From Survey Analysis
Jump to: navigation, search

A measure of the extent to which there is a correspondence between the observed values of a Dependent Variable and the predicted values from a Predictive Model. There are two commonly-used and equivalent ways of computing the variance:

  1. The ratio of the varianceof the predicted values to the variance of the observed values. (This formula only works where the predicted values are, in some sense, optimal.)[note 1]
  2. The square of the Pearson's Product-Moment Correlation coefficient.

Also known as

  • R-Squared
  • R^2
  • Proportion of variance explained.

See also

Adjusted R-Square


  1. For example, Maximum Likelihood estimates.

A more up-to-date version of this content is on www.displayr.com.