This study estimated the potential bias from "worst-case" scenarios of selective non-participation in NAEP, and examined the extent to which statistical methods can correct for that bias.
The purpose of this simulation was to assess the improvements in standard errors of estimates that could be expected if students participating in NAEP were pre-assigned to test booklets that were adapted to their level of performance based on their state assessment scores.
This paper is a demonstration of how partitioning analysis can be used to help separate changes in reading and mathematical proficiency from changes in school populations over assessment years, using NAEP data.