Dear users,

My question is – is there a way to calculate confidence intervals that takes into account the fact that some effects are nested within studies and are therefore dependent?

You're not really giving us much to go on.

The whole reason for using a multi-level model is to ensure a more or less robust way to estimate the higher level parameters - or fixed effects.

You see in your model, you could be stating that within groups effect size is related to "covariate", though groups can have a variable intercept (if the covariate is numeric). Or that effect sizes are dependent in part on a covariate and in part on group level variation (captured by the random effect) - if the covariate is a factor.

In both cases, if your specified hierarchical structure is correct, you are already correcting for the fact that effects are nested within studies simply due to the fact that you are using a multi-level model.

So either you're not telling us everything we need to know, or your problem is already solved and you just need to calculate CI.

Lets assume it's the latter, which means you should look at ?confint.merMod

Hope this helps,