| Date: | Tue, 27 Jan 2004 12:00:24 -0600 |
|---|
| Reply-To: | Paul R Swank <Paul.R.Swank@UTH.TMC.EDU> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Paul R Swank <Paul.R.Swank@UTH.TMC.EDU> |
|---|
| Subject: | Re: repeated measures analysis of variance |
|---|
|
| In-Reply-To: | <BAY7-F125VAEA7Cu6hf0005411a@hotmail.com> |
|---|
| Content-Type: | text/plain; charset="us-ascii" |
|---|
The time_1 contrast represents the difference between y1 and y2. Thus, grupo
by time_1 is the difference between groups on the difference between y 1 and
y2. The second contrast should be time_2, representing the differences
between y2 and y3 but seems to be a duplicate of the first one.
Paul R. Swank, Ph.D.
Professor, Developmental Pediatrics
Medical School
UT Health Science Center at Houston
-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of lamack
lamack
Sent: Tuesday, January 27, 2004 6:15 AM
To: SAS-L@LISTSERV.UGA.EDU
Subject: repeated measures analysis of variance
Dear all, I have a question about repeated measures anova interpretation.
Running the sas code bellow I have:
1) The within subject test indicate that the interaction of time and group
is significant. The main effect of time is not significant. However, the
significant interaction indicates that the groups are changing over time and
they are changing in diferent ways.
2)from Tests of Hypotheses for Between Subjects Effects I have a significant
diference between gropus.
From sas output, my question is. What is the interpretation for the follwing
sas output getting by: repeated time 3 profile/summary?
Contrast Variable: time_1
Source DF Type III SS Mean Square F Value Pr > F
Mean 1 1.2071429 1.2071429 0.50 0.4829
grupo 1 304.0960317 304.0960317 126.80 <.0001
Error 34 81.5428571 2.3983193
Contrast Variable: time_1
Source DF Type III SS Mean Square F Value Pr > F
Mean 1 1.2071429 1.2071429 0.50 0.4829
grupo 1 304.0960317 304.0960317 126.80 <.0001
Error 34 81.5428571 2.3983193
The sas Code.
data test;
input id group y1-y3@@;
cards;
1 b 82 76 77
2 b 84 79 79
3 b 85 82 82
4 b 76 74 75
5 b 83 82 83
6 b 89 83 85
7 b 85 83 83
8 b 77 76 76
9 b 91 88 91
10 b 86 81 82
11 b 85 82 82
12 b 90 89 89
13 b 84 83 83
14 b 84 82 82
15 b 83 80 80
16 b 78 77 76
17 b 82 80 81
18 b 80 78 79
19 b 81 79 79
20 b 88 84 84
21 b 87 84 85
22 f 75 79 79
23 f 76 77 77
24 f 72 74 73
25 f 80 85 84
26 f 75 78 78
27 f 71 72 71
28 f 69 74 74
29 f 78 80 80
30 f 69 73 73
31 f 72 75 74
32 f 76 79 79
33 f 74 77 77
34 f 73 75 75
35 f 82 85 85
36 f 79 85 84
run;
proc glm data = dados;
class group;
model y1-y3 = gruop/nouni;
repeated time 3 profile/summary;
run;
run;
_________________________________________________________________
MSN Hotmail, o maior webmail do Brasil. http://www.hotmail.com
| 
