PMEANS Subcommand (MANOVA: Univariate command)
PMEANS
(predicted
means) displays a table of the predicted cell means of the dependent
variable, adjusted for the effect of covariates in the cell and unadjusted
for covariates. For comparison, PMEANS
also displays the observed cell means.
- Output from
PMEANS
can be computationally expensive. -
PMEANS
without any additional specifications displays a table showing, for each cell, the observed mean of the dependent variable, the predicted mean adjusted for the effect of covariates in that cell (ADJ. MEAN), the predicted mean unadjusted for covariates (EST. MEAN), and the raw and standardized residuals from the estimated means. - Cells are numbered in output from
PMEANS
so that the levels vary most rapidly on the factor that is named last in theMANOVA
variables specification. A table showing the levels of each factor corresponding to each cell number is displayed at the beginning of theMANOVA
output. - Predicted means are suppressed for any design in
which the
MUPLUS
keyword appears. - Covariates are not predicted.
- In designs with covariates and multiple error terms,
use the
ERROR
subcommand to designate which error term’s regression coefficients are to be used in calculating the standardized residuals.
For univariate analysis, the following keywords
are available on the PMEANS
subcommand:
TABLES. Additional
tables showing adjusted predicted means for specified factors or combinations
of factors. Enter the names of factors or combinations
of factors in parentheses after this keyword. For each factor or combination,
MANOVA
displays the predicted
means (adjusted for covariates) collapsed over all other factors.
PLOT. A plot of the predicted means for each cell.
Example
MANOVA DEP BY A(1,4) B(1,3)
/PMEANS TABLES(A, B, A BY B)
/DESIGN = A, B.
-
PMEANS
displays the default table of observed and predicted means for DEP and raw and standardized residuals in each of the 12 cells in the model. - The
TABLES
specification onPMEANS
displays tables of predicted means for A (collapsing over B), for B (collapsing over A), and for all combinations of A and B. - Because A and B are the only factors in the model, the means
for A by B in the
TABLES
specification come from every cell in the model. The means are identical to the adjusted predicted means in the defaultPMEANS
table, which always includes all non-empty cells. - Predicted means for A by B can be requested in the
TABLES
specification, even though the A by B effect is not in the design.