The Conventional Wisdom About Group Mean Scores
Author:
- 1 December 1995
- journal article
- Published by Wiley in Journal of Educational Measurement
- Vol. 32 (4) , 385-396
Abstract
Not infrequently, investigators assume that reliability for groups is greater than reliability for persons, and/or that error variance for groups is less than error variance for persons. Using generalizability theory, it is shown that this conventional wisdom is not necessarily true. Examples are provided from the course evaluation literature and the performance testing literatureKeywords
This publication has 5 references indexed in Scilit:
- Sampling Variability of Performance AssessmentsJournal of Educational Measurement, 1993
- Generalizability TheoryEducational Measurement: Issues and Practice, 1992
- The Generalizability of Class MeansReview of Educational Research, 1977
- STUDENT EVALUATIONS OF TEACHING: THE GENERALIZABILITY OF CLASS MEANSJournal of Educational Measurement, 1976
- A Long-Term Project in Psychology: Lives Through Time . Jack Block. In collaboration with Norma Haan. Bancroft, Berkeley, Calif., 1971. xxii, 314 pp., illus. $12.50.Science, 1972