Output from probe2.sas

Source
0 Graphs

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 1

               SUBJNO     GROUP     P1    P2    P3    P4    P5
                  1      1Lo_STM    20    21    42    32    32
                  2      1Lo_STM    67    29    56    39    41
                  3      1Lo_STM    37    25    28    31    34
                  4      1Lo_STM    42    38    36    19    35
                  5      1Lo_STM    57    32    21    30    29
                  6      1Lo_STM    39    38    54    31    28
                  7      1Lo_STM    43    20    46    42    31
                  8      1Lo_STM    35    34    43    35    42
                  9      1Lo_STM    41    23    51    27    30
                 10      1Lo_STM    39    24    35    26    32
                 11      2Hi_STM    47    25    36    21    27
                 12      2Hi_STM    53    32    48    46    54
                 13      2Hi_STM    38    33    42    48    49
                 14      2Hi_STM    60    41    67    53    50
                 15      2Hi_STM    37    35    45    34    46
                 16      2Hi_STM    59    37    52    36    52
                 17      2Hi_STM    67    33    61    31    50
                 18      2Hi_STM    43    27    36    33    32
                 19      2Hi_STM    64    53    62    40    43
                 20      2Hi_STM    41    34    47    37    46

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 2 Two Sample Profile Analysis

                       General Linear Models Procedure
                           Class Level Information

                      Class    Levels    Values
                      GROUP         2    1Lo_STM 2Hi_STM


                   Number of observations in data set = 20



Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 3 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance
                     Repeated Measures Level Information

      Dependent Variable         P1       P2       P3       P4       P5

       Level of POSITION          1        2        3        4        5


Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 4 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance

  Partial Correlation Coefficients from the Error SS&CP Matrix / Prob > |r|

         DF = 18           P1        P2        P3        P4        P5
         P1          1.000000  0.381718  0.399535  0.149753  0.254986
                       0.0001    0.1068    0.0901    0.5406    0.2921
         P2          0.381718  1.000000  0.391231  0.146756  0.322161
                       0.1068    0.0001    0.0977    0.5488    0.1786
         P3          0.399535  0.391231  1.000000  0.428327  0.426038
                       0.0901    0.0977    0.0001    0.0673    0.0689
         P4          0.149753  0.146756  0.428327  1.000000  0.560026
                       0.5406    0.5488    0.0673    0.0001    0.0126
         P5          0.254986  0.322161  0.426038  0.560026  1.000000
                       0.2921    0.1786    0.0689    0.0126    0.0001



                            E = Error SS&CP Matrix

        POSITI.N represents the nth successive difference in POSITION
                  POSITI.1         POSITI.2         POSITI.3         POSITI.4

 POSITI.1      2313.300000      -740.800000       274.700000        16.200000
 POSITI.2      -740.800000      2026.000000     -1103.800000      -230.600000
 POSITI.3       274.700000     -1103.800000      1996.100000      -505.000000
 POSITI.4        16.200000      -230.600000      -505.000000       917.600000


Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 5 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance

         Partial Correlation Coefficients from the Error SS&CP Matrix
    of the Variables Defined by the Specified Transformation / Prob > |r|
              DF = 18     POSITI.1  POSITI.2  POSITI.3  POSITI.4

              POSITI.1    1.000000 -0.342188  0.127836  0.011119
                            0.0001    0.1516    0.6020    0.9640
              POSITI.2   -0.342188  1.000000 -0.548883 -0.169127
                            0.1516    0.0001    0.0149    0.4888
              POSITI.3    0.127836 -0.548883  1.000000 -0.373141
                            0.6020    0.0149    0.0001    0.1156
              POSITI.4    0.011119 -0.169127 -0.373141  1.000000
                            0.9640    0.4888    0.1156    0.0001



              Test for Sphericity: Mauchly's Criterion = 0.31758
  Chisquare Approximation = 18.830336 with 9 df   Prob > Chisquare = 0.0267

                      Applied to Orthogonal Components:
             Test for Sphericity: Mauchly's Criterion = 0.5692268
   Chisquare Approximation = 9.250404 with 9 df   Prob > Chisquare = 0.4145

                    H = Type III SS&CP Matrix for POSITION
        POSITI.N represents the nth successive difference in POSITION
                  POSITI.1         POSITI.2         POSITI.3         POSITI.4

 POSITI.1          4351.25          -4041.5          3200.75            -1357
 POSITI.2          -4041.5           3753.8          -2972.9           1260.4
 POSITI.3          3200.75          -2972.9          2354.45           -998.2
 POSITI.4            -1357           1260.4           -998.2            423.2


               Manova Test Criteria and Exact F Statistics for
                     the Hypothesis of no POSITION Effect
       H = Type III SS&CP Matrix for POSITION   E = Error SS&CP Matrix

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 6 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance

                             S=1    M=1    N=6.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.21986464    13.3059         4        15  0.0001
 Pillai's Trace             0.78013536    13.3059         4        15  0.0001
 Hotelling-Lawley Trace     3.54825288    13.3059         4        15  0.0001
 Roy's Greatest Root        3.54825288    13.3059         4        15  0.0001
                 H = Type III SS&CP Matrix for POSITION*GROUP
        POSITI.N represents the nth successive difference in POSITION
                  POSITI.1         POSITI.2         POSITI.3         POSITI.4

 POSITI.1            26.45            -20.7            19.55            -55.2
 POSITI.2            -20.7             16.2            -15.3             43.2
 POSITI.3            19.55            -15.3            14.45            -40.8
 POSITI.4            -55.2             43.2            -40.8            115.2


               Manova Test Criteria and Exact F Statistics for
                  the Hypothesis of no POSITION*GROUP Effect
    H = Type III SS&CP Matrix for POSITION*GROUP   E = Error SS&CP Matrix

                             S=1    M=1    N=6.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.83905852     0.7193         4        15  0.5919
 Pillai's Trace             0.16094148     0.7193         4        15  0.5919
 Hotelling-Lawley Trace     0.19181198     0.7193         4        15  0.5919
 Roy's Greatest Root        0.19181198     0.7193         4        15  0.5919

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 7 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance
               Tests of Hypotheses for Between Subjects Effects

Source                  DF     Type III SS    Mean Square   F Value     Pr > F
GROUP                    1      1772.41000     1772.41000      8.90     0.0080

Error                   18      3583.14000      199.06333



Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 8 Two Sample Profile Analysis

                       General Linear Models Procedure
                    Repeated Measures Analysis of Variance
          Univariate Tests of Hypotheses for Within Subject Effects

Source: POSITION
                                                                 Adj  Pr > F
     DF      Type III SS      Mean Square   F Value   Pr > F    G - G    H - F
      4    3371.30000000     842.82500000     14.48   0.0001   0.0001   0.0001
Source: POSITION*GROUP
                                                                 Adj  Pr > F
     DF      Type III SS      Mean Square   F Value   Pr > F    G - G    H - F
      4      79.94000000      19.98500000      0.34   0.8479   0.8068   0.8479
Source: Error(POSITION)

     DF      Type III SS      Mean Square
     72    4191.96000000      58.22166667

                     Greenhouse-Geisser Epsilon = 0.8009
                            Huynh-Feldt Epsilon = 1.0487


                            E = Error SS&CP Matrix
               P1             P2             P3             P4             P5

P1         2546.9            597          946.6          257.9          385.9
P2            597          960.4          569.2          155.2          299.4
P3          946.6          569.2           2204          686.2          599.8
P4          257.9          155.2          686.2         1164.5          573.1
P5          385.9          299.4          599.8          573.1          899.3

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 9 Two Sample Profile Analysis

                       General Linear Models Procedure
                      Multivariate Analysis of Variance

  Partial Correlation Coefficients from the Error SS&CP Matrix / Prob > |r|
         DF = 18           P1        P2        P3        P4        P5

         P1          1.000000  0.381718  0.399535  0.149753  0.254986
                       0.0001    0.1068    0.0901    0.5406    0.2921
         P2          0.381718  1.000000  0.391231  0.146756  0.322161
                       0.1068    0.0001    0.0977    0.5488    0.1786
         P3          0.399535  0.391231  1.000000  0.428327  0.426038
                       0.0901    0.0977    0.0001    0.0673    0.0689
         P4          0.149753  0.146756  0.428327  1.000000  0.560026
                       0.5406    0.5488    0.0673    0.0001    0.0126
         P5          0.254986  0.322161  0.426038  0.560026  1.000000
                       0.2921    0.1786    0.0689    0.0126    0.0001



Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 10 Two Sample Profile Analysis

                       General Linear Models Procedure
                      Multivariate Analysis of Variance

                     H = Type III SS&CP Matrix for GROUP
               P1             P2             P3             P4             P5

P1         396.05          293.7          373.8         298.15         511.75
P2          293.7          217.8          277.2          221.1          379.5
P3          373.8          277.2          352.8          281.4            483
P4         298.15          221.1          281.4         224.45         385.25
P5         511.75          379.5            483         385.25         661.25


Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 11 Two Sample Profile Analysis

                       General Linear Models Procedure
                      Multivariate Analysis of Variance
                              Canonical Analysis

         H = Type III SS&CP Matrix for GROUP   E = Error SS&CP Matrix

                                Adjusted       Approx       Squared
                 Canonical      Canonical     Standard     Canonical
                Correlation    Correlation     Error      Correlation
           1      0.666276       0.588174     0.127573      0.443924
                                Eigenvalues of INV(E)*H
                                  = CanRsq/(1-CanRsq)

                 Eigenvalue    Difference    Proportion    Cumulative
            1       0.7983         .           1.0000        1.0000
                     Test of H0: The canonical correlations in the
                       current row and all that follow are zero

               Likelihood
                  Ratio      Approx F      Num DF      Den DF    Pr > F
          1    0.55607569      2.2353           5          14    0.1083

                       NOTE: The F statistic is exact.


                          Total Canonical Structure
                                                CAN1

                                P1            0.5506
                                P2            0.6453
                                P3            0.5575
                                P4            0.6033
                                P5            0.9770


                         Between Canonical Structure
                                                CAN1

                                P1            1.0000
                                P2            1.0000
                                P3            1.0000
                                P4            1.0000
                                P5            1.0000

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 12 Two Sample Profile Analysis

                       General Linear Models Procedure
                      Multivariate Analysis of Variance
                              Canonical Analysis

                          Within Canonical Structure
                                                CAN1

                                P1            0.4413
                                P2            0.5330
                                P3            0.4478
                                P4            0.4914
                                P5            0.9597

                     Standardized Canonical Coefficients
                                                CAN1

                                P1            0.1710
                                P2            0.2293
                                P3           -0.0706
                                P4           -0.0423
                                P5            1.1546

                          Raw Canonical Coefficients
                                                CAN1

                                P1      0.0137368147
                                P2      0.0291163481
                                P3      -0.006082556
                                P4      -0.004946857
                                P5      0.1273983836
               Manova Test Criteria and Exact F Statistics for
                  the Hypothesis of no Overall GROUP Effect
         H = Type III SS&CP Matrix for GROUP   E = Error SS&CP Matrix

                             S=1    M=1.5    N=6

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.55607569     2.2353         5        14  0.1083
 Pillai's Trace             0.44392431     2.2353         5        14  0.1083
 Hotelling-Lawley Trace     0.79831634     2.2353         5        14  0.1083
 Roy's Greatest Root        0.79831634     2.2353         5        14  0.1083

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 13 Output from CANDISC Procedure

                       Canonical Discriminant Analysis
                  20 Observations        19 DF Total
                   5 Variables           18 DF Within Classes
                   2 Classes              1 DF Between Classes

                           Class Level Information

              GROUP       Frequency        Weight     Proportion
              1Lo_STM            10       10.0000       0.500000
              2Hi_STM            10       10.0000       0.500000

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 14 Output from CANDISC Procedure

                       Canonical Discriminant Analysis

                Multivariate Statistics and Exact F Statistics

                             S=1    M=1.5    N=6

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.55607569     2.2353         5        14  0.1083
 Pillai's Trace             0.44392431     2.2353         5        14  0.1083
 Hotelling-Lawley Trace     0.79831634     2.2353         5        14  0.1083
 Roy's Greatest Root        0.79831634     2.2353         5        14  0.1083

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 15 Output from CANDISC Procedure

                       Canonical Discriminant Analysis

                                Adjusted       Approx       Squared
                 Canonical      Canonical     Standard     Canonical
                Correlation    Correlation     Error      Correlation
           1      0.666276       0.588174     0.127573      0.443924
                                Eigenvalues of INV(E)*H
                                  = CanRsq/(1-CanRsq)

                 Eigenvalue    Difference    Proportion    Cumulative
            1       0.7983         .           1.0000        1.0000
                     Test of H0: The canonical correlations in the
                       current row and all that follow are zero

               Likelihood
                  Ratio      Approx F      Num DF      Den DF    Pr > F
          1    0.55607569      2.2353           5          14    0.1083

                       NOTE: The F statistic is exact.

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 16 Scores on discriminant function

      OBS    SUBJNO    P1    P2    P3    P4    P5     GROUP       CAN1
        1       1      20    21    42    32    32    1Lo_STM    -1.55249
        2       2      67    29    56    39    41    1Lo_STM     0.35288
        3       3      37    25    28    31    34    1Lo_STM    -0.85760
        4       4      42    38    36    19    35    1Lo_STM    -0.27230
        5       5      57    32    21    30    29    1Lo_STM    -0.96851
        6       6      39    38    54    31    28    1Lo_STM    -1.37415
        7       7      43    20    46    42    31    1Lo_STM    -1.46685
        8       8      35    34    43    35    42    1Lo_STM     0.28514
        9       9      41    23    51    27    30    1Lo_STM    -1.49059
       10      10      39    24    35    26    32    1Lo_STM    -1.13188
       11      11      47    25    36    21    27    2Hi_STM    -1.61121
       12      12      53    32    48    46    54    2Hi_STM     1.91812
       13      13      38    33    42    48    49    2Hi_STM     1.13079
       14      14      60    41    67    53    50    2Hi_STM     1.61654
       15      15      37    35    45    34    46    2Hi_STM     0.84410
       16      16      59    37    52    36    52    2Hi_STM     1.91647
       17      17      67    33    61    31    50    2Hi_STM     1.62509
       18      18      43    27    36    33    32    2Hi_STM    -1.03029
       19      19      64    53    62    40    43    2Hi_STM     1.22381
       20      20      41    34    47    37    46    2Hi_STM     0.84293

Profile Analysis Example: PROBE2 (Timm, Ex. 3.16) 17 Scores on discriminant function Univariate t-test on discriminant scores

                               TTEST PROCEDURE

Variable: CAN1

GROUP         N                 Mean              Std Dev            Std Error
------------------------------------------------------------------------------
1Lo_STM      10          -0.84763477           0.72432100           0.22905041
2Hi_STM      10           0.84763477           1.21464361           0.38410403
Variances        T       DF    Prob>|T|
---------------------------------------
Unequal    -3.7907     14.7      0.0018
Equal      -3.7907     18.0      0.0013

For H0: Variances are equal, F' = 2.81    DF = (9,9)    Prob>F' = 0.1395