By Jason Burns
A criticism of the No Child Left Behind Act (NCLB) was that it encouraged a “race to the bottom.” Though it set a lofty goal of 100% proficiency by 2013, it allowed states to design their own assessments and set their own standards for determining proficiency. As a result, states had an incentive to create easy tests and set low thresholds for proficiency which enabled them to more easily meet their goals.
A strength of the Common Core assessments that were administered for the first time this spring, advocates claim, is that using common assessments will make it easier to compare student performance across states. Ideally, policymakers and educators should then be able to use this information to improve the education that students receive.
But as results from the Common Core assessments are being released, it appears that states may still be overstating the performance of their students.
According to the Partnership for Assessment of Readiness for College and Careers (PARCC), one of the two Common Core testing consortia, five million students in 11 states and the District of Columbia took the PARCC assessments earlier this year. And this fall, results from the administration of those assessments have been released by participating states.
Earlier this year, PARCC went through an arduous process (see here and here) to set recommended cut scores, categories of achievement that are used to interpret students’ raw scores and classify them in one of five levels of achievement. As states have released their assessment results, though, widely divergent results between states suggests that some are setting their own, lower, cut scores to inflate the achievement of their students.
In October, PARCC results released by the Ohio Department of Education showed that most students were proficient in all grades and subjects, from a low of 51% in 8th grade math to a high of 90% in Integrated Math II. On its face, this suggests that Ohio students are high achieving. However, criticism quickly emerged that the Ohio Department of Education had lowered cut scores to give the appearance of higher achievement. In Louisiana, just 32% of 8th graders demonstrated math proficiency and in New Jersey, only 24% of students did so.
Later in October, Arkansas was also accused of manipulating PARCC results, but state officials changed course and later decided to use the cut scores set by PARCC.
The Implications of Inflated Scores
While some have argued that the test-related issues that have been observed this fall serve to undermine the credibility of the Common Core movement in general, such concern may be overblown. It is certainly concerning that states may be deliberately misrepresenting the performance of their students as this may create a false sense of achievement for students as well as educators. At the same time, it is much easier to address this now than when states were able to create their own assessments and set their own standards.
Students are still taking the same assessments and receiving the same score regardless of the state in which they took the test, meaning the issue is not with the assessments themselves or the manner in which they are scored. Rather, the issue lies in how states are choosing to represent this information. To promote greater transparency, families, educators, and other stakeholders could pressure states to use the cut scores set by PARCC when determining achievement or publish the cut scores states used so that more accurate interpretations of students’ achievement can be made.