NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED211606
Record Type: RIE
Publication Date: 1981-Jan
Pages: 44
Abstractor: N/A
Reference Count: 0
ISBN: N/A
ISSN: N/A
Effects of Varying Item Domain and Sample Duration on Technical Characteristics of Daily Measures in Reading.
Fuchs, Lynn; And Others
Three related studies were conducted to examine the effects of variations in procedures used for curriculum-based assessment of reading proficiency: the first addressed the question of the influence of sample duration on the concurrent validity of the measure; the second addressed the question of the influence of sample duration on the level, slope, and variability of performance over repeated measurements; and the third examined the effect that varying the size of the pool from which items are drawn has on slope and variability of performance on the measure. Results of the studies provided evidence that sample duration is an important consideration in curriculum-based measurement because of its probable impact on variability and slope. Increasing sample duration from 30 seconds to a three minute sample reduced day-to-day variability in performance and resulted in a more rapid increase in student performance. The results with respect to sampling from domains of differing sizes indicated that measurement samples drawn from smaller domains are more sensitive to variations in instruction, but somewhat more variable. The optimum daily measurement procedure would seem to involve sampling from a pool of stimulus items well beyond that defined by the short-term objectives, but not in excess of an annual goal. (Author/GK)
Editor, IRLD, 350 Elliott Hall, 75 East River Road, University of Minnesota, Minneapolis, MN 55455 ($3.00)
Publication Type: Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: Office of Special Education (ED), Washington, DC.
Authoring Institution: Minnesota Univ., Minneapolis. Inst. for Research on Learning Disabilities.
Identifiers: Test Curriculum Overlap