Meta-analysis of correlated designs (repeated measures) in the context of neural responses to facial expressions of emotion /
This case study refers to a meta-analysis of correlated designs, specifically a systematic review examining the neural activation produced by the same individuals to different kinds of facial expressions of emotion. This method stands in contrast with most published meta-analyses that examine effect...
| Main Author: | |
|---|---|
| Format: | eBook |
| Language: | English |
| Published: |
London :
SAGE Publications Ltd,
2018.
|
| Series: | SAGE Research Methods. Cases.
|
| Subjects: | |
| Online Access: | Connect to the full text of this electronic book |
| Summary: | This case study refers to a meta-analysis of correlated designs, specifically a systematic review examining the neural activation produced by the same individuals to different kinds of facial expressions of emotion. This method stands in contrast with most published meta-analyses that examine effects between groups or independent designs, but correlated or repeated-measures designs are standard in neurophysiological and neuroimaging studies as they are a more efficient way of collecting data. Effect sizes are efficient summaries of the results of statistical tests as they capture both the magnitude and the direction of the effects, for both significant and non-significant findings. However, independent and dependent effect size statistics need to be addressed at the proper level, similarly to what takes place when selecting the appropriate statistical technique to analyze independent or paired samples. In correlated designs, effect sizes should be corrected for the correlation between the paired variables, but this is often not possible in meta-analytic reviews given that the primary studies do not report this value. Alternatively, it is possible to impute an estimated correlation or to perform a sensitivity analysis over a range of plausible levels of correlation. One recommendation that stems out of this work is the appeal for more detailed reports of statistical results that should anticipate future secondary analyses. |
|---|---|
| Physical Description: | 1 online resource : illustrations. |
| Bibliography: | Includes bibliographical references and index. |
| ISBN: | 9781526434180 1526434180 |