Investigating the Potential Effects of Exclusion on Mapping State Proficiency Standards onto NAEP Scales

The NAEP sampling design calls for a random sample of students to be selected from each school participating in the assessment. Students with disabilities (SD) and/or English language learners (ELL) can be excluded from NAEP if, in the opinion of school officials, they cannot meaningfully participate in the assessment, even with the options of accommodations. At present, no adjustment is made for these excluded students in the reported NAEP estimates. Since exclusion rates vary widely across states, concerns have been expressed about the comparability of the estimated state NAEP distributions. To the degree that comparability is impaired, estimates of the NAEP equivalent scores are also affected.

An alternative approach to constructing estimates of state NAEP distributions that are intended to be more comparable has been proposed by several researchers. This approach yields statistics that are called full population estimates (FPEs), which attempt to estimate the state NAEP distribution for all students in the cohort, including those that have been excluded from the NAEP assessment.

FPEs of the state NAEP distributions for reading and mathematics in grades 4 and 8 have been constructed for both the 2003 and 2005 administrations. Employing the methodology described in the R&D Report, NAEP score equivalents to the states’ proficiency standards were calculated using the FPEs in place of the reported estimates of the state NAEP distributions. The results are presented in the following eight tables:

The format of the tables is the same as that in the R&D Report Mapping 2005 State Proficiency Standards Onto the NAEP Scales. Because FPE average scores are in general a few points smaller than the standard estimates, the corresponding NAEP score equivalents are generally smaller than those presented in the R&D Report. As before, comparisons of results between administrations for each subject-grade combination for a state should be made with caution because of changes in relevant state policies that may have occurred between 2002–03 and 2004–05.

Read more about comparing NAEP and state proficiency standards.

Read more about full population estimates.

Last updated 23 September 2009 (EP)