Student Assessments, Non‐test‐takers, and School Accountability

Abstract

Much attention has focused recently on using student test scores to evaluate public schools. The No Child Left Behind Act of 2002 requires states to test students and evaluate each school's progress toward having all students meet or exceed state standards. Under the law, however, schools only need to test 95% of their students. When some students do not take the test, variability arises in a school's evaluation as the score of each student who did not take the test remains unknown. Using a statewide assessment administered to 11th graders in Illinois, we investigate this source of variation. In our data, 8% of students do not take the test. By applying a bounding technique to the unknown scores of the non‐test‐takers, we show that classifying schools as failing or passing against some fixed threshold frequently can be misleading. We also provide evidence that some schools may be strategically selecting some students to not take the test and, by so doing, increasing the school's test scores.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.