Fixed- and random-effects models in meta-analysis

被引:2368
作者
Hedges, LV [1 ]
Vevea, JL
机构
[1] Univ Chicago, Dept Educ, Chicago, IL 60637 USA
[2] Univ Chicago, Dept Psychol, Chicago, IL 60637 USA
[3] Univ Chicago, Dept Sociol, Chicago, IL 60637 USA
[4] Univ N Carolina, Dept Psychol, Chapel Hill, NC USA
关键词
D O I
10.1037/1082-989X.3.4.486
中图分类号
B84 [心理学];
学科分类号
04 [教育学]; 0402 [心理学];
摘要
There are 2 families of statistical procedures in meta-analysis: fixed- and random-effects procedures. They were developed for somewhat different inference goals: making inferences about the effect parameters in the studies that have been observed versus making inferences about the distribution of effect parameters in a population of studies from a random sample of studies. The authors evaluate the performance of confidence intervals and hypothesis tests when each type of statistical procedure is used for each type of inference and confirm that each procedure is best for making the kind of inference for which it was designed. Conditionally random-effects procedures (a hybrid type) are shown to have properties in between those of fixed- and random-effects procedures.
引用
收藏
页码:486 / 504
页数:19
相关论文
共 18 条
[2]
CHANG L, 1992, THESIS MICHIGAN STAT
[3]
METAANALYSIS IN CLINICAL-TRIALS [J].
DERSIMONIAN, R ;
LAIRD, N .
CONTROLLED CLINICAL TRIALS, 1986, 7 (03) :177-188
[4]
Hedges L.V., 2014, STAT METHODS METAANA
[5]
CLUSTERING ESTIMATES OF EFFECT MAGNITUDE FROM INDEPENDENT STUDIES [J].
HEDGES, LV ;
OLKIN, I .
PSYCHOLOGICAL BULLETIN, 1983, 93 (03) :563-573
[6]
METAANALYSIS [J].
HEDGES, LV .
JOURNAL OF EDUCATIONAL STATISTICS, 1992, 17 (04) :279-296
[7]
A RANDOM EFFECTS MODEL FOR EFFECT SIZES [J].
HEDGES, LV .
PSYCHOLOGICAL BULLETIN, 1983, 93 (02) :388-395
[8]
Hedges LV., 1981, J EDUC BEHAV STAT, V6, P107, DOI [10.3102/10769986006002107, DOI 10.3102/10769986006002107, 10.2307/1164588]
[9]
Hunter J.E., 1990, METHODS METAANALYSIS