The Common Core And Failing Schools

In observing all the recent controversy surrounding the Common Core State Standards (CCSS), I have noticed that one of the frequent criticisms from one of the anti-CCSS camps, particularly since the first rounds of results from CCSS-aligned tests have started to be released, is that the standards are going to be used to label more schools as “failing," and thus ramp up the test-based accountability regime in U.S. public education.

As someone who is very receptive to a sensible, well-designed dose of test-based accountability, but sees so little of it in current policy, I am more than sympathetic to concerns about the proliferation and misuse of high-stakes testing. On the other hand, anti-CCSS arguments that focus on testing or testing results are not really arguments against the standards per se. They also strike me as ironic, as they are based on the same flawed assumptions that critics of high-stakes testing should be opposing.

Standards themselves are about students. They dictate what students should know at different points in their progression through the K-12 system. Testing whether students meet those standards makes sense, but how we use those test results is not dictated by the standards. Nor do standards require us to set bars for “proficient," “advanced," etc., using the tests.

That said, it is true that the new CCSS-aligned tests will, in most states, result in lower proficiency rates than did the old tests, and I have no doubt that many (but not all) policymakers and advocates will use those lower rates to ratchet up their "failing schools" rhetoric.

But that only means that they are misinterpreting the results by conflating student and school performance – that is, operating under the mistaken assumption that how highly students score on tests, and/or whether they meet cut scores that determine their “proficiency," is entirely (or even mostly) a function of school effectiveness.

When they argue that the Common Core will result in more “failing schools," opponents of high-stakes testing are, rather ironically, co-opting this flawed assumption when they should be opposing it, regardless of which standards are in place. They seem to be taking for granted the existence of a problem, instead of confronting its cause – the fact that this is a misinterpretation of the data, one which has plagued U.S. public education for a long time. In fact, to whatever degree testing data can be used to gauge schools' actual effectiveness, proficiency cut scores need not (I would argue should not) play any role in those measures.

Based on my prior experience, I would anticipate the following reaction to these points: Matt, you’re being naïve if you believe that there’s any hope policymakers will use the standards correctly and interpret the data appropriately. They have shown time and time again that they are unwilling or unable to do so.

That may very well be true, but, again, it doesn’t change a simple fact: This misinterpretation is already causing harm under the old/existing standards. Anyone who opposes education policies simply because some people misinterpret testing data surrounding those policies will have trouble finding anything to support.

Now, to be clear: If your argument is that the Common Core is poorly-designed vis-a-vis existing systems, or that there's insufficient evidence it will work better, or any other potentially falsifiable argument about the actual standards, I fully respect those arguments. We should continue to debate (and test) them. This is all very new, and empirical evidence on the CCSS is extremely scarce.

If, on the other hand, you make no mention about the quality or efficacy of the standards, and focus instead on how the testing results will be interpreted, you're having a different conversation, one that actually obscures the conversation we should be having. The Common Core will not create -- cannot create -- any "failing schools." Arguing against the CCSS based on whether or not it will exacerbate the "narrative of failure" is in many respects an implicit endorsement of the latter's shaky measurement premises, and contributes little of substance to a discussion that really could use some additional substance.

One final note: I'm hoping the new tests will serve as an opportunity to increase awareness of the fact that the proportion of students who score above the proficiency line in any given year tells us very little about schools' actual test-based performance. If, for example, a school's proficiency rate is 70 percent in one year and 35 percent the next year (due to new standards/tests being adopted), that strikes me as a perfect time to point out how the school's actual effectiveness could not possibly have changed so much over the course of a single year, and that perhaps the measures we are using to judge that effectiveness are not appropriate for this purpose. I seriously doubt that's what will happen, but one can hope.

- Matt Di Carlo