The conventional wisdom about group mean scores

被引:21
作者
Brennan, RL [1 ]
机构
[1] AMER COLL TESTING,IOWA CITY,IA
关键词
D O I
10.1111/j.1745-3984.1995.tb00473.x
中图分类号
G44 [教育心理学];
学科分类号
0402 ; 040202 ;
摘要
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 literature.
引用
收藏
页码:385 / 396
页数:12
相关论文
共 9 条
[1]  
Brennan R. L., 1992, ED MEAS, V11, P27
[2]  
BRENNAN RL, 1992, ELEMENTS GENERALIZAB
[3]  
CRONBACH LJ, 1972, DEPENDABILITY BEHAVI
[4]  
Feldt L.S., 1989, ED MEASUREMENT, P105
[5]  
Joint Committee on Standards for Educational Evaluation, 1981, STAND EV ED PROGR PR
[6]   STUDENT-EVALUATIONS OF TEACHING - GENERALIZABILITY OF CLASS MEANS [J].
KANE, MT ;
GILLMORE, GM ;
CROOKS, TJ .
JOURNAL OF EDUCATIONAL MEASUREMENT, 1976, 13 (03) :171-183
[7]   GENERALIZABILITY OF CLASS MEANS [J].
KANE, MT ;
BRENNAN, RL .
REVIEW OF EDUCATIONAL RESEARCH, 1977, 47 (02) :267-292
[8]   SAMPLING VARIABILITY OF PERFORMANCE ASSESSMENTS [J].
SHAVELSON, RJ ;
BAXTER, GP ;
GAO, XH .
JOURNAL OF EDUCATIONAL MEASUREMENT, 1993, 30 (03) :215-232
[9]  
SHAVELSON RJ, 1991, GENERALIZABILITY THE