SAS COMMANDS FOR CRD WITH SAMPLING

options linesize=72;

data crdsamp;

input trt rep sample cwt;

*COMMENT: Each experimental unit will be

entered on its own line;

cards;

8 1 1 3.5

8 1 2 4.0

8 1 3 3.0

8 1 4 4.5

8 2 1 2.5

8 2 2 4.5

8 2 3 5.5

8 2 4 5.0

8 3 1 3.0

8 3 2 3.0

8 3 3 2.5

8 3 4 3.0

12 1 1 5.0

12 1 2 5.5

12 1 3 4.0

12 1 4 3.5

12 2 1 3.5

12 2 2 3.5

12 2 3 3.0

12 2 4 4.0

12 3 1 4.5

12 3 2 4.0

12 3 3 4.0

12 3 4 5.0

16 1 1 5.0

16 1 2 4.5

16 1 3 5.0

16 1 4 4.5

16 2 1 5.5

16 2 2 6.0

16 2 3 5.0

16 2 4 5.0

16 3 1 5.5

16 3 2 4.5

16 3 3 6.5

16 3 4 5.5

proc print;

proc anova;

classes rep trt;

model cwt=trt rep*trt;

*COMMENT: rep*trt equals the experimental error;

*COMMENT: The sampling error is not included in

the model;

*COMMENT The sampling error will appear at the top

of the output under the Error SOV under the Model SOV;

test h=trt e=rep*trt;

*COMMENT: The F-values for the first ANOVA table on

the page should be ignored because they were

calculated using the incorrect denominator term. By

default, SAS used the error term at the top of the

page (i.e, sampling error) as the denominator of the

F-test;

*COMMENT: This statement tells the computer to use

the experimental error as the denominator of the

F-test. By default, SAS uses the error at the top of

the page as the denominator of the F-test. A new

ANOVA with the correct F-values will be printed;

means trt/lsd e=rep*trt;

*COMMENT: We are telling the computer to use the

experimental error in the calculation of the LSD. By

default, SAS will use the error at the top of the output

for calculating the LSD;

title 'crd with sampling - expt error used as denominator';

CRD WITH SAMPLING LOG FILE

PAGE 1

The SAS System 17:28 Monday, November 27, 1995

NOTE: Copyright(c) 1989 by SAS Institute Inc., Cary, NC USA.

NOTE: SAS (r) Proprietary Software Release 6.07.01

Licensed to NORTH DAKOTA STATE UNIVERSITY, Site 0001220004.

NOTE: Running on IBM Model RS/6000 Serial Number 000132666600.

This message is contained in the SAS news file, and is presented

upon initialization. Edit the file "news" in the "misc" directory

to display site-specific news and information in the program log.

The command line option "-nonews" will prevent this display.

NOTE: AUTOEXEC processing beginning; file is /usr/local/sas/autoexec.sas.

NOTE: SAS initialization used:

real time 0.536 seconds

cpu time 0.300 seconds

NOTE: AUTOEXEC processing completed.

1 options linesize=72;

2 data crdsamp;

3 input trt rep sample cwt;

4 *COMMENT: Each experimental unit will be

5 entered on its own line;

6 cards;

NOTE: The data set WORK.CRDSAMP has 36 observations and 4 variables.

NOTE: DATA statement used:

real time 0.168 seconds

cpu time 0.090 seconds

43 proc print;

NOTE: The PROCEDURE PRINT printed page 1.

NOTE: PROCEDURE PRINT used:

real time 0.061 seconds

cpu time 0.050 seconds

44 proc anova;

45 classes rep trt;

46 model cwt=trt rep*trt;

47 *COMMENT: rep*trt equals the experimental error;

48 *COMMENT: The sampling error is not included in

49 the model;

50 *COMMENT: This analysis is for the unpooled error

51 term;

52 *COMMENT The sampling error will appear at the top

53 of the output under the Error SOV under the Model SOV;

54 test h=trt e=rep*trt;

55 *COMMENT: The F-values for the first ANOVA table on

56 the page should be ignored because they were

Page 2 The SAS System

17:28 Monday, November 27, 1995

57 calculated using the incorrect denominator term. By

58 default, SAS used the error term at the top of the

59 page (i.e, sampling error) as the denominator of the

60 F-test;

61 *COMMENT: This statement tells the computer to use

62 the experimental error as the denominator of the

63 F-test. By default, SAS uses the error at the top of

64 the page as the denominator of the F-test. A new

65 ANOVA with the correct F-values will be printed;

66 means trt/lsd e=rep*trt;

67 *COMMENT: We are telling the computer to use the

68 experimental error in the calculation of the LSD. By

69 default, SAS will use the error at the top of the output

70 for calculating the LSD;

71 title 'crd with sampling - expt error used as denominator';

NOTE: Means from the MEANS statement are not adjusted for other terms

in the model. For adjusted means, use the LSMEANS statement.

NOTE: The PROCEDURE ANOVA printed pages 2-4.

NOTE: PROCEDURE ANOVA used:

real time 0.376 seconds

cpu time 0.120 seconds

NOTE: PROCEDURE ANOVA used:

real time 0.239 seconds

cpu time 0.070 seconds

NOTE: The SAS System used:

real time 1.523 seconds

cpu time 0.730 seconds

NOTE: SAS Institute Inc., SAS Circle, PO Box 8000, Cary, NC 27512-8000

CRD WITH SAMPLING LISTING FILE

The SAS System Page 1

17:28 Monday, November 27, 1995

OBS	TRT	REP	SAMPLE	CWT
1	8	1	1	3.5
2	8	1	2	4.0
3	8	1	3	3.0
4	8	1	4	4.5
5	8	2	1	2.5
6	8	2	2	4.5
7	8	2	3	5.5
8	8	2	4	5.0
9	8	3	1	3.0
10	8	3	2	3.0
11	8	3	3	2.5
12	8	3	4	3.0
13	12	1	1	5.0
14	12	1	2	5.5
15	12	1	3	4.0
16	12	1	4	3.5
17	12	2	1	3.5
18	12	2	2	3.5
19	12	2	3	3.0
20	12	2	4	4.0
21	12	3	1	4.5
22	12	3	2	4.0
23	12	3	3	4.0
24	12	3	4	5.0
25	16	1	1	5.0
26	16	1	2	4.5
27	16	1	3	5.0
28	16	1	4	4.5
29	16	2	1	5.5
30	16	2	2	6.0
31	16	2	3	5.0
32	16	2	4	5.0
33	16	3	1	5.5
34	16	3	2	4.5
35	16	3	3	6.5
36	16	3	4	5.5

crd with sampling - expt error used as denominator Page 2

17:28 Monday, November 27, 1995

Analysis of Variance Procedure

Class Level Information

Class	Levels	Values
REP	3	1	2	3
TRT	3	8	12	16

Number of observations in data set = 36

Tests of Hypotheses using the Anova MS for REP*TRT as an error term

Sources of		Sums of	Mean
variation	df	square	square	F	PR > F
TRT	2	15.041667	7.520833	15.33	0.0001

crd with sampling - expt error used as denominator Page 4