Skip to content
Prev 2148 / 5636 Next

[R-meta] Overlapping CIs with significant difference among subgroups

Dear Rafael,

To avoid any confusion:

Yes, the standard error is the square root of the *sampling variance*, 
which is the variance of a statistic (for example, a mean, or an 
estimated proportion, or an estimated relative risk, there are many 
other examples of estimates).

Particularly, for a mean the sampling variance is the (estimated) 
population variance, divided by n. The standard error is the square root 
of the sampling variance.

Both the sampling variance and the standard error describe how precise 
an estimate is, which is reflected by their (inverse) dependence on the 
sample size.

By contrast, the population variance and its square root (the standard 
deviation) describe the variability of a measure in a population - which 
is a very different thing.

These are general definitions, there is nothing special with meta-analysis.

Best,

Gerta

Am 04.06.2020 um 16:48 schrieb Rafael Rios: