What The Standards-Based Movement (Really) Got Wrong

Late last year, Education Week published a commentary called What the standards-based movement got wrong. While I think the title was provocative, the author missed the boat in her diagnosis of the problems of the standards movement.

This is my attempt to answer the same question. I have written similar pieces before, but this one is more focused on standards than accountability, and it is also more oriented toward diagnosing the problems of standards than it is pointing out the successes (of which I think there are many—see here for a three-year-old-but-still-quite-accurate summary). 

So, what do I think the standards movement has gotten wrong in its two decades, and how can the problem be resolved?

1. The standards themselves have sometimes been quite poor. The list of criticisms has been long: they’re vague, too focused on practices, not focused enough on content, too procedural, poorly constructed in certain mathematics content areas (e.g., fractions). Of course, if the standards themselves are bad, then the standards-based movement will almost certainly fail to achieve its desired goals.
2. The tests have been even poorer than the standards. For instance, state tests during the NCLB era were poorly aligned with state standards and disproportionately focused on certain easy-to-test content at the expense of others. They were excessively procedural (even relative to the excessively procedural standards), and their all-multiple-choice format probably didn’t help.
3. The design of accountability systems was also very poor. These systems have almost always measured school performance based on status, rather than growth (and the worst possible measure of status, at that—proficiency rates), all but ensuring that schools serving poor children will be targeted, regardless of their actual effectiveness. These design issues have all but ensured that negative unintended consequences, such as teaching to the test (or, more pervasively, cheating), would appear in these kinds of schools (though the magnitude of these consequences has likely been overstated).
4. The interventions that have gone with these accountability systems have generally not been effective. For instance, the public school choice policies under NCLB didn’t work because few students had reasonable within-district non-failing alternatives to transfer to. And the supplemental education services under NCLB didn’t work because students didn’t attend enough hours and because the services were not provided in the form of tutoring that works best. Even related policies, such as School Improvement Grants, have not been particularly effective on average, though they have been more effective in particular places.

Still, with all of these critiques, I remain a supporter of standards-based reform.

There is a great deal of evidence that standards-based accountability can work to improve student outcomes (see my similar pieces above for reviews of this evidence or see this review by Figlio and Loeb). And this is in spite of all the issues raised above. There’s also good conceptual reason to think that some form of standards can improve teaching and learning. The alternative to a standards-driven system, given our decentralized governance structures, is a system where we have 10,000+ districts, 100,000+ schools, and 3,000,000+ teachers creating their own curricula. Suffice it to say that I’m skeptical that there exists the capacity to do curriculum well at this kind of scale, even if I thought it was efficient. In short, standards put some guardrails on curriculum and instruction that I expect is especially important for historically underserved groups.

Fortunately, I think there are some promising signs that we are moving in the right direction on some of the issues raised above. For instance, it appears that standards are getting better, on average. For instance, math education experts (see for instance here and here) generally regard the standards as more coherent than the state standards they replaced. Also, new tests seem to have applied some lessons from the previous generation—they are more comprehensive in their coverage and do a better job at assessing deeper conceptual understanding than previous renditions. And state accountability systems, though far from perfect, seem to have adjusted based on many of the lessons learned from previous design iterations.

Still, I think there is work to be done. Standards, assessments, and accountability policies will never be perfect. But with continued effort they hold very real promise for improving teaching and learning in America’s schools. Ongoing research, such as that being conducted by IES’s Center on Standards, Alignment, Instruction, and Learning (C-SAIL), on which I am a co-principal investigator, is probing key assumptions of the newest generation of standards-based reform. We are also testing these theories in an ongoing intervention study to improve teachers’ alignment and students’ learning. My hope is that this is just part of a broad research effort oriented toward continually adjusting these policies over time in ways that make them more effective.