A widespread critique of high-stakes testing – regardless of the test involved – is that it distorts the incentives of educators, from system leaders down to teachers. One particularly well-documented phenomenon is the practice of redirecting (scarce) resources towards those students on the threshold of whatever arbitrary bar has been deemed the cutoff for the test’s high-stakes sanction.
The ability to use data to diagnose and target students’ needs is important. But from a pedagogical or scientific perspective, there is no reason to give threshold students any more focus and assistance than those who scored below them – who may need help even more urgently – or than those who scored a few questions above them but whose skills may be at a very similar level. Certainly, focusing on threshold students does not help establish “high standards” for every child. But given the perverse incentives created by a system of high-stakes testing, in which the outcome that matters is how many students cross a particular cutoff point, it is simply rational resource allocation for administrators and teachers to zero in on those students who are right on the edge of clearing the bar. In the case of the new NECAP graduation requirement, in which the cutoff for a diploma is a line between a score of 1 and 2, we are talking about the students who scored a “high 1.”
On Wednesday night, at a forum focused on the NECAP graduation requirement, a representative of the Rhode Island Department of Education (RIDE) may have inadvertently admitted that encouragement for these practices can be found even at the highest levels of the system.
Andrea Castaneda, Chief for Accelerated School Performance at RIDE, was defending the new testing policy when she spotlighted the remedial math/college credit program RIDE organized at CCRI for 100 at-risk students this summer. The crux of her comments was that this was a great service that RIDE was able to deliver thanks to the urgency created by placing over 4,000 students at risk of not graduating from high school; in other words, “Look, good things are happening because of this policy.”
We have heard this story before. It seems to be one of RIDE’s main talking points in support of the NECAP graduation requirement, and at the Board of Education’s August retreat it constituted Ms. Castaneda’s closing pitch after a long presentation focused on selling the policy to on-the-fence Board members.
From what I have heard, the CCRI program was a good one. Certainly, a class for 100 students is not a real response to a crisis facing thousands, but it definitely seems like a positive program and we should be glad it was offered.
What I had not heard before, however, was who was targeted to participate in these classes. On Wednesday, Ms. Castaneda let this information slip, explaining that RIDE asked local districts to identify, and I quote directly, “high 1s” to join the CCRI program.
High 1s. The threshold kids.
Clearly, the “high 1s” do need extra supports and assistance to fill gaps in their math skills. But then again, so do the “low 1s,” and so, presumably, do the “low 2s.” What is so worrying about this statement from a high-ranking RIDE official is that it calls into question the basic talking point that the NECAP graduation requirement is ensuring all students have the skills and knowledge they need to succeed. It suggests that instead of helping all students cross the bar of proficiency (which would welcome all 1s, if not 2s as well), RIDE targeted their extra training to those falling just an inch short. To learn that these statistical games may be happening in a program administered at the state level is very concerning.
Of course, when I spoke with Ms. Castaneda after the forum she backtracked. What she had really meant, she said, was that RIDE asked districts to identify students on the higher-performing end of the at-risk population. The really low scorers, she explained, probably did not have enough math skills to be able to learn from the remedial math classes offered at CCRI anyway. But she said all that would have taken too long to explain, so she had simply used the shorthand “high 1s.”
Whatever the case may be, we should take this as a reminder that the distorting effects of high-stakes testing continue to crop up. Education should not be about getting students to jump arbitrary hurdles. RIDE is absolutely right when they say we should be working to ensure every student has the supports they need to succeed, starting in pre-kindergarten and continuing until high school and beyond. But if a policy sets an arbitrary bar as an obstacle to graduation at the eleventh hour, RIDE must be ready to deal with the perverse ways this incentivizes educators to game the system. And RIDE should certainly not be engaging in these practices itself. There are already enough games playing themselves out in Rhode Island classrooms every day because of this policy, and not to the benefit of our students.