Ad Hoc Committee on NAEP Testing and Reporting of Students with Disabilities and English Language Learners
An Ad Hoc Committee of the National Assessment Governing Board is considering the following options for testing and reporting of students with disabilities (SD) and English Language Learners (ELL) on the National Assessment of Educational Progress (NAEP). The options are not mutually exclusive. Some could go into effect quickly while others would be for medium-term or long-range implementation.
The National Assessment is a representative-sample survey, designed to produce valid, comparable data on the academic achievement of large groups of students. It is prohibited by law from providing results for individual children or schools. The policy options are under consideration because of concern that differences in accommodation and exclusion rates among the states and districts participating in NAEP may jeopardize the fairness and validity of state comparisons and other NAEP data and trends.
- Retain current procedures—Testing conditions on NAEP for SD and ELL follow those on state tests with limited exceptions. Accommodation and exclusion rates posted in appendix of NAEP reports. No adjusted scores or cautionary flags.
- Adopt uniform national rules for accommodations and exclusions
- For Students with Disabilities—Determine testing conditions according to the severity, category, and/or nature of disability or based on brief screener exam.
- For English Language Learners—Determine whether to take NAEP in English by English language proficiency screener. Provide NAEP in Spanish if below cut-score.
- Provide incentive for schools to encourage testing of SD and ELL students by scoring excluded students at the 5th percentile nationwide instead of the current practice of exclusions not affecting group average. Incentive may be needed to accept uniform rules because student participation in NAEP is voluntary by law.
- Conduct targeted testing at ability level
- Offer to all students, using assessment booklets at different levels of difficulty—low, medium, and high.
- Offer less difficult or "accessible" booklets to SD and ELL only. Might be similar in concept to NCLB "alternate assessments" but must be on NAEP scale.
Determine level by brief locator test or percentile score on state assessment. Follow standard testing procedures.
- Adjust scores—Use full population estimates or variant to adjust for exclusions. Present as principal means of reporting in NAEP Report Cards, as alternate presentation in appendix, or as prominent display on NAEP website.
- Add cautionary flags
- For exclusions, if 5 percent or more of sample is excluded from NAEP testing, a cautionary flag would accompany a state's scores. This would be similar to rule in TIMSS and PIRLS international assessments. Might also flag if exclusion rate changed more than 3 percentage points from prior assessment year.
- For accommodations, flag if 10 percent or more of sample is tested under non-standard conditions OR accommodation rate changed more than 5 percentage points from prior assessment year.
- Use "reasonable" target exclusion rates (rather than a uniform rate) that vary by demography and testing practice of states. Flag if actual rates exceed targets or change by a defined margin.
- Research validity of accommodations most widely-used on state tests—Results may expand or reduce the list of accommodations prohibited by NAEP because they alter a fundamental attribute of the assessment, e.g. reading-aloud the reading assessment or allowing calculators on all sections of math. Studies may include extended time to help determine if time should be deemed fundamental.
- Offer a screener exam to determine whether students can "meaningfully participate" in the National Assessment without an accommodation that is provided on state tests but is not permitted by NAEP. Currently, these students are routinely excused from the National Assessment.
- Change rules for IEPs to have NAEP considered separately from state tests—Rules for preparing individualized education programs (IEPs) for SD students may be altered by state action or revised by federal regulation, guidance, or law. Separate consideration for participating in NAEP from IEP for state tests because the National Assessment is required to produce valid representative-sample group results for the nation, states, and urban districts and may not provide data or impose consequences on individual students and schools.
- Make minor changes in NAEP report language and placement of information about exclusions and accommodations.