Reliability Analysis for the Internationally Administered 2002 Series GED Tests. GED Testing Service[R] Research Studies, 2009-3

Saved in:
Bibliographic Details
Title: Reliability Analysis for the Internationally Administered 2002 Series GED Tests. GED Testing Service[R] Research Studies, 2009-3
Language: English
Authors: Setzer, J. Carl, He, Yi, American Council on Education, GED Testing Service
Source: GED Testing Service. 2009.
Availability: GED Testing Service. Available from: American Council on Education. One Dupont Circle NW Suite 250, Washington, DC 20036. Tel: 202-939-9490; Fax: 202-659-8875' e-mail: ged@ace.nche.edu; Web site: http://www.GEDtest.org
Peer Reviewed: N
Page Count: 24
Publication Date: 2009
Document Type: Reports - Evaluative
Education Level: Elementary Secondary Education
Descriptors: Educational Research, Error of Measurement, Scores, Test Reliability, Item Analysis, International Education, Test Results, Robustness (Statistics), Goodness of Fit, Academic Standards, Comparative Testing, Measurement Techniques, Replication (Evaluation), General Education
Assessment and Survey Identifiers: General Educational Development Tests
Abstract: Reliability Analysis for the Internationally Administered 2002 Series GED (General Educational Development) Tests Reliability refers to the consistency, or stability, of test scores when the authors administer the measurement procedure repeatedly to groups of examinees (American Educational Research Association [AERA], American Psychological Association [APA], and National Council on Measurement in Education [NCME], 1999). If a given test yields widely discrepant scores for the same individual on separate test administrations, and the individual does not change significantly on the measured attribute, then the scores on the test are not reliable. Conversely, if a test produces the same or similar scores for an individual on separate administrations, then the scores from the test are considered reliable. Reliability is inversely related to the amount of measurement error in test scores. That is, the more measurement error present in test scores, the less reliable the test. Reliability is a crucial index of test quality. Standard practices require test developers to evaluate and report the reliability of their test scores. The purpose of this report was to estimate and evaluate the reliability of the internationally administered 2002 Series GED Tests, which have been developed and maintained by the GED Testing Service (GEDTS) since 1963. (Contains 3 tables and 3 footnotes.)
Abstractor: ERIC
Number of References: 12
Entry Date: 2010
Accession Number: ED510052
Database: ERIC
Description
Abstract:Reliability Analysis for the Internationally Administered 2002 Series GED (General Educational Development) Tests Reliability refers to the consistency, or stability, of test scores when the authors administer the measurement procedure repeatedly to groups of examinees (American Educational Research Association [AERA], American Psychological Association [APA], and National Council on Measurement in Education [NCME], 1999). If a given test yields widely discrepant scores for the same individual on separate test administrations, and the individual does not change significantly on the measured attribute, then the scores on the test are not reliable. Conversely, if a test produces the same or similar scores for an individual on separate administrations, then the scores from the test are considered reliable. Reliability is inversely related to the amount of measurement error in test scores. That is, the more measurement error present in test scores, the less reliable the test. Reliability is a crucial index of test quality. Standard practices require test developers to evaluate and report the reliability of their test scores. The purpose of this report was to estimate and evaluate the reliability of the internationally administered 2002 Series GED Tests, which have been developed and maintained by the GED Testing Service (GEDTS) since 1963. (Contains 3 tables and 3 footnotes.)