality of my teaching was going to be evaluated by an assessment I had not made or seen.

What was truly stressful, though, as journalist Linda Pearlstein describes here and as teacher Amanda Sheaffer describes so poignantly in this letter to her third grade students, was knowing that I was cheating my students out of a rich and meaningful education. Now that my own children are in school, I have careful conversations with their teachers about this and they agree: the standardization, the pacing, the constant assessments, the high-stakes testing all gets in the way of what most teachers are really yearning to teach and what students, such as my children, are truly yearning to learn.

Michael Miles, the Harrison District superintendent Goldstein profiles, is wrong when he insultingly says the options for an art teacher are between narrowing standardization and "doing coloring." What an incredibly impoverished view of art education, and of the teaching and learning processes in general. How utterly depressing that these are the ideas of the people "reforming" our public schools.

Goldstein didn't interview a Colorado teacher "who didn't mind the extra testing or even thought it helped her do her job better" as Russo implored her to do because, as John Thompson comments here, she was probably hard pressed to find one. I was recently critical of journalists for "balancing" their stories for the sake of balance at the expense of truth and accuracy. However, I praise Goldstein for resisting that habit in this particular article, for relaying truth via narrative (as, to his credit, Yglesias remarks) about what the recent education reform policies enacted in places such as Colorado are doing to teaching and learning, to teachers and students, to tell the story that those who teach and those who are parents see unfold almost every school day.

To conclude, this piece in The Economist provides a brilliant historical analogy for why the standardization approach to education that Goldstein chronicles, although well-intentioned, is the wrong one.

At the dinner table a few months ago my heart sank when my son Caleb, a voracious reader and creative soul, announced, "Mommy, I love reading, but I hate reading tests." As testing takes over, I'm afraid this will simply turn into: "I hate school."

No reform, no measure of "quality" is worth that.


  1. Thank you for saying what needs to be said.

  2. Teachers appear to be just as good or better as assessing where kids are at (assuming their year isn't crammed with 25 tests), for example, the grades teachers give better predict college success than college entrance exams do. The tests have made themselves seem more useful than they are by displacing from the day all the other means through which teachers can assess kids--and we're already paying the teachers to be there.
    David Berliner also reports that he has data that teachers are just as good at identifying reading difficulties. Standardized tests are the potato chip that makes us but the drink (test-prep curriculum). There's little evidence we need them at all, except for diagnosing disabilities and some broad matrix sampling so we know broad trends.

    -Karl Wheatley

  3. Thanks, Anon for reading & commenting!

    Hey there, Karl. So glad you made it here. I generally agree with you about standardized tests.

    I refer to reading tests as an example (I don't know about math because I don't have any kind of a math background, but as an ESOL teacher, I would also use English language proficiency assessments as an example) because I do think sometimes it can be useful to have normed assessments that all students in the same class, school, or district take so that their educators and parents have a general idea of where their students are compared to their peers. It's only one small piece of the puzzle, but it can sometimes be an informative one. Such information has helped me and I've seen it used appropriately and well by my children's teachers. That being said, districts need to make sure the tests are of high quality, that teachers find them useful and worthwhile, and that they're being used for their intended purpose--as tools to inform instruction. In general, to figure out what students have learned, I agree that teacher-generated assessments are far superior.

    In fact, equipping kids with strategies to assess themselves, for example to tell if a book is too easy or hard, is even better. My sons were taught in kindergarten and 1st grade strategies such as the 5-finger rule and taking a "picture walk" prior to reading a book. It gave them the confidence to try books, but the agency to say "this book is too hard for me. I'm going to put it down and find something else to read." It saved them a lot of frustration and helped to build their confidence.

  4. One of the problems with testing and this test culture is that the idea of intrinsic value is lost. As teachers, I really feel that our job is to instill a love of learning, to foster a child's curiosity, to make them WANT to learn. As you said, Rachel, tests are excellent diagnostic tools. So are thermometers but we want our doctors to do more than just take our temperature. I see the weariness in my students after 4 BAS tests and a DC CAS and I know I have my work cut out for me in keeping their interest for the rest of the year.

  5. The idea that increasing the number of standardized tests makes it OK because how can 25 (or 52 in a district in NC) tests be considered high stakes since instead of one there are many more defies logic. Testing that counts beyond use in the classroom is high stakes, period. Keep writing. We need to say this loud and clear, over and over again.

  6. Just an aside. Not everyone "loves" Matt Y. He was a staunch supporter of the runup to the Iraq invasion. He is very slippery about being pinned down on opinions. And his writing is often pretty awful. Like many of his young counterparts on the right, he seems to know how to put words together that please his elders, and I envision him eagerly awaiting the pat on the head as reward.

    That said, he is great on urban/land use/transportation policy. Nothing is cut and dried, eh?

    Just found your blog via ED and it looks like a place I will enjoy. Keep up the hard work.

  7. I have actually thought the same thing about MY's writing, but then again good writing is often in the eye of the reader. I've heard the same thing about him on urban planning.

    I suppose part of the problem is that we're in the era of the all-purpose pundit columnist/ blogger, where there's pressure to produce lots of opinions on a wide variety of topics, but very little in the way of expertise or research or even humility.

    In any case, thanks so much, robo, for the comment & encouragement.