[R-meta] How to set the sampling variance when the Effect Size is Fisher-Zr?
Viechtbauer, Wolfgang (SP)
wolfg@ng@viechtb@uer @ending from m@@@trichtuniver@ity@nl
Thu Oct 25 12:40:46 CEST 2018
To simulate a correlation, you have to simulate raw data for 2 variables with a given true correlation, and then you compute the observed correlation. mvrnorm() from MASS allows you to simulate the raw data from a multivariate normal distribution and then use cor() to get the observed correlation. Repeat for k studies. Then you can apply the r-to-z transformation to the observed correlations.
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of FANG JUNYAN
Sent: Thursday, 25 October, 2018 12:36
To: r-sig-meta-analysis using r-project.org
Subject: [R-meta] How to set the sampling variance when the Effect Size is Fisher-Zr?
I want to do a simulation study about how the Meta-regression performs when two moderators included.
The equation is yi<-0+xi*b1+xj*b2+e.
yi refers to the outcome variable(ES), 0 is the intercept, b1 and b2 are moderators, xi and xj are regression coefficients, e is the sampling variance.
Since most empirical meta-analysis study take Pearson Correlation as Effect Size, I want to take Fisher-Zr as the yi. While most of the simulation studies I read took SMD or OR as the outcome variable, I was wondering how to generate the e when the ES was Fisher-Zr.
More information about the R-sig-meta-analysis