07-30-2016 04:16 PM
I'm a Phd student and I'm doing a meta-analysis with a dataset where we are collecting data from some studies performed in my State; in this dataset, we have 360 observations where each observation is the average of two, three or four evaluations. However, I don't have the original data, so I can't evaluate sampling variance from each study. However, I'd like to measure, for each variable, some kind of variance between experiments, because I'd like to know if I must exclude some observations or if I can keep all them.
I hope that you can understand what I'm trying to explain and there is some help with this dataset.
Since now, thanks!
07-30-2016 05:48 PM
Although you may get some help here this sounds like a statistical methodology question. Given that, you may get a faster response if you post on cross validated stats.stackexchange.com.