Experimental designs pose the ability to control all variables and thus provide the most powerful support available for a hypothesis of causation. Post facto designs are generally constructed for reporting purposes only as the researcher has no control over the variables.