While analyzing a data, allocation into similar groups is done to ensure -

Both groups (study and control) should be comparable in respect of all the possible varibles, which may influence the outcome. So that, except for the factor under study will not influence the outcome → confoundation and bias can be minimized.