[R-meta] Moderator analysis test of residual heterogeneity confusion
Viechtbauer, Wolfgang (SP)
wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Tue Sep 24 12:28:24 CEST 2019
Whether analyzing 'one moderator at a time' is a better approach is debatable, but that's a different issue.
So, leaving that issue aside, the results below suggest that all of the heterogeneity (that is not already accounted for by the moderator included in the model) is due to differences between studies (and none of it is due to heterogeneity in the true outcomes within studies). The test for residual heterogeneity even suggests that there is no noteworthy heterogeneity (either within or between studies) left with the moderator included in the model.
From: Mia Daucourt [mailto:miadaucourt using gmail.com]
Sent: Thursday, 19 September, 2019 3:28
To: James Pustejovsky
Cc: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis using r-project.org
Subject: Re: [R-meta] Moderator analysis test of residual heterogeneity confusion
Thank you for the advice! I have added the observation-level random effect and will not run the overfitted model anymore and just stick to one moderator at a time.
How would I interpret the results of the test for residual heterogeneity for the single moderator below?
(P.S. this is the only moderator that does not have a significant value for the between-study variance parameter at the observation level and only does at the study level).
Model2_HMEcomp <- rma.mv(yi, vi., mods = ~ factor(hne_component)-1, random = ~ 1 | study/count, data=Zcalc, test="t")
Multivariate Meta-Analysis Model (k = 456; method: REML)
logLik Deviance AIC BIC AICc
112.1356 -224.2713 -204.2713 -163.2233 -203.7678
estim sqrt nlvls fixed factor
sigma^2.1 0.0136 0.1166 51 no study
sigma^2.2 0.0000 0.0000 456 no study/count
Test for Residual Heterogeneity:
QE(df = 448) = 409.9810, p-val = 0.9007
Test of Moderators (coefficients 1:8):
F(df1 = 8, df2 = 448) = 6.2947, p-val < .0001
Thanks so much for your guidance!
More information about the R-sig-meta-analysis