The amount of variance accounted for by treatment can be estimated with omega squared or with the squared multiple correlation coefficient. Monte Carlo methods were employed to compare omega squared, the squared multiple correlation coefficient, and the squared multiple correlation coefficient to which a shrinkage formula had been applied, in terms of bias and precision. The squared multiple correlation coefficient to which the shrinkage formula had been applied and omega squared produced estimates which were negligibly biased. The bias in the squared multiple correlation coefficient, while consistently positive, decreased as the sample size increased and was too small to be of practical importance when the sample size was 50 or larger. Omega squared, the squared multiple correlation coefficient and the squared multiple correlation coefficient to which the shrinkage formula had been applied were all most precise with large samples and least precise when treatment effects were moderate in magnitude. (Author/MV)