It's hard to implement the ideal experiment and have it go well, so expts in ecology are not clearly "expt" vs "observational" and are often the latter.. 2) Within observational study designs, some only account for observable confounding variables. Others deal with unobservables ones too. 3/n
