Abstract
Contrast coefficients are arbitrary weights given to subgroups in a study, allowing for the identification of unexpected subgroup effects. Weights should add up to unit (1), and different contrast coefficient patterns can be tested for best fit of the data. An example of four studies of 10 patients per study assessing the fall in systolic blood pressure after different treatments was used. Meta-analyses including linear contrast testing provided better data fit than did traditional fixed and random effect meta-analysis.
This is a preview of subscription content, log in via an institution.
Notes
- 1.
More information of contrast test models and novel models have been recently published.
References
More information of contrast test models and novel models have been recently published.
Abdi, Williams (2010) Contrast analysis. www.utdallas.edu
Krizan (2010) Behav Res Meth 42:863–870
Shuster (2010) Stat Med 29:1259–1265
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Cleophas, T.J., Zwinderman, A.H. (2017). Contrast Coefficients Meta-analysis. In: Modern Meta-Analysis. Springer, Cham. https://doi.org/10.1007/978-3-319-55895-0_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-55895-0_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-55894-3
Online ISBN: 978-3-319-55895-0
eBook Packages: MedicineMedicine (R0)