Analysis of Covariance

Part of the Springer Texts in Statistics book series (STS)


This chapter examines the analysis of partition models, also known as analysis of covariance, a method traditionally used for improving the analysis of designed experiments. Sections 1 and 2 present the theory of estimation and testing for general partitioned models. Sections 3 and 4 present nontraditional applications of the theory. Section 3 applies the partitioned model results to the problem of fixing up balanced ANOVA problems that have lost their balance due to the existence of some missing data. Although applying analysis of covariance to missing data problems is not a traditional experimental design application, it is an application that was used for quite some time until computational improvements made it largely unnecessary. Section 4 uses the analysis of covariance results to derive the analysis for balanced incomplete block designs. Section 5 presents Milliken and Graybill’s (1970) test of a linear model versus a nonlinear alternative. I personally find the techniques of Sections 1 and 2 to be some of the most valuable tools available for deriving results in linear model theory.


  1. Christensen, R. (1996). Analysis of variance, design, and regression: Applied statistical methods. London: Chapman and Hall.Google Scholar
  2. Christensen, R. (1997). Log-linear models and logistic regression (2nd ed.). New York: Springer.Google Scholar
  3. Christensen, R. (2015). Analysis of variance, design, and regression: Linear modeling for unbalanced data (2nd ed.). Boca Raton: Chapman and Hall/CRC Press.Google Scholar
  4. Christensen, R., & Utts, J. (1992). Testing for nonadditivity in log-linear and logit models. Journal of Statistical Planning and Inference, 33, 333–343.MathSciNetCrossRefGoogle Scholar
  5. Cox, D. R. (1958). Planning of experiments. New York: Wiley.Google Scholar
  6. LaMotte, L. R. (2014). The Gram-Schmidt construction as a basis for linear models. The American Statistician, 68, 52–55.MathSciNetCrossRefGoogle Scholar
  7. Mandel, J. (1961). Nonadditivity in two-way analysis of variance. Journal of the American Statistical Association, 56, 878–888.MathSciNetCrossRefGoogle Scholar
  8. Mandel, J. (1971). A new analysis of variance model for nonadditive data. Technometrics, 13, 1–18.CrossRefGoogle Scholar
  9. Milliken, G. A., & Graybill, F. A. (1970). Extensions of the general linear hypothesis model. Journal of the American Statistical Association, 65, 797–807.CrossRefGoogle Scholar
  10. Rao, C. R. (1973). Linear statistical inference and its applications (2nd ed.). New York: Wiley.Google Scholar
  11. Snedecor, G. W., & Cochran, W. G. (1980). Statistical methods (7th ed.). Ames: Iowa State University Press.Google Scholar
  12. St. Laurent, R. T. (1990). The equivalence of the Milliken-Graybill procedure and the score test. The American Statistician, 44, 36–37.Google Scholar
  13. Sulzberger, P. H. (1953). The effects of temperature on the strength of wood, plywood and glued joints. Department of supply, Report ACA-46, Aeronautical Research Consultative Committee, Australia.Google Scholar
  14. Tukey, J. W. (1949). One degree of freedom for nonadditivity. Biometrics, 5, 232–242.CrossRefGoogle Scholar
  15. Williams, E. J. (1959). Regression analysis. New York: Wiley.Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of Mathematics and StatisticsUniversity of New MexicoAlbuquerqueUSA

Personalised recommendations