(symbol: MSE) the average amount of error variance within a data set, given as the typical squared distance of a score from the mean score for the set. Mean square error may be calculated in both analysis of variance and regression analysis. In the former it is referred to more specifically as the within-groups mean square and used as the denominator when calculating an F ratio; in the latter it is known as a residual mean square (or mean square residual) and gives the mean difference between actual scores and those predicted by a regression model. A large mean square error indicates that scores are not homogeneous within groups or are not consistent with prediction, such that there is more “noise” than “signal.” For example, a large mean square error in gender research would show no significant differences between groups of males and groups of females.