Figure 1.3b

ESTIMATES FOR Y

CONSTANT
PARAMETER COEFF. STD. ERR. T-VALUE SIG. OF T LOWER .95 CL UPPER .95 CL
CAT 1 7 .0000000000 70053 9. 99245 000 5.47368 8.52632
PARAMETER COEFF. STD. ERR. T-VALVE SIG. OF T LOWER .95 CL UPPER .95 CL
2 -|.0000000000 70053 -l.42749 179 ,-2.52632 .52632
DRUG
PARAMETER COEFF. STD. ERR. T-VALVE SIG. OF T LOWER .95 CL UPPER .95 CL
3 0.0 - 99070 0.0 1.000 -2.15854 2.15854
4 -2 .0000000000 » 99070 -2.01878 -066 -4. 15854 -15854
CAT BY DRUG
PARAMETER COEFF. STD. ERR. T-VALUE SIG. OF T LOWER .95 CL UPPER .95 CL
5 -2.0000000000 - 99070 -2.01878 066 —4.15854 15854
6 4 .0000000000 - 990TO 4.03756 002 1.84146 6. 15854

1.4 Use of the PRINT Subcommand

Additional printed output can be obtained by using the PRINT subcommand. For instance, tests of
homogeneity of within-cells variance are produced by specifying

MANOVA Y BY CAT(1,2) DRUG(l,3)/
PRINT<HOMOGENEITY ( BARTLETT, COCHRAN) /

The output (Figure 1.4a) includes Bartlett's test and Cochran's test. The significance (P) of both
tests is also given.

Figure 1.4a

UNIVARIATE HOMOGENEITY OF VARIANCE TESTS

VARIABLE .. Y
COCHRANS C(2,6) - -30189, P -  .829 (APPROX.)
BARTLETT-BOX F(5,185) < »38601, P - .858

The cell statistics, including the mean, standard deviation, number of observations, and the
95% confidence intervals for the population means can be obtained using

MANOVA Y BY CAT(1,2) DRUG(l1,3)/
PRINT<CELLINFO ( MEANS)) /

The output from the above PRINT subcommand is given in Figure 1.4b. -

Figure 1.4b

CELL MEANS AND STANDARD DEVIATIONS
VARIABLE .. Y

FACTOR CODE MEAN STD. DEV. N 95 PERCENT CONF. INTERVAL

CAT l
DRUG l 4.00000 4.00000 3 -5.93666 13.93666
DRUG 2 8.00000 2.00000 3 3.03167 12.96833
DRUG 3 6.00000 2.00000 3 1.03167 10.96833

CAT 2
DRUG 1 10.00000 4.00000 3 06334 ' 19.93666
DRUG 2 2.00000 2.00000 3 -2. 96833 6 .96833
DRUG 3 12.00000 3.00000 3 4.54751 19.45249
FOR ENTIRE SAMPLE 7.00000 4.31141 18 4 .85599 9.14401

MANOVA 5