A note on the analysis of repeated measurements of the same subjects

https://doi.org/10.1016/0021-9681(62)90116-9Get rights and content

Abstract

When repeated measurements have been made on the same subject, care must be taken in choosing the indices which are to be used to summarize the measurements, or erroneous conclusions may be reached. For example, when two measurements have been made, it is natural to think of analysing the first, and the difference between second and first. These indices are not independent, however, and to treat them as independent is to introduce spurious correlations into the results. The best solution is to use as indices the functions of the repeated measurements given by orthogonal polynomials. In the case of two measurements these are the mean of the two and their difference.

References (17)

  • O. Turpeinen et al.

    Effect on serum-cholesterol level of replacement of dietary milk fat by soybean oil

    Lancet

    (1960)
  • R.A. Fisher et al.

    Statistical Tables for Biological, Agricultural and Medical Research

    (1957)
  • L.H.C. Tippett

    Tracts for Computers, XV

    (1927)
  • M.G. Kendall et al.

    Tracts for Computers, XXIV

    (1939)
  • A.B. Hill

    Principles of Medical Statistics

    (1961)
  • I.T.T. Higgins et al.

    Ventilatory capacity in miners. A five-year follow-up study

    Brit. J. industr. Med.

    (1962)
  • S.C. Pearce

    A method of studying manner of growth

    Biometrics

    (1960)
  • R.A. Fisher

    Studies in crop variation—I. An examination of the yield of dressed grain from Broadbalk

    J. agric. Sci.

    (1920)
There are more references available in the full text version of this article.

Cited by (259)

View all citing articles on Scopus
View full text