Coefficient of Alienation

Perhaps the most bizarre sounding statistical term. The coefficient of alienation (a.k.a., coefficient of non-determination) represents the proportion of variance in the dependent variable that is not accounted for by the independent variable(s). It is the coefficient of determination's counterpart. It is estimated by 1 - r². Effectively, it is a measure of the non-association between two variables.

The coefficient of alienation is used most commonly in the area of psychometrics; specifically, as an element within the formulation of the standard error of measurement. Greater levels of alienation are associated with greater standard error of measurement values. Although perhaps not obvious, the coefficient of alienation is also used in multiple regression in the context of measuring multicollinearity. That is, 'tolerance' is in fact identical to the coefficient of alienation.

See Also: Coefficient of alienation, Tolerance