Output from m1way.sas

Source
0 Graphs

Dog Food: One Way Completely Randomized Manova Design 1

                   OBS    FORMULA    DOG    START    AMOUNT
                     1     OLD        1       0        100
                     2     OLD        2       1         97
                     3     OLD        3       1         88
                     4     OLD        4       0         92
                     5     NEW        1       0         95
                     6     NEW        2       1         85
                     7     NEW        3       2         82
                     8     NEW        4       3         89
                     9     MAJOR      1       1         77
                    10     MAJOR      2       5         84
                    11     MAJOR      3       3         78
                    12     MAJOR      4       1         89
                    13     ALPS       1       2         72
                    14     ALPS       2       3         82
                    15     ALPS       3       4         85
                    16     ALPS       4       0         74

Dog Food: One Way Completely Randomized Manova Design 2

             Plot of AMOUNT*START.  Symbol is value of FORMULA.

      |
      |
  100 +O
   99 +
   98 +
   97 +              O
   96 +
   95 +N
   94 +
   93 +
A  92 +O
m  91 +
o  90 +
u  89 +              M                           N
n  88 +              O
t  87 +
   86 +
e  85 +              N                                         A
a  84 +                                                                      M
t  83 +
e  82 +                            N             A
n  81 +
   80 +
   79 +
   78 +                                          M
   77 +              M
   76 +
   75 +
   74 +A
   73 +
   72 +                            A
      |
      -+-------------+-------------+-------------+-------------+-------------+
       0             1             2             3             4             5

                                Time to start eating


Dog Food: One Way Completely Randomized Manova Design 3 MANOVA for equality of means

                       General Linear Models Procedure
                           Class Level Information

                    Class    Levels    Values
                    FORMULA       4    OLD NEW MAJOR ALPS


                   Number of observations in data set = 16


Dog Food: One Way Completely Randomized Manova Design 4 MANOVA for equality of means

                       General Linear Models Procedure

Dependent Variable: START   Time to start eating
                                    Sum of           Mean
Source                  DF         Squares         Square   F Value     Pr > F
Model                    3      9.68750000     3.22916667      1.50     0.2634

Error                   12     25.75000000     2.14583333

Corrected Total         15     35.43750000

                  R-Square            C.V.       Root MSE           START Mean
                  0.273369        86.80689       1.464866             1.687500

Source                  DF     Type III SS    Mean Square   F Value     Pr > F
FORMULA                  3      9.68750000     3.22916667      1.50     0.2634

Dog Food: One Way Completely Randomized Manova Design 5 MANOVA for equality of means

                       General Linear Models Procedure

Dependent Variable: AMOUNT   Amount eaten
                                    Sum of           Mean
Source                  DF         Squares         Square   F Value     Pr > F
Model                    3     585.6875000    195.2291667      6.00     0.0097

Error                   12     390.2500000     32.5208333

Corrected Total         15     975.9375000

                  R-Square            C.V.       Root MSE          AMOUNT Mean
                  0.600128        6.664957       5.702704             85.56250

Source                  DF     Type III SS    Mean Square   F Value     Pr > F
FORMULA                  3     585.6875000    195.2291667      6.00     0.0097



Dog Food: One Way Completely Randomized Manova Design 6 MANOVA for equality of means

                       General Linear Models Procedure

Level of       ------------START------------     ------------AMOUNT-----------
FORMULA    N       Mean              SD              Mean              SD
OLD        4     0.50000000       0.57735027       94.2500000       5.31507291
NEW        4     1.50000000       1.29099445       87.7500000       5.61990510
MAJOR      4     2.50000000       1.91485422       82.0000000       5.59761854
ALPS       4     2.25000000       1.70782513       78.2500000       6.23832242

Dog Food: One Way Completely Randomized Manova Design 7 MANOVA for equality of means

                       General Linear Models Procedure
                      Multivariate Analysis of Variance

          Characteristic Roots and Vectors of: E Inverse * H, where
        H = Type III SS&CP Matrix for FORMULA   E = Error SS&CP Matrix

        Characteristic   Percent        Characteristic Vector  V'EV=1
             Root
                                                 START         AMOUNT
            2.03961854     98.47           -0.10279413     0.04639418
            0.03174562      1.53            0.16973304     0.02111246


                Manova Test Criteria and F Approximations for
                 the Hypothesis of no Overall FORMULA Effect
        H = Type III SS&CP Matrix for FORMULA   E = Error SS&CP Matrix

                             S=2    M=0    N=4.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.31886605     2.8267         6        22  0.0341
 Pillai's Trace             0.70178019     2.1623         6        24  0.0829
 Hotelling-Lawley Trace     2.07136416     3.4523         6        20  0.0166
 Roy's Greatest Root        2.03961854     8.1585         3        12  0.0031

         NOTE: F Statistic for Roy's Greatest Root is an upper bound.
                NOTE: F Statistic for Wilks' Lambda is exact.
          Characteristic Roots and Vectors of: E Inverse * H, where
  H = Contrast SS&CP Matrix for Equality of Groups   E = Error SS&CP Matrix

        Characteristic   Percent        Characteristic Vector  V'EV=1
             Root
                                                 START         AMOUNT
            2.03961854     98.47           -0.10279413     0.04639418
            0.03174562      1.53            0.16973304     0.02111246



Dog Food: One Way Completely Randomized Manova Design 8 MANOVA for equality of means

                       General Linear Models Procedure
                      Multivariate Analysis of Variance

                Manova Test Criteria and F Approximations for
            the Hypothesis of no Overall Equality of Groups Effect
  H = Contrast SS&CP Matrix for Equality of Groups   E = Error SS&CP Matrix

                             S=2    M=0    N=4.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.31886605     2.8267         6        22  0.0341
 Pillai's Trace             0.70178019     2.1623         6        24  0.0829
 Hotelling-Lawley Trace     2.07136416     3.4523         6        20  0.0166
 Roy's Greatest Root        2.03961854     8.1585         3        12  0.0031

         NOTE: F Statistic for Roy's Greatest Root is an upper bound.
                NOTE: F Statistic for Wilks' Lambda is exact.
          Characteristic Roots and Vectors of: E Inverse * H, where
    H = Contrast SS&CP Matrix for Ours vs. Theirs   E = Error SS&CP Matrix

        Characteristic   Percent        Characteristic Vector  V'EV=1
             Root
                                                 START         AMOUNT
            1.66869411    100.00           -0.10378699     0.04626967
            0.00000000      0.00            0.16912776     0.02138397


               Manova Test Criteria and Exact F Statistics for
             the Hypothesis of no Overall Ours vs. Theirs Effect
    H = Contrast SS&CP Matrix for Ours vs. Theirs   E = Error SS&CP Matrix

                             S=1    M=0    N=4.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.37471511     9.1778         2        11  0.0045
 Pillai's Trace             0.62528489     9.1778         2        11  0.0045
 Hotelling-Lawley Trace     1.66869411     9.1778         2        11  0.0045
 Roy's Greatest Root        1.66869411     9.1778         2        11  0.0045
          Characteristic Roots and Vectors of: E Inverse * H, where
       H = Contrast SS&CP Matrix for Old - New   E = Error SS&CP Matrix

        Characteristic   Percent        Characteristic Vector  V'EV=1
             Root
                                                 START         AMOUNT
            0.32912079    100.00           -0.11606288     0.04455347
            0.00000000      0.00            0.16095138     0.02476175

Dog Food: One Way Completely Randomized Manova Design 9 MANOVA for equality of means

                       General Linear Models Procedure
                      Multivariate Analysis of Variance

               Manova Test Criteria and Exact F Statistics for
                the Hypothesis of no Overall Old - New Effect
       H = Contrast SS&CP Matrix for Old - New   E = Error SS&CP Matrix

                             S=1    M=0    N=4.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.75237707     1.8102         2        11  0.2091
 Pillai's Trace             0.24762293     1.8102         2        11  0.2091
 Hotelling-Lawley Trace     0.32912079     1.8102         2        11  0.2091
 Roy's Greatest Root        0.32912079     1.8102         2        11  0.2091
          Characteristic Roots and Vectors of: E Inverse * H, where
    H = Contrast SS&CP Matrix for Major vs. Alps   E = Error SS&CP Matrix

        Characteristic   Percent        Characteristic Vector  V'EV=1
             Root
                                                 START         AMOUNT
            0.07354926    100.00            0.02814930     0.04926128
            0.00000000      0.00            0.19642697    -0.01309513


               Manova Test Criteria and Exact F Statistics for
              the Hypothesis of no Overall Major vs. Alps Effect
    H = Contrast SS&CP Matrix for Major vs. Alps   E = Error SS&CP Matrix

                             S=1    M=0    N=4.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.93148963     0.4045         2        11  0.6768
 Pillai's Trace             0.06851037     0.4045         2        11  0.6768
 Hotelling-Lawley Trace     0.07354926     0.4045         2        11  0.6768
 Roy's Greatest Root        0.07354926     0.4045         2        11  0.6768

Dog Food: One Way Completely Randomized Manova Design 10 MANOVA for equality of means

                       General Linear Models Procedure

Dependent Variable: START   Time to start eating

Contrast                DF     Contrast SS    Mean Square   F Value     Pr > F
Equality of Groups       3      9.68750000     3.22916667      1.50     0.2634
Ours vs. Theirs          1      7.56250000     7.56250000      3.52     0.0850
Old - New                1      2.00000000     2.00000000      0.93     0.3534
Major vs. Alps           1      0.12500000     0.12500000      0.06     0.8134

Dog Food: One Way Completely Randomized Manova Design 11 MANOVA for equality of means

                       General Linear Models Procedure

Dependent Variable: AMOUNT   Amount eaten

Contrast                DF     Contrast SS    Mean Square   F Value     Pr > F
Equality of Groups       3     585.6875000    195.2291667      6.00     0.0097
Ours vs. Theirs          1     473.0625000    473.0625000     14.55     0.0025
Old - New                1      84.5000000     84.5000000      2.60     0.1329
Major vs. Alps           1      28.1250000     28.1250000      0.86     0.3707