noun
: an experiment for checking the results of other experiments by maintaining the same conditions except in some one particular and thus inferring the causal significance of this varied factor — compare blank determination , controlled experiment