Public schools are often criticized for the poor performance of American students on the National Assessment of Educational Progress (NAEP). Is it the performance of students or the definition of proficient?
I recently read an excellent article by James Harvey in School Administrator. He outlined the challenges created by the proficient standard for NAEP. Most people consider “proficient” work to be performance on grade level. However, the United States Department of Education has consistently noted that grade-level performance is aligned with the “basic” standard of NAEP. If we look at the basic level for NAEP, nearly 82 percent of American students are actually performing on grade level. Many organizations, including the National Academy of Sciences, have indicated that NAEP’s proficiency benchmark is of doubtful validity. Some have gone so far as to call it “wishful thinking.”
Statistical analyses have investigated use of the NAEP proficiency benchmark for international comparisons. The result? No nation was able to demonstrate that 40 percent of its 4th graders are proficient in reading by NAEP’s definition. In this case, American students are competing well with their international peers. And yet, American public schools are repeatedly criticized for “poor performance” on NAEP.
The challenge of defining proficiency occurs in state assessments like PARCC as well. The standards for PARCC closely align with the standards for NAEP. So, achieving a level 3 performance on a PARCC test could indicate a student is performing on grade level; however, because a student has to achieve a level 4 to be deemed at a “proficient” level, many critics use the PARCC results to claim that our schools are not performing well. It becomes a vicious cycle.
It is time for policy makers and educators to discuss the implications of our use of testing and proficiency standards. We need an honest conversation. Even the College Board re-evaluated the standard it was using to predict college readiness. It is time for NAEP, PARCC, and others to do the same.