1996
DOI: 10.1037/1082-989x.1.2.170
|View full text |Cite
|
Sign up to set email alerts
|

Meta-analysis of experiments with matched groups or repeated measures designs.

Abstract: Tests for experiments with matched groups or repeated measures designs use error terms that involve the correlation between the measures as well as the variance of the data. The larger the correlation between the measures, the smaller the error and the larger the test statistic. If an effect size is computed from the test statistic without taking the correlation between the measures into account, effect size will be overestimated. Procedures for computing effect size appropriately from matched groups or repeat… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

6
1,132
1
8

Year Published

1998
1998
2024
2024

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 1,535 publications
(1,147 citation statements)
references
References 31 publications
6
1,132
1
8
Order By: Relevance
“…It is important to note that although the repeated measures are correlated, the standardizer can be obtained by treating the cell variances as independent estimates of the population variances (Dunlap, Cortina, Vaslow, & Burke, 1996). Option B is consistent with the intent of the approach recommended by Glass et al (1981, p. 120) for calculating the standardizer.…”
Section: Within-subjects Designsmentioning
confidence: 79%
“…It is important to note that although the repeated measures are correlated, the standardizer can be obtained by treating the cell variances as independent estimates of the population variances (Dunlap, Cortina, Vaslow, & Burke, 1996). Option B is consistent with the intent of the approach recommended by Glass et al (1981, p. 120) for calculating the standardizer.…”
Section: Within-subjects Designsmentioning
confidence: 79%
“…The goal of the meta-analysis is to obtain a pure number (free of the original measurement unit), called the effect size, which is an index of the relation between treatment and outcome that can be compared across studies (for further detail see Rosenthal, 1991;Schulze, 2004;Wolf, 1986). The effect size used in this study is the standardized mean difference d as defined by Cohen (Cohen, 1977) applied to the correlated design (Dunlap, Cortina, Vaslow, and Burke, 1996). This is accomplished by standardizing the raw effect size expressed in the measurement unit of the dependent variable (e.g.…”
Section: Resultsmentioning
confidence: 99%
“…The t-test values for correlated observations (paired t-test) are larger than t values for independent observations (unpaired t-test) because the existence of the correlation between measures reduces the standard error of the difference between the means, making the difference across conditions more prominent. Dunlap et al (Dunlap et al, 1996) argue that this correlation does not change the size of the effect when it is calculated with the mean and the standard deviation of the mean. It simply makes the effect more noticeable.…”
Section: Meta-analyses Of Multiple Experimentsmentioning
confidence: 99%
See 1 more Smart Citation
“…Results showed that moderate mean-level (Dunlap et al, 1996). YGPI = Yatabe-Guilford Personality Inventory.…”
Section: Mean-level Changementioning
confidence: 99%