A Comparison of Analysis of Covariate-Adjusted Residuals and Analysis of Covariance

Various methods to control the influence of a covariate on a response variable are compared. These methods are ANOVA with or without homogeneity of variances (HOV) of errors and Kruskal–Wallis (K–W) tests on (covariate-adjusted) residuals and analysis of covariance (ANCOVA). Covariate-adjusted residuals are obtained from the overall regression line fit to the entire data set ignoring the treatment levels or factors. It is demonstrated that the methods on covariate-adjusted residuals are only appropriate when the regression lines are parallel and covariate means are equal for all treatments. Empirical size and power performance of the methods are compared by extensive Monte Carlo simulations. We manipulated the conditions such as assumptions of normality and HOV, sample size, and clustering of the covariates. The parametric methods on residuals and ANCOVA exhibited similar size and power when error terms have symmetric distributions with variances having the same functional form for each treatment, and covariates have uniform distributions within the same interval for each treatment. In such cases, parametric tests have higher power compared to the K–W test on residuals. When error terms have asymmetric distributions or have variances that are heterogeneous with different functional forms for each treatment, the tests are liberal with K–W test having higher power than others. The methods on covariate-adjusted residuals are severely affected by the clustering of the covariates relative to the treatment factors when covariate means are very different for treatments. For data clusters, ANCOVA method exhibits the appropriate level. However, such a clustering might suggest dependence between the covariates and the treatment factors, so makes ANCOVA less reliable as well.

[1]  C. Borror An Introduction to Statistical Methods and Data Analysis, 5th Ed. , 2002 .

[2]  S. Siegel,et al.  Nonparametric Statistics for the Behavioral Sciences , 2022, The SAGE Encyclopedia of Research Design.

[3]  Rikard Berthilsson,et al.  A Statistical Theory of Shape , 1998, SSPR/SPR.

[4]  R Fisher,et al.  Design of Experiments , 1936 .

[5]  L. Leemis Applied Linear Regression Models , 1991 .

[6]  Michael G. Akritas,et al.  Nonparametric ANCOVA with two and three covariates , 2004 .

[7]  W. Atchley,et al.  Statistical Properties of Ratios. I. Empirical Results , 1976 .

[8]  D. Duvall,et al.  Variation in oxygen consumption of the western diamondback rattlesnake (Crotalus atrox): implications for sexual size dimorphism , 1998, Journal of Comparative Physiology B.

[9]  G. H. Albrecht,et al.  Ratios as a size adjustment in morphometrics. , 1993, American journal of physical anthropology.

[10]  Jerry M. Manheimer,et al.  Anova of Residuals and Ancova: Correcting an Illusion by Using Model Comparisons and Graphs , 1985 .

[11]  R. Kuehl Design of Experiments: Statistical Principles of Research Design and Analysis , 1999 .

[12]  L. Toothaker Book Review : Nonparametric Statistics for the Behavioral Sciences (Second Edition) , 1989 .

[13]  G. Uetz,et al.  Estimating fitness : a comparison of body condition indices , 1996 .

[14]  R. Lyman Ott.,et al.  An introduction to statistical methods and data analysis , 1977 .

[15]  Mark Anderson,et al.  Design of Experiments: Statistical Principles of Research Design and Analysis , 2001, Technometrics.

[16]  C. Kowalski,et al.  ANCOVA for nonparallel slopes: the Johnson-Neyman technique. , 1994, International journal of bio-medical computing.

[17]  J. Neter,et al.  Applied Linear Regression Models , 1983 .

[18]  G. C. Packard,et al.  The Misuse of Ratios, Indices, and Percentages in Ecophysiological Research , 1988, Physiological Zoology.

[19]  Douglas A. Penfield,et al.  The Effects of Type I Error Rate and Power of the ANCOVA F Test and Selected Alternatives Under Nonnormality and Variance Heterogeneity , 2001 .

[20]  Bradley E. Huitema,et al.  The analysis of covariance and alternatives , 1980 .

[21]  Richard K. Burdick,et al.  Analysis of Messy Data Volume III: Analysis of Covariance , 2002, Technometrics.

[22]  L. Low,et al.  The Analysis of Covariance and Alternatives , 1983 .

[23]  Michael G. Akritas,et al.  Nonparametric models and methods for ANCOVA with dependent data , 2004 .

[24]  Departament de Ciències Ambientals On the misuse of residuals in ecology : testing regression residuals vs . the analysis of covariance , 2001 .

[25]  Michael G. Akritas,et al.  Nonparametric models and methods for nonlinear analysis of covariance , 2000 .

[26]  Scott E. Maxwell,et al.  Another look at ANCOVA versus blocking. , 1984 .

[27]  G. A. Miller,et al.  Misunderstanding analysis of covariance. , 2001, Journal of abnormal psychology.

[28]  David Raubenheimer,et al.  Analysis of covariance: an alternative to nutritional indices , 1992 .