Tuesday, February 24, 2015

Adaptive Students and Adaptive Tests

One the great unexamined assumptions of the test-driven accountability fans is that students will actually give a rat's rear about the tests.

I'm not sure why the test fans make this assumption. Maybe they were the kinds of students who took every single test with the utmost seriousness, whether it mattered or not. Maybe they have convinced themselves that the tests are super-duper important and they can't imagine how anyone would think otherwise. Maybe it's been a really long time since they met someone who was fifteen years old.

If you like your "news" in anecdote form, there's this piece that's been bouncing around the internet that gathers a few choice tweets about the PARCC. Not exactly conclusive, but I'm betting few high school teachers read it and shake their heads, thinking, "My land, but I never heard of students saying such a thing." No, in high school land, we already know that one of the challenges of the Big Standardized Test is convincing students it isn't a complete waste of their time.

But this unexamined assumption really hits the fan when we get to adaptive testing.

The idea of an adaptive test is, of course, that it adjusts to the student level-- the smarter the student appears to be, the more the test ramps up the question.

But that only works if the student is motivated to do his best no matter what, if his reward is knowing he's done the best he could possibly do. However, if a student thinks the reward associated with the test is to be done with the test, adaptive testing looks completely different.

If my reward is to be done with a minimum of fuss, then I can adapt to my adaptive test easily, because here's the deal-- the more questions I answer incorrectly, the easier the questions get, and the quicker I can finish. And as I was writing this this morning, stories are starting to wander into facebook of exactly this happening-- students who have been burned out on endless pointless testing are starting to figure out how to game the new super-duper tests.

My reward for answering questions well is that the test gets longer and harder.

Adaptive tests can only work for a body of students who are driven to do their very best and show that impersonal, inhuman, pointless, no-stakes, computerized test just what they're made of. It's for students who are incapable of analyzing and adapting to their testing environment, but who are simply stuck in a default full-on setting. For everyone else, this is like some bizarro video game where the more fights you lose, the sooner you get to the final boss and the easier he is to beat. Tell me my students don't have the critical thinking skills to figure that one out, or the adaptive response skills to adjust to it.


  1. OMG I am a second grade teacher where we give the Map test 3 times a year. I have a student who I feel is truly exceptional he pretty much gets every thing put in front of him right even on pretests before he is taught how to do what is on the test. When he took the map and adaptive test in the fall in reading his score was as low as my students who are on IEP's those who are reading at a kindergarten level. However when he is given an ORR to see where his reading level truly is he is read and a Fountis and Pinnell level of Q and can answer almost any kind of question you throw at him. Same with the math scores...so we thought oh he did not spend enough time on the test and just flew through them...sooo....with much deliberation the district said retest him and talk about the importance of doing your best. (One thing I have learned is that he does not like to get the wrong answer at all....kids like him usually feel that means he is a failure for many reasons) we he retook the test and......the exact same scores....I know of no child who could do something like this. The questions are supposed to be roaming questions so no one has the same question as the child next to him so how did he do this? I asked him if they were the same questions and he said pretty much. Before winter break he took the tests again. This time reading was a better score putting him in enrichment..The highest level. However he did the same with math. I know he is better than his score. When we went over his scores and how he did not make growth ( forgive me but this is mandatory in our school) he said I did not do the hard ones I did the easy ones. This is what he did...he actually figured out what the test was doing and did the easy questions thus not an accurate score.....SMH

  2. I have gifted children and that's exactly what a gifted child will do - they figure out the system, think outside the box, and defeat the system so they can have the path of least resistance. There has to be a personalized, delicious "carrot" dangled in front of a gifted child in order to get the best of out them. UNLESS you have a bragger....then it's the competition of the getting the highest score that motivates them. I was one of those when I was a kid. These days, no one finds out the score until months later, so there are no bragging rights.
    Our children deserve better than these psychological assessments (PARCC).

  3. The thing that I particularly don't like about the adaptive tests is that we only get a score and don't know which questions the student actually answered, never mind what their answers were, never mind that these are multiple choice questions to begin with....