[R-meta] I squared for multilevel meta-analysis

[Michael Dewey]

Dear Amy

When you look at your data does it seem plausible that there is little 
between-study variability? If so that is your answer. I^2 can be zero.

Michael

On 02/04/2024 03:44, Amy Zadow via R-sig-meta-analysis wrote:
> Hi Wolfgang and experts,
> We are running multilevel meta-analysis in metafor and having difficulties with I squared. We looking at climate aggregated to the group level and its relationship with outcomes such as psychological health. We have an I squared of 2 percent. Have you seen this before and do you have any recommendations?
> Another question - can we use t values (for the bivariate relationship between variables taken from hierarchical linear modeling studies) within the meta-analysis?
> Please let me know if you need more information. Thank you - your help is very much appreciated. Best wishes, Amy
> 
> Dr Amy Zadow Organisational Psychologist  Senior Research Fellow
> PSC Global Observatory | Centre for Workplace Excellence| University of South Australia
> IPC CWE-17 | David Pank Building (DP1-21)| 160 Currie Street
> City West Campus |GPO Box 2471 ADELAIDE SA 5001.
> M/ +61 400 069 335 P/ +61 8 8302 9096  E/ amy.zadow using unisa.edu.au<mailto:amy.zadow using unisa.edu.au>
> 
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-meta-analysis mailing list @ R-sig-meta-analysis using r-project.org
> To manage your subscription to this mailing list, go to:
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis
> 

-- 
Michael