A measure of experimental error. Variance arising from unrecognized or uncontrolled factors in an experiment with which the variance of unrecognized factors is compared in tests of significance.