a computation of the probability that an observed effect or difference may have arisen purely as a result of chance or as a result of experimental error —note 1. if this probability is less than an agreed small value (termed the level of significance and often accepted as 1:20), the effect or difference is said to be significant, leading to a decision to accept or reject the statistical hypothesis under consideration —note 2. a nonsignificant result does not, however, constitute a proof that the result is due to chance only —see variance-ratio testThis definition last updated 09/25/2008