analyses:berkey1995
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revision | Next revisionBoth sides next revision | ||
analyses:berkey1995 [2020/06/26 06:46] – Wolfgang Viechtbauer | analyses:berkey1995 [2021/10/06 15:48] – Wolfgang Viechtbauer | ||
---|---|---|---|
Line 96: | Line 96: | ||
Again, the results match the findings from Berkey et al. (1995): The residual amount of heterogeneity is now $\hat{\tau}^2 = 0.157$ and the estimated model is $log(RR) = -0.6303 - 0.0268 (x - 33.46)$, where $x$ is the distance from the equator (in degrees latitude). The standard errors of the model coefficients are $SE[b_0] = 0.1591$ and $SE[b_1] = 0.0110$. | Again, the results match the findings from Berkey et al. (1995): The residual amount of heterogeneity is now $\hat{\tau}^2 = 0.157$ and the estimated model is $log(RR) = -0.6303 - 0.0268 (x - 33.46)$, where $x$ is the distance from the equator (in degrees latitude). The standard errors of the model coefficients are $SE[b_0] = 0.1591$ and $SE[b_1] = 0.0110$. | ||
- | The amount of variance (heterogeneity) accounted for by the absolute latitude moderator is provided in the output above. It can also be obtained with: | + | The amount of variance (heterogeneity) accounted for by the absolute latitude moderator is provided in the output above. It can also be obtained with:((A warning will be issued with respect to the likelihood ratio test that is also part of this output since LRTs should be based on ML/REML estimation.)) |
<code rsplus> | <code rsplus> | ||
anova(res.RE, | anova(res.RE, |
analyses/berkey1995.txt · Last modified: 2022/10/19 18:55 by Wolfgang Viechtbauer