The Effects of Heteroscedasticity on Tests of Equivalence

Tests of equivalence, which are designed to assess the similarity of group means, are becoming more popular, yet very little is known about the statistical properties of these tests. Monte Carlo methods are used to compare the test of equivalence proposed by Schuirmann with modified tests of equivalence that incorporate a heteroscedastic error term. It was found that the latter were more accurate than the Schuirmann test in detecting equivalence when sample sizes and variances were unequal.

[1]  Ronald C. Serlin,et al.  Equivalence confidence intervals for two-group comparisons of means , 1998 .

[2]  Satterthwaite Fe.  An approximate distribution of estimates of variance components. , 1946 .

[3]  Carl J. Huberty,et al.  Statistical Practices of Educational Researchers: An Analysis of their ANOVA, MANOVA, and ANCOVA Analyses , 1998 .

[4]  Robert A. Cribbie,et al.  Recommendations for applying tests of equivalence. , 2004, Journal of clinical psychology.

[5]  J. Epstein,et al.  Insuring sample equivalence across internet and paper-and-pencil assessments , 2001, Comput. Hum. Behav..

[6]  A. Bostrom,et al.  Equivalence testing for use in psychosocial and services research: An introduction with examples , 1996 .

[7]  J. L. Rogers,et al.  Using significance tests to evaluate equivalence between two experimental groups. , 1993, Psychological bulletin.

[8]  Donald J. Schuirmann.  A comparison of the Two One-Sided Tests Procedure and the Power Approach for assessing the equivalence of average bioavailability , 1987, Journal of Pharmacokinetics and Biopharmaceutics.

[9]  W. Tryon.  Evaluating statistical difference, equivalence, and indeterminacy using inferential confidence intervals: an integrated alternative method of conducting null hypothesis statistical tests. , 2001, Psychological methods.

[10]  R. Grissom,et al.  Heterogeneity of variance in clinical data. , 2000, Journal of consulting and clinical psychology.

[11]  Holger Dette,et al.  An extension of Welch's approximate t-solution to comparative bioequivalence trials , 1994 .

[12]  B. L. Welch.  THE SIGNIFICANCE OF THE DIFFERENCE BETWEEN TWO MEANS WHEN THE POPULATION VARIANCES ARE UNEQUAL , 1938 .

[13]  T. Micceri.  The unicorn, the normal curve, and other improbable creatures. , 1989 .

[14]  F. E. Satterthwaite.  An approximate distribution of estimates of variance components. , 1946, Biometrics.

[15]  W. Westlake,et al.  Symmetrical confidence intervals for bioequivalence trials. , 1976, Biometrics.

[16]  N. Anderson,et al.  Equivalence and predictive validity of paper‐and‐pencil and computerized adaptive formats of the Differential Aptitude Tests , 1998 .

[17]  H. Scheffé.  Practical Solutions of the Behrens-Fisher Problem , 1970 .

[18]  A R Feinstein,et al.  Claims of Equivalence in Medical Research: Are They Supported by the Evidence? , 2000, Annals of Internal Medicine.

[19]  W. Edwards Deming,et al.  Teaching of Statistics , 1940 .

[20]  A note on Welch's approximate t-solution to bioequivalence assessment , 1996 .

[21]  Leland Wilkinson,et al.  Statistical Methods in Psychology Journals Guidelines and Explanations , 2005 .