Monday, June 7, 2010

12 + 12 = 48 = half-right?

On New York’s state test for fourth-graders, it is. The New York Post writes about scoring guidelines for students, given to them by “an outraged Brooklyn teacher,” that allow partial credit for wrong or no answers. Shocker, right?

Not exactly. Many states have always scored their tests like this. That’s the point of having kids show their work: even if they do the computation incorrectly, they get credit for understanding how to set up the problem. It is worth asking just what level of omission or inaccuracy is deemed acceptable—whether teachers are encouraged to accept even the most fumbling scrawlings—but the general practice of valuing the solution process as much as the final answer has become ingrained into pedagogy. This would be common knowledge if people had a better understanding of what is on standardized tests and how they are graded.

Reducing an emphasis on computation through this sort of scoring is similar to how spelling and sentence structure often don’t count on constructed responses. I sat through an information session years ago where a Maryland official told educators that their students could write bullet lists instead of essays and still get the full score on the written portion of the test, which was, after all, a reading test and not a writing test.


  1. just to be sure i understand, are you taking offense with the awarding of partial credit for math exams?

  2. Is this another of those "d'oh!" articles, like the infamous Richard Cohen column blasting schools because SATs were dropping (because a much larger pool of students were taking SATs)? Maybe they should be giving more tests when newspaper reporters and columnists are hired.

  3. Since the rules have been tightened, my anecdote is old, but it supports your account. In every type of school, regardless of how low the English scores were, Writing tests would have a 95+% pass rate. When monitoring tests back then, I was allowed to read the answers. Almost every one just repeated the topic sentence over and over with stream of consciousness-like variations. Clearly the purpose was to get enough words on paper.

    After testing this year, I praised a student who took the entire time and reread and editted her answers. The other top students argued back forcefully. They had done what the teacher had said was required so they could then lay their heads on the desks and go to sleep. The idea of self-expression or getting ready for college seems to have been completely pushed out and forgotten.

    There is little new under the sun about the politics of testing, even in excellent tests. AP answers are supposed to be frontloaded with information, which was the opposite of what I had been teaching. I didn't have big complaints with the graders' conventions, but one case was interesting. The grader grimmaced with disgust throwing a practice test on the table saying that "He's not near as smart as he thinks he is." The essay was concise and mature, but something about its easy confidence, not showing off his facts but applying evidence to a thesis, made it seem like an elite college essay, not a high school competitive essay. He clearly wasn't conforming to the College Board convention, and it was interesting how their expert grader picked up on it and condemned his playing the college game, not the AP game.

  4. To me, this sounds valid: "...even if they do the computation incorrectly, they get credit for understanding how to set up the problem." That is, unless they chronically do the computation incorrectly.

  5. Anonymous, read again. I am not taking offense; I am explaining it.

  6. linda, anonymous here

    to me it sounds like you are a bit dismissive of the policy, since you liken it to students getting credit for bulleted lists over essays.

    the ny post is really all over this story (a second editorial yesterday) but to me it seems like sensationalism to act like partial credit is so shocking.

    math tests *should* always include partial credit. i've written about this extensively on under the name bronxmathteach. it seems that people have a lack of understanding of what a math teacher might consider worth partial credit.

    in fact, fumbled scrawlings will get no points, but if there is clear mathematical thinking (as there were in the examples cited), then students should be partially rewarded, despite their errors.

    i wish there were more honest pieces out there that consulted math teachers about these things before posting SHOCKING exposes!

  7. That's what I was trying to say, Anonymous. That this is the way it's been done, there's nothing new or shocking about it.

  8. partial credit for equations -- yes but basic math no way...