Randomization

Overall effect moderated by the presence of randomization.

estimate stderror meta tau
0.529 0.109 Observational 0.581
0.467 0.039 Randomized 0.327

COMMENT: Observational studies seem to give somewhat higher ESs but not significantly so. Note especially huge heterogeneity of observational studies. That causes the huge SE of MA estimate, primarily producing ns difference btw those two sets.

Wald test p-value

Testing the difference in the uncorrected MA estimates between effects from observational and randomized studies.

## [1] 0.5919586

Contour enhanced funnel plots

Year of Publication

## 
## Attaching package: 'ggplot2'
## The following objects are masked from 'package:psych':
## 
##     %+%, alpha

Linear mixed-effects model (taking into effect clustering of ESs due to originating from the same study. Using square root of variance to make them normal)

Estimate Std. Error t value
(Intercept) -0.089 0.092 -0.966
scale(Publication.Year) -0.351 0.088 -3.967
scale(H5.Index.GS.Journal.March.2016) -0.202 0.096 -2.104

Comment: all the variables were centered for easier interpretation of model coefficients. See the beta for Publication Year (-.351). The higher the publication year, the lower the variance (better precision), controlling for H5.

Thus, practices regarding the precision of studies (mainly due to N) seem to have improved throughout last years.

Scatterplot

Size of the points indicate the H5 index (the bigger the higher) of the journal that the ES is published in.