|
|
Practice
Computer
Programs
SAS
headstart
SAS
procedures
Program
1
Program
2
Program
3
Program 4
Program
5
Program
6
Program
7
Program
8
Program
9
Program
10
Program
11
Program
12
Program
13
Program
14
Topical
Bibliographies (Readings)
Datasets
and Stories
Software
Help
Search
for
|
|
Program
4
OPTIONS ls=75 NODATE PAGENO=1;
TITLE 'Fixed effect model applied to CRF design, p. 369 of Kirk, 1994';
DATA p369;
INPUT a b score;
CARDS;
1 1 24
1 1 33
1 1 37
1 1 29
1 1 42
1 2 44
1 2 36
1 2 25
1 2 27
1 2 43
1 3 38
1 3 29
1 3 28
1 3 47
1 3 48
2 1 30
2 1 21
2 1 39
2 1 26
2 1 34
2 2 35
2 2 40
2 2 27
2 2 31
2 2 22
2 3 26
2 3 27
2 3 36
2 3 46
2 3 45
3 1 21
3 1 18
3 1 10
3 1 31
3 1 20
3 2 41
3 2 39
3 2 50
3 2 36
3 2 34
3 3 42
3 3 52
3 3 53
3 3 49
3 3 64
;
PROC PRINT;
PROC MEANS;
VAR score;
BY a b;
PROC GLM;
CLASS a b;
MODEL score=a b a*b;
MEANS a b a*b/SIDAK TUKEY;
RUN;
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 1
OBS A B SCORE
1 1 1 24
2 1 1 33
3 1 1 37
4 1 1 29
5 1 1 42
6 1 2 44
7 1 2 36
8 1 2 25
9 1 2 27
10 1 2 43
11 1 3 38
12 1 3 29
13 1 3 28
14 1 3 47
15 1 3 48
16 2 1 30
17 2 1 21
18 2 1 39
19 2 1 26
20 2 1 34
21 2 2 35
22 2 2 40
23 2 2 27
24 2 2 31
25 2 2 22
26 2 3 26
27 2 3 27
28 2 3 36
29 2 3 46
30 2 3 45
31 3 1 21
32 3 1 18
33 3 1 10
34 3 1 31
35 3 1 20
36 3 2 41
37 3 2 39
38 3 2 50
39 3 2 36
40 3 2 34
41 3 3 42
42 3 3 52
43 3 3 53
44 3 3 49
45 3 3 64
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 2
Analysis Variable : SCORE
---------------------------------- A=1 B=1 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 33.0000000 6.9641941 24.0000000 42.0000000
----------------------------------------------------------
---------------------------------- A=1 B=2 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 35.0000000 8.8034084 25.0000000 44.0000000
----------------------------------------------------------
---------------------------------- A=1 B=3 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 38.0000000 9.5131488 28.0000000 48.0000000
----------------------------------------------------------
---------------------------------- A=2 B=1 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 30.0000000 6.9641941 21.0000000 39.0000000
----------------------------------------------------------
---------------------------------- A=2 B=2 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 31.0000000 6.9641941 22.0000000 40.0000000
----------------------------------------------------------
---------------------------------- A=2 B=3 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 36.0000000 9.5131488 26.0000000 46.0000000
----------------------------------------------------------
---------------------------------- A=3 B=1 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 20.0000000 7.5166482 10.0000000 31.0000000
----------------------------------------------------------
---------------------------------- A=3 B=2 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 40.0000000 6.2048368 34.0000000 50.0000000
----------------------------------------------------------
---------------------------------- A=3 B=3 --------------------------------
N Mean Std Dev Minimum Maximum
----------------------------------------------------------
5 52.0000000 7.9686887 42.0000000 64.0000000
----------------------------------------------------------
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 4
General Linear Models Procedure
Class Level Information
Class Levels Values
A 3 1 2 3
B 3 1 2 3
Number of observations in data set = 45
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 5
General Linear Models Procedure
Dependent Variable: SCORE
Sum of Mean
Source DF Squares Square F Value Pr > F
Model 8 2970.00000 371.25000 5.94 0.0001
Error 36 2250.00000 62.50000
Corrected Total 44 5220.00000
R-Square C.V. Root MSE SCORE Mean
0.568966 22.58770 7.90569 35.0000
Source DF Anova SS Mean Square F Value Pr > F
A 2 190.00000 95.00000 1.52 0.2324
B 2 1543.33333 771.66667 12.35 0.0001
A*B 4 1236.66667 309.16667 4.95 0.0028
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 6
General Linear Models Procedure
Tukey's Studentized Range (HSD) Test for variable: SCORE
NOTE: This test controls the type I experimentwise error rate, but
generally has a higher type II error rate than REGWQ.
Alpha= 0.05 df= 36 MSE= 62.5
Critical Value of Studentized Range= 3.457
Minimum Significant Difference= 7.0561
Means with the same letter are not significantly different.
Tukey Grouping Mean N A
A 37.333 15 3
A
A 35.333 15 1
A
A 32.333 15 2
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 7
General Linear Models Procedure
Sidak T tests for variable: SCORE
NOTE: This test controls the type I experimentwise error rate, but
generally has a higher type II error rate than REGWQ.
Alpha= 0.05 df= 36 MSE= 62.5
Critical Value of T= 2.50
Minimum Significant Difference= 7.2283
Means with the same letter are not significantly different.
Sidak Grouping Mean N A
A 37.333 15 3
A
A 35.333 15 1
A
A 32.333 15 2
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 8
General Linear Models Procedure
Tukey's Studentized Range (HSD) Test for variable: SCORE
NOTE: This test controls the type I experimentwise error rate, but
generally has a higher type II error rate than REGWQ.
Alpha= 0.05 df= 36 MSE= 62.5
Critical Value of Studentized Range= 3.457
Minimum Significant Difference= 7.0561
Means with the same letter are not significantly different.
Tukey Grouping Mean N B
A 42.000 15 3
A
A 35.333 15 2
B 27.667 15 1
1 Fixed effect model applied to CRF design, p. 369 of Kirk, 1994 9
General Linear Models Procedure
Sidak T tests for variable: SCORE
NOTE: This test controls the type I experimentwise error rate, but
generally has a higher type II error rate than REGWQ.
Alpha= 0.05 df= 36 MSE= 62.5
Critical Value of T= 2.50
Minimum Significant Difference= 7.2283
Means with the same letter are not significantly different.
Sidak Grouping Mean N B
A 42.000 15 3
A
A 35.333 15 2
B 27.667 15 1
Level of Level of ------------SCORE------------
A B N Mean SD
1 1 5 33.0000000 6.96419414
1 2 5 35.0000000 8.80340843
1 3 5 38.0000000 9.51314880
2 1 5 30.0000000 6.96419414
2 2 5 31.0000000 6.96419414
2 3 5 36.0000000 9.51314880
3 1 5 20.0000000 7.51664819
3 2 5 40.0000000 6.20483682
3 3 5 52.0000000 7.96868873
Comments: peng@indiana.edu
Dr. Peng's Home Page: Dr.
Chao-Ying Joanne Peng
Copyright
1999, The Trustees of Indiana
University
|
