What I want to do is combine the results of several experiments. I have two groups (treated- non-treated), one measurable variable, and several experiments that I want to analyse together to determine if the treatment give rise to a significant difference in the measured variable.
I don't have it in front of me, but most of the analysis is standard statistics so I think you can do it. However, it doesn't have a "Meta Analysis" function by definition I think,
If I have 3 columns: "Measurable variable", Treatment and Experiment. Would you use a Fit model to analyse that? Giving the "Experiment" variable a Random effect in attributes? Would that be correct?