The Complicated Economics of Testing in the Era of Common Core Standards
Two days earlier this month were extremely educational.
On January 9, I was with a group of state leaders, convened by CCSSO, wrestling with Common Core implementation and all of the issues tied to it (assessments, accountability, human capital, etc.).
The combined effect of these events is my heightened concern about the chances for the kind of implementation that produces the ground-shaking results so many have forecast.
The good news is that lots of talented people are engaged in this work. That gives me hope.
The bad news is that the work is extremely complicated and this is getting obscured by lots of sycophantic cheerleading. (Two bright spots from today: Rick Hess’s general skepticism about CC implementation and USED’s Joanne Weiss’s encouragement of a “continuous improvement” mindset that will accept setbacks and lead to course corrections.)
I’m going to write more about these matters in the days to come. But for now, I’d like to call your attention to a recent report. Before I went to work for a state department of education, I probably wouldn’t have read “Strength in Numbers,” the study on state assessments by Brookings’ Matthew Chingos.
But my time at an SEA and today’s CC struggles have taught me just how important and complex (and underappreciated) assessment issues are. Tests are expensive and they serve as the cornerstone for numerous other high-priority activities. Most importantly, Common Core’s fate is tied to that of the two testing consortia working feverishly to prepare for administration in 2014–15.
Chingos’s paper focuses on costs specifically, but it touches on other matters as well. One of the big takeaways is just how much variation there is in state assessment costs. The best predictor of a state’s testing expense is its enrollment (and that is the only variable that is predictive). But, importantly, per-pupil costs are negatively correlated with enrollment. States get major savings from economies of scale. There are huge fixed costs associated with test development, for example, that are similar whether you have 100,000 kids or 10 million.
A key lesson is that if a state wants to limit costs, rather than reducing the number of test items or relying more on multiple-choice questions (both of which can compromise test quality), it might join a multi-state consortia, which spreads fixed costs around. Good news for the two consortia.
Another interesting data point is that while we spend approaching $2 billion annually on state tests, this is a miniscule fraction of the total US K–12 spending. Given their importance, perhaps we ought to be willing to spend more on tests.
Something you might not know is that federal funding for PARCC and SB runs out six months before their tests are first administered, meaning participating states must pony up before the virtues of these common assessments are fully realized. This is a big deal.
Holding the consortia together is going to be tough sledding, as there are multiple centrifugal forces that will try to pull them apart: disagreements over cut scores, new political leadership in the states, possible cost overruns, problems with technology, and, probably most powerful of all, inertia—the fact that states have generally run their own standards-and-assessments regimes without much external meddling.
The testing consortia are similar to Common Core itself in an important way: For state leaders, their early days were full of political revenue and virtually devoid of cost. In days gone by, leaders could happily talk about benefits, and they did so with relish (consider how many times you’ve heard the terms “dramatic” and “transformational” thrown around).
But now, states are realizing more each day just how difficult it is to implement CCSS in a way that will be different than the hundreds of previous standards and assessments efforts states have endured in years gone by.
The same is about to be true of common assessments. The costs—political, practical, financial, and otherwise—are about to hit. I think that the chickens will come home to roost when states have to budget for the first time for the costs of the consortia. That’s not far away, and it will force them to do some calculations about whether the upsides of being a member of PARCC or SB outweigh the downsides.
I’m not predicting anything dire yet. But fast forward to a year from now.
There will be lots of governors and state chiefs in office who ascended after CC was adopted and will therefore feel less loyalty to it. They will be aware that the new standards are tougher, that their teacher education programs aren’t adequately preparing graduates for this shift, and that professional development programs aren’t up to snuff.
They’ll be told that their schools’ test scores are about to fall off a cliff and that huge numbers of teachers, thanks to new evaluation systems, are about to be rated as ineffective.
Then they’ll be told, “Oh, by the way, please cut a big state check to pay for this wonderful new reality.”
In other words, assessments, beyond being technically complicated to produce and administer, may very well determine the future of Common Core.
This blog entry first appeared on the Fordham Institute’s Flypaper blog.