An energy drink manufacturer wants to find out the effect of different amounts of their product on the maximum anaerobic power level for rowers. Would an experimental or observational study design be more appropriate?