The Influence of Student Sampling Plan on Standard Error for School PAAC.
Saved in:
| Title: | The Influence of Student Sampling Plan on Standard Error for School PAAC. |
|---|---|
| Language: | English |
| Authors: | Lee, Guemin, Fitzpatrick, Anne R. |
| Peer Reviewed: | N |
| Page Count: | 23 |
| Publication Date: | 2001 |
| Sponsoring Agency: | Office of Educational Research and Improvement (ED), Washington, DC. Council of Chief State School Officers, Washington, DC. |
| Contract Number: | R279A50006 |
| Document Type: | Reports - Research Speeches/Meeting Papers |
| Descriptors: | Academic Achievement, Cutting Scores, Elementary Education, Elementary School Students, Error of Measurement, Estimation (Mathematics), Generalizability Theory, Junior High School Students, Mathematics Tests, Performance Factors, Sampling |
| Geographic Terms: | U.S.; California |
| Abstract: | The percentage of students at/above a cut point (PAAC) is one of the most common measures used for reporting school-level performance relative to a proficiency standard (L. Cronbach, N. Bradburn, and D. Horvitz, 1994). The two purposes of this study were to introduce procedures for estimating standard errors for school PAACs under a generalizability theory model and to examine the influence of different student sampling plans on the standard errors. The tests used were mathematics tests for grades 4 and 8 from a statewide assessment. More than 25,000 students took each test form within a grade. A strong relationship between the standard error for school PAAC and the number of students in a school was found. Infinite- and finite-population assumptions for students provide somewhat different standard errors when relatively small numbers of students were used for estimating school PAACs. (Contains 2 tables, 3 figures, and 15 references.) (Author/SLD) |
| Notes: | Paper presented at the Annual Meeting of the National Council on Measurement in Education (Seattle, WA, April 11-13, 2001). |
| Journal Code: | RIEAUG2001 |
| Entry Date: | 2001 |
| Accession Number: | ED451256 |
| Database: | ERIC |
| Abstract: | The percentage of students at/above a cut point (PAAC) is one of the most common measures used for reporting school-level performance relative to a proficiency standard (L. Cronbach, N. Bradburn, and D. Horvitz, 1994). The two purposes of this study were to introduce procedures for estimating standard errors for school PAACs under a generalizability theory model and to examine the influence of different student sampling plans on the standard errors. The tests used were mathematics tests for grades 4 and 8 from a statewide assessment. More than 25,000 students took each test form within a grade. A strong relationship between the standard error for school PAAC and the number of students in a school was found. Infinite- and finite-population assumptions for students provide somewhat different standard errors when relatively small numbers of students were used for estimating school PAACs. (Contains 2 tables, 3 figures, and 15 references.) (Author/SLD) |
|---|
