PARTITION Subcommand (MANOVA: Univariate command)

PARTITION subdivides the degrees of freedom that are associated with a factor. This process permits you to test the significance of the effect of a specific contrast or group of contrasts of the factor instead of the overall effect of all contrasts of the factor. The default is a single degree of freedom for each partition.

  • Specify the factor name in parentheses following the PARTITION subcommand.
  • Specify an integer list in parentheses after the optional equals sign to indicate the degrees of freedom for each partition.
  • Each value in the partition list must be a positive integer, and the sum of the values cannot exceed the degrees of freedom for the factor.
  • The degrees of freedom that are available for a factor are one less than the number of levels of the factor.
  • The meaning of each degree of freedom depends on the contrast type for the factor. For example, with deviation contrasts (the default for between-subjects factors), each degree of freedom represents the deviation of the dependent variable in one level of the factor from its grand mean over all levels. With polynomial contrasts, the degrees of freedom represent the linear effect, the quadratic effect, and so on.
  • If your list does not account for all the degrees of freedom, MANOVA adds one final partition containing the remaining degrees of freedom.
  • You can use a repetition factor of the form n* to specify a series of partitions with the same number of degrees of freedom.
  • To specify a model that tests only the effect of a specific partition of a factor in your design, include the number of the partition in parentheses on the DESIGN subcommand (see the example below).
  • If you want the default single degree-of-freedom partition, you can omit the PARTITION subcommand and simply enter the appropriate term on the DESIGN subcommand.

Example

MANOVA OUTCOME BY TREATMNT(1,12)
  /PARTITION(TREATMNT) = (3*2,4) 
  /DESIGN TREATMNT(2).
  • The factor TREATMNT has 12 categories (hence, 11 degrees of freedom).
  • PARTITION divides the effect of TREATMNT into four partitions, containing, respectively, 2, 2, 2, and 4 degrees of freedom. A fifth partition is formed to contain the remaining 1 degree of freedom.
  • DESIGN specifies a model in which only the second partition of TREATMNT is tested. This partition contains the third and fourth degrees of freedom.
  • Because the default contrast type for between-subjects factors is DEVIATION, this second partition represents the deviation of the third and fourth levels of TREATMNT from the grand mean.