This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 1
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGE08:01 Monday, April 19, 2004
Obs DRUG DISEASE ISBP MISS BSBP FSBP
1 1 1 42 0 142 184
2 1 1 44 0 136 180
3 1 1 36 0 118 154
4 1 1 13 0 114 127
5 1 1 19 0 107 126
6 1 1 22 0 116 138
7 1 2 33 0 125 158
8 1 2 26 0 115 141
9 1 2 33 0 132 165
10 1 2 21 0 115 136
11 1 3 31 0 119 150
12 1 3 -3 0 92 89
13 1 3 25 0 121 146
14 1 3 25 0 117 142
15 1 3 24 0 111 135
16 2 1 28 0 109 137
17 2 1 23 0 116 139
18 2 1 34 0 126 160
19 2 1 42 0 139 181
20 2 1 13 0 112 125
21 2 2 34 0 116 150
22 2 2 33 0 134 167
23 2 2 31 0 127 158
24 2 2 36 0 119 155
25 2 3 3 0 76 79
26 2 3 26 0 98 124
27 2 3 28 0 118 146
28 2 3 32 0 115 147
29 2 3 4 0 100 104
30 2 3 16 0 97 113
31 3 1 1 0 95 96
32 3 1 29 0 102 131
33 3 1 19 0 114 133
34 3 2 11 0 104 115
35 3 2 9 0 115 124
36 3 2 7 0 126 133
37 3 2 1 0 104 105
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 2
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGE08:01 Monday, April 19, 2004
Obs DRUG DISEASE ISBP MISS BSBP FSBP
38 3 2 -6 0 97 91
39 3 3 21 0 125 146
40 3 3 1 0 107 108
41 3 3 9 0 99 108
42 3 3 3 0 109 112
43 4 1 24 0 121 145
44 4 1 9 0 95 104
45 4 1 22 0 99 121
46 4 1 -2 0 104 102
47 4 1 15 0 107 122
48 4 2 27 0 113 140
49 4 2 12 0 117 129
50 4 2 12 0 82 94
51 4 2 -5 0 96 91
52 4 2 16 0 89 105
53 4 2 15 0 107 122
54 4 3 22 0 105 127
55 4 3 7 0 83 90
56 4 3 25 0 121 146
57 4 3 5 0 101 106
58 4 3 12 0 92 104
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 3
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM Procedure
Class LevelInformation
Class Levels Values
DRUG 4 1 2 3 4
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 4
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 3 9895.46897 3298.48966 6.98 0.0005
Error 54 25502.60000 472.27037
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.279548 16.79248 21.73178 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 9895.468966 3298.489655 6.98 0.0005
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept 115.5000000 B 5.43294562 21.26 <.0001
DRUG 1 29.2333333 B 7.81035143 3.74 0.0004
DRUG 2 23.5000000 B 7.81035143 3.01 0.0040
DRUG 3 1.3333333 B 8.29896152 0.16 0.8730
DRUG 4 0.0000000 B . . .
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 5
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 144.733333 5.611122 <.0001 1
2 139.000000 5.611122 <.0001 2
3 116.833333 6.273425 <.0001 3
4 115.500000 5.432946 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3 4
1 0.4731 0.0016 0.0004
2 0.4731 0.0110 0.0040
3 0.0016 0.0110 0.8730
4 0.0004 0.0040 0.8730
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 6
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEREGRESSION OF FOLLOWUP ON BASELINE 08:01 Monday, April 19, 2004
The GLM Procedure
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 7
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEREGRESSION OF FOLLOWUP ON BASELINE 08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 1 30786.10565 30786.10565 373.82 <.0001
Error 56 4611.96332 82.35649
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.869711 7.012427 9.075048 129.4138
Source DF Type I SS Mean Square F Value Pr > F
BSBP 1 30786.10565 30786.10565 373.82 <.0001
Source DF Type III SS Mean Square F Value Pr > F
BSBP 1 30786.10565 30786.10565 373.82 <.0001
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -52.35507204 9.47658426 -5.52 <.0001
BSBP 1.64445394 0.08505371 19.33 <.0001
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 8
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON Baseline B.P. 08:01 Monday, April 19, 2004
The GLM Procedure
Class LevelInformation
Class Levels Values
DRUG 4 1 2 3 4
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 9
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON Baseline B.P. 08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: BSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 3 2358.44770 786.14923 4.70 0.0055
Error 54 9025.98333 167.14784
Corrected Total 57 11384.43103
R-Square Coeff Var Root MSE BSBP Mean
0.207164 11.69641 12.92857 110.5345
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 2358.447701 786.149234 4.70 0.0055
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept 102.0000000 B 3.23214170 31.56 <.0001
DRUG 1 16.6666667 B 4.64649645 3.59 0.0007
DRUG 2 11.4666667 B 4.64649645 2.47 0.0168
DRUG 3 6.0833333 B 4.93717800 1.23 0.2232
DRUG 4 0.0000000 B . . .
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 10
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON Baseline B.P. 08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG BSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 118.666667 3.338142 <.0001 1
2 113.466667 3.338142 <.0001 2
3 108.083333 3.732156 <.0001 3
4 102.000000 3.232142 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: BSBP
i/j 1 2 3 4
1 0.2756 0.0392 0.0007
2 0.2756 0.2871 0.0168
3 0.0392 0.2871 0.2232
4 0.0007 0.0168 0.2232
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 11
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., ADJUSTED FOR BASELINE B.P.
08:01 Monday, April 19, 2004
The GLM Procedure
Class LevelInformation
Class Levels Values
DRUG 4 1 2 3 4
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 12
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., ADJUSTED FOR BASELINE B.P.
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 4 32112.35031 8028.08758 129.50 <.0001
Error 53 3285.71866 61.99469
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.907178 6.084105 7.873671 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 1326.24466 442.08155 7.13 0.0004
BSBP 1 22216.88134 22216.88134 358.37 <.0001
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -44.52743930 B 8.67952910 -5.13 <.0001
DRUG 1 3.08505894 B 3.14889757 0.98 0.3317
DRUG 2 5.50998722 B 2.98508703 1.85 0.0705
DRUG 3 -8.21078682 B 3.04878232 -2.69 0.0095
DRUG 4 0.00000000 B . . .
BSBP 1.56889646 0.08287623 18.93 <.0001
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 13
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG EFFECT ON FOLLOWUP B.P., ADJUSTED FOR BASELINE B.P.
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
Coefficients for DRUG Least Square Means
DRUG Level
Effect 1 2 3 4
Intercept 1 1 1 1
DRUG 1 1 0 0 0
DRUG 2 0 1 0 0
DRUG 3 0 0 1 0
DRUG 4 0 0 0 1
BSBP 110.534483 110.534483 110.534483 110.534483
DRUG FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 131.974779 2.141777 <.0001 1
2 134.399707 2.047445 <.0001 2
3 120.678933 2.281993 <.0001 3
4 128.889720 2.091638 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3 4
1 0.4080 0.0008 0.3317
2 0.4080 <.0001 0.0705
3 0.0008 <.0001 0.0095
4 0.3317 0.0705 0.0095
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 14
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM Procedure
Class Level Information
Class Levels Values
DRUG 4 1 2 3 4
DISEASE 3 1 2 3
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 15
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 11 15468.63563 1406.23960 3.25 0.0024
Error 46 19929.43333 433.24855
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.436991 16.08378 20.81462 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 10191.88576 3397.29525 7.84 0.0002
DISEASE 2 2181.69748 1090.84874 2.52 0.0917
DRUG*DISEASE 6 3018.68300 503.11383 1.16 0.3434
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept 114.6000000 B 9.30858261 12.31 <.0001
DRUG 1 17.8000000 B 13.16432377 1.35 0.1829
DRUG 2 4.2333333 B 12.60388043 0.34 0.7385
DRUG 3 3.9000000 B 13.96287391 0.28 0.7813
DRUG 4 0.0000000 B . . .
DISEASE 1 4.2000000 B 13.16432377 0.32 0.7511
DISEASE 2 -1.1000000 B 12.60388043 -0.09 0.9308
DISEASE 3 0.0000000 B . . .
DRUG*DISEASE 1 1 14.9000000 B 18.22518099 0.82 0.4178
DRUG*DISEASE 1 2 18.7000000 B 18.81009436 0.99 0.3254
DRUG*DISEASE 1 3 0.0000000 B . . .
DRUG*DISEASE 2 1 25.3666667 B 18.22518099 1.39 0.1707
DRUG*DISEASE 2 2 39.7666667 B 18.42221570 2.16 0.0361
DRUG*DISEASE 2 3 0.0000000 B . . .
DRUG*DISEASE 3 1 -2.7000000 B 20.64043947 -0.13 0.8965
DRUG*DISEASE 3 2 -3.8000000 B 18.81009436 -0.20 0.8408
DRUG*DISEASE 3 3 0.0000000 B . . .
DRUG*DISEASE 4 1 0.0000000 B . . .
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 16
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM ProcedureDependent Variable: FSBP
Parameter EstimateStandard
Error t Value Pr > |t|
DRUG*DISEASE 4 2 0.0000000 B . . .
DRUG*DISEASE 4 3 0.0000000 B . . .
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 17
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 144.633333 5.448445 <.0001 1
2 141.577778 5.448445 <.0001 2
3 117.366667 6.140739 <.0001 3
4 115.633333 5.222893 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3 4
1 0.6935 0.0018 0.0004
2 0.6935 0.0050 0.0013
3 0.0018 0.0050 0.8307
4 0.0004 0.0013 0.8307
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
DISEASE FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 134.675000 4.936621 <.0001 1
2 133.650000 4.844340 <.0001 2
3 121.083333 4.702524 <.0001 3
Least Squares Means for effect DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3
1 0.8828 0.0522
2 0.8828 0.0691
3 0.0522 0.0691
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 18
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., UNADJUSTED
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG DISEASE FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 1 151.500000 8.497534 <.0001 1
1 2 150.000000 10.407312 <.0001 2
1 3 132.400000 9.308583 <.0001 3
2 1 148.400000 9.308583 <.0001 4
2 2 157.500000 10.407312 <.0001 5
2 3 118.833333 8.497534 <.0001 6
3 1 120.000000 12.017328 <.0001 7
3 2 113.600000 9.308583 <.0001 8
3 3 118.500000 10.407312 <.0001 9
4 1 118.800000 9.308583 <.0001 10
4 2 113.500000 8.497534 <.0001 11
4 3 114.600000 9.308583 <.0001 12
Least Squares Means for effect DRUG*DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 23
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Class Level Information
Class Levels Values
DRUG 4 1 2 3 4
DISEASE 3 1 2 3
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 24
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 12 32632.94394 2719.41200 44.26 <.0001
Error 45 2765.12502 61.44722
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.921885 6.057181 7.838828 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 1004.31971 334.77324 5.45 0.0028
DISEASE 2 77.44095 38.72047 0.63 0.5371
DRUG*DISEASE 6 480.01760 80.00293 1.30 0.2759
BSBP 1 17164.30831 17164.30831 279.33 <.0001
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -44.08633346 B 10.12112729 -4.36 <.0001
DRUG 1 -0.53427757 B 5.07762456 -0.11 0.9167
DRUG 2 3.81185569 B 4.74671300 0.80 0.4262
DRUG 3 -11.27319523 B 5.33623888 -2.11 0.0402
DRUG 4 0.00000000 B . . .
DISEASE 1 -3.38659762 B 4.97844733 -0.68 0.4998
DISEASE 2 -1.52147765 B 4.74671300 -0.32 0.7500
DISEASE 3 0.00000000 B . . .
DRUG*DISEASE 1 1 6.41776239 B 6.88237665 0.93 0.3561
DRUG*DISEASE 1 2 3.71120124 B 7.14046097 0.52 0.6058
DRUG*DISEASE 1 3 0.00000000 B . . .
DRUG*DISEASE 2 1 1.76391853 B 7.00741651 0.25 0.8024
DRUG*DISEASE 2 2 3.30885035 B 7.27268983 0.45 0.6513
DRUG*DISEASE 2 3 0.00000000 B . . .
DRUG*DISEASE 3 1 14.89669169 B 7.84420858 1.90 0.0640
DRUG*DISEASE 3 2 -2.11408942 B 7.08463646 -0.30 0.7668
DRUG*DISEASE 3 3 0.00000000 B . . .
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 25
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureDependent Variable: FSBP
Parameter EstimateStandard
Error t Value Pr > |t|
DRUG*DISEASE 4 1 0.00000000 B . . .
DRUG*DISEASE 4 2 0.00000000 B . . .
DRUG*DISEASE 4 3 0.00000000 B . . .
BSBP 1.58054117 0.09456791 16.71 <.0001
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 26
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 131.823986 2.190358 <.0001 1
2 134.484721 2.095324 <.0001 2
3 121.969614 2.328955 <.0001 3
4 128.981942 2.122921 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3 4
1 0.3673 0.0042 0.3857
2 0.3673 0.0003 0.0821
3 0.0042 0.0003 0.0276
4 0.3857 0.0821 0.0276
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
DISEASE FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 131.002058 1.872085 <.0001 1
2 128.324076 1.852009 <.0001 2
3 128.619063 1.827475 <.0001 3
Least Squares Means for effect DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3
1 0.3097 0.3740
2 0.3097 0.9121
3 0.3740 0.9121
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 27
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON FOLLOWUP B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG DISEASE FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 1 133.114854 3.383973 <.0001 1
1 2 132.273413 4.060386 <.0001 2
1 3 130.083690 3.508369 <.0001 3
2 1 132.807144 3.627652 <.0001 4
2 2 136.217196 4.121088 <.0001 5
2 3 134.429823 3.333471 <.0001 6
3 1 130.854866 4.572114 <.0001 7
3 2 115.709205 3.507901 <.0001 8
3 3 119.344772 3.919740 <.0001 9
4 1 127.231370 3.541742 <.0001 10
4 2 129.096490 3.333471 <.0001 11
4 3 130.617967 3.634277 <.0001 12
Least Squares Means for effect DRUG*DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 34
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON CHANGE IN B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Class Level Information
Class Levels Values
DRUG 4 1 2 3 4
DISEASE 3 1 2 3
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 35
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON CHANGE IN B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: ISBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 12 6575.030151 547.919179 8.92 <.0001
Error 45 2765.125021 61.447223
Corrected Total 57 9340.155172
R-Square Coeff Var Root MSE ISBP Mean
0.703953 41.52073 7.838828 18.87931
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 1004.319714 334.773238 5.45 0.0028
DISEASE 2 77.440949 38.720474 0.63 0.5371
DRUG*DISEASE 6 480.017598 80.002933 1.30 0.2759
BSBP 1 2315.691646 2315.691646 37.69 <.0001
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -44.08633346 B 10.12112729 -4.36 <.0001
DRUG 1 -0.53427757 B 5.07762456 -0.11 0.9167
DRUG 2 3.81185569 B 4.74671300 0.80 0.4262
DRUG 3 -11.27319523 B 5.33623888 -2.11 0.0402
DRUG 4 0.00000000 B . . .
DISEASE 1 -3.38659762 B 4.97844733 -0.68 0.4998
DISEASE 2 -1.52147765 B 4.74671300 -0.32 0.7500
DISEASE 3 0.00000000 B . . .
DRUG*DISEASE 1 1 6.41776239 B 6.88237665 0.93 0.3561
DRUG*DISEASE 1 2 3.71120124 B 7.14046097 0.52 0.6058
DRUG*DISEASE 1 3 0.00000000 B . . .
DRUG*DISEASE 2 1 1.76391853 B 7.00741651 0.25 0.8024
DRUG*DISEASE 2 2 3.30885035 B 7.27268983 0.45 0.6513
DRUG*DISEASE 2 3 0.00000000 B . . .
DRUG*DISEASE 3 1 14.89669169 B 7.84420858 1.90 0.0640
DRUG*DISEASE 3 2 -2.11408942 B 7.08463646 -0.30 0.7668
DRUG*DISEASE 3 3 0.00000000 B . . .
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 36
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON CHANGE IN B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureDependent Variable: ISBP
Parameter EstimateStandard
Error t Value Pr > |t|
DRUG*DISEASE 4 1 0.00000000 B . . .
DRUG*DISEASE 4 2 0.00000000 B . . .
DRUG*DISEASE 4 3 0.00000000 B . . .
BSBP 0.58054117 0.09456791 6.14 <.0001
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 37
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON CHANGE IN B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG ISBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 21.2895030 2.1903578 <.0001 1
2 23.9502380 2.0953245 <.0001 2
3 11.4351316 2.3289553 <.0001 3
4 18.4474594 2.1229207 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: ISBP
i/j 1 2 3 4
1 0.3673 0.0042 0.3857
2 0.3673 0.0003 0.0821
3 0.0042 0.0003 0.0276
4 0.3857 0.0821 0.0276
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
DISEASE ISBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 20.4675757 1.8720848 <.0001 1
2 17.7895931 1.8520092 <.0001 2
3 18.0845802 1.8274747 <.0001 3
Least Squares Means for effect DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: ISBP
i/j 1 2 3
1 0.3097 0.3740
2 0.3097 0.9121
3 0.3740 0.9121
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 38
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEDRUG BY DISEASE EFFECTS ON CHANGE IN B.P., ADJUSTED FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG DISEASE ISBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 1 22.5803717 3.3839729 <.0001 1
1 2 21.7389305 4.0603864 <.0001 2
1 3 19.5492069 3.5083688 <.0001 3
2 1 22.2726611 3.6276523 <.0001 4
2 2 25.6827129 4.1210882 <.0001 5
2 3 23.8953402 3.3334706 <.0001 6
3 1 20.3203833 4.5721139 <.0001 7
3 2 5.1747222 3.5079012 0.1471 8
3 3 8.8102892 3.9197398 0.0295 9
4 1 16.6968869 3.5417419 <.0001 10
4 2 18.5620068 3.3334706 <.0001 11
4 3 20.0834845 3.6342770 <.0001 12
Least Squares Means for effect DRUG*DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 39
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESSMENT OF DRUG BY BSBP INTERACTION -- TEST OF HOMOGENEITY OF SLOPES
08:01 Monday, April 19, 2004
The GLM Procedure
Class LevelInformation
Class Levels Values
DRUG 4 1 2 3 4
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 40
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESSMENT OF DRUG BY BSBP INTERACTION -- TEST OF HOMOGENEITY OF SLOPES
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 7 32453.95927 4636.27990 78.74 <.0001
Error 50 2944.10969 58.88219
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.916829 5.929409 7.673473 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 271.33816 90.44605 1.54 0.2167
BSBP 1 18288.60540 18288.60540 310.60 <.0001
BSBP*DRUG 3 341.60897 113.86966 1.93 0.1361
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -23.34840426 B 16.58997516 -1.41 0.1655
DRUG 1 -54.31126788 B 26.20952684 -2.07 0.0434
DRUG 2 -18.58725894 B 22.13291416 -0.84 0.4050
DRUG 3 -6.17007090 B 29.73375116 -0.21 0.8365
DRUG 4 0.00000000 B . . .
BSBP 1.36125887 B 0.16155576 8.43 <.0001
BSBP*DRUG 1 0.51283950 B 0.23464594 2.19 0.0336
BSBP*DRUG 2 0.23335626 B 0.20607458 1.13 0.2629
BSBP*DRUG 3 -0.00719433 B 0.27892730 -0.03 0.9795
BSBP*DRUG 4 0.00000000 B . . .
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 41
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESSMENT OF DRUG EFFECT WITH NESTED SLOPE MODEL
08:01 Monday, April 19, 2004
The GLM Procedure
Class LevelInformation
Class Levels Values
DRUG 4 1 2 3 4
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 42
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESSMENT OF DRUG EFFECT WITH NESTED SLOPE MODEL
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 7 32453.95927 4636.27990 78.74 <.0001
Error 50 2944.10969 58.88219
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.916829 5.929409 7.673473 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 271.33816 90.44605 1.54 0.2167
BSBP(DRUG) 4 22558.49031 5639.62258 95.78 <.0001
Parameter EstimateStandard
Error t Value Pr > |t|
Intercept -23.34840426 B 16.58997516 -1.41 0.1655
DRUG 1 -54.31126788 B 26.20952684 -2.07 0.0434
DRUG 2 -18.58725894 B 22.13291416 -0.84 0.4050
DRUG 3 -6.17007090 B 29.73375116 -0.21 0.8365
DRUG 4 0.00000000 B . . .
BSBP(DRUG) 1 1.87409836 0.17017184 11.01 <.0001
BSBP(DRUG) 2 1.59461513 0.12793150 12.46 <.0001
BSBP(DRUG) 3 1.35406454 0.22737673 5.96 <.0001
BSBP(DRUG) 4 1.36125887 0.16155576 8.43 <.0001
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 43
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESEMENT OF HOMOGENEITY OF SLOPES IN THE COMPLETE DESIGN
08:01 Monday, April 19, 2004
The GLM Procedure
Class Level Information
Class Levels Values
DRUG 4 1 2 3 4
DISEASE 3 1 2 3
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 44
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGEASSESEMENT OF HOMOGENEITY OF SLOPES IN THE COMPLETE DESIGN
NOTE: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whoseestimates are followed by the letter 'B' are not uniquely estimable.
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 48
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGENO INTERACTION MODEL WITH COMMON SLOPE FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Class Level Information
Class Levels Values
DRUG 4 1 2 3 4
DISEASE 3 1 2 3
Number of observations 58
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 49
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGENO INTERACTION MODEL WITH COMMON SLOPE FOR BSBP
08:01 Monday, April 19, 2004
The GLM Procedure
Dependent Variable: FSBP
Source DFSum of
Squares Mean Square F Value Pr > F
Model 6 32152.92635 5358.82106 84.22 <.0001
Error 51 3245.14262 63.63025
Corrected Total 57 35398.06897
R-Square Coeff Var Root MSE FSBP Mean
0.908324 6.163838 7.976857 129.4138
Source DF Type III SS Mean Square F Value Pr > F
DRUG 3 1263.43283 421.14428 6.62 0.0007
DISEASE 2 40.57604 20.28802 0.32 0.7284
BSBP 1 19702.97372 19702.97372 309.65 <.0001
KUTNER BLOOD PRESSURE DATA -- UNBALANCED DESIGN 50
BSBP = BASELINE, FSBP = FOLLOWUP, ISBP = F - B CHANGENO INTERACTION MODEL WITH COMMON SLOPE FOR BSBP
08:01 Monday, April 19, 2004
The GLM ProcedureLeast Squares Means
DRUG FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 131.910254 2.188607 <.0001 1
2 134.399091 2.089746 <.0001 2
3 120.822956 2.322981 <.0001 3
4 128.868373 2.140399 <.0001 4
Least Squares Means for effect DRUGPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3 4
1 0.4022 0.0013 0.3525
2 0.4022 <.0001 0.0780
3 0.0013 <.0001 0.0123
4 0.3525 0.0780 0.0123
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.
DISEASE FSBP LSMEANStandard
Error Pr > |t|LSMEAN
Number
1 130.189924 1.868146 <.0001 1
2 128.220025 1.855264 <.0001 2
3 128.590556 1.856401 <.0001 3
Least Squares Means for effect DISEASEPr > |t| for H0: LSMean(i)=LSMean(j)
Dependent Variable: FSBP
i/j 1 2 3
1 0.4545 0.5524
2 0.4545 0.8904
3 0.5524 0.8904
NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.