Homoscedasticity is a statistical term which means the homogeneity of variance. It is a necessary condition for the large majority of parametric statistics. Its meaning varies with the context:

• In the univariate case, it means that the expected value of the variance of a statistical population is the same when that population is sampled repeatedly.
• In the regression context, it means that the variance of the dependant variable is the same for all values of the independant variables.
• In the analysis of variance context, it means that the variances are equal for all groups.