NotesFAQContact Us
Search Tips
ERIC Number: ED262096
Record Type: Non-Journal
Publication Date: 1985-Apr-3
Pages: 14
Abstractor: N/A
Reference Count: N/A
NAEP Analysis Procedures and Methodology.
Beaton, Albert E.
This paper overviews technical developments in data analysis procedures for the National Assessment of Educational Progress (NAEP) reading data during 1984. The highlight of the reshaping of the NAEP data has been the scaling using item response theory (IRT). AT this point in the data analysis, an IRT-based scale appears appropriate for reading proficiency. A single dimension that spans the three grade levels (4, 8, and 11) and the three age levels (9, 13, and 17) has been located. A report of results on a scale representing a hypothetical test with known properties is in preparation. Effects of changing the administration of exercises from a tape recording to pencil-and-paper have been examined. Data from past NAEPs are being rescaled onto the new reading proficiency scales for analysis of trends. Present technology has been adapted to mesh with the new Balanced Incomplete Block (BIB) spiralling. The following activities are discussed in detail: (1) the multifaceted approach to establishing the dimensionality of the reading parameters; (2) estimation of reading parameters; (3) proficiency imputation; (4) the reading proficiency scale; and (5) trend data. (BS)
Publication Type: Speeches/Meeting Papers; Reports - Descriptive
Education Level: N/A
Audience: Researchers
Language: English
Sponsor: National Inst. of Education (ED), Washington, DC.
Authoring Institution: National Assessment of Educational Progress, Princeton, NJ.
Identifiers - Assessments and Surveys: National Assessment of Educational Progress