Hi - I am conducting a meta-analysis and I have a matrix of f-statistics, Ns and dfs from a series of studies that tested for an interaction in a 2x2 anova. I'd like to test whether the 2x2 interaction is significant in the aggregate. Similarly, I have a matrix of chi-square statistics that I'd like to meta-analyze. How can I input these test statistics into the "metafor" package and conduct these meta-analyses? I see how to input raw data or standardizes effect sizes (e.g., d) into "metafor", but not test statistics. Best, David -- View this message in context: http://r.789695.n4.nabble.com/Meta-analysis-of-test-statistics-in-metafor-package-tp3868126p3868126.html Sent from the R help mailing list archive at Nabble.com.
Meta-analysis of test statistics in "metafor" package?
2 messages · davidyeager, Kevin E. Thorpe
On 10/03/2011 01:07 PM, davidyeager wrote:
Hi - I am conducting a meta-analysis and I have a matrix of f-statistics, Ns and dfs from a series of studies that tested for an interaction in a 2x2 anova. I'd like to test whether the 2x2 interaction is significant in the aggregate. Similarly, I have a matrix of chi-square statistics that I'd like to meta-analyze. How can I input these test statistics into the "metafor" package and conduct these meta-analyses? I see how to input raw data or standardizes effect sizes (e.g., d) into "metafor", but not test statistics. Best, David
I believe the MAd package may have what you are looking for. It can call metafor for you. Kevin
Kevin E. Thorpe Biostatistician/Trialist, Applied Health Research Centre (AHRC) Li Ka Shing Knowledge Institute of St. Michael's Assistant Professor, Dalla Lana School of Public Health University of Toronto email: kevin.thorpe at utoronto.ca Tel: 416.864.5776 Fax: 416.864.3016