Hi All,

Just a quick question. SAS Can Be used t generate treatment combinations, But I have problem because I never analyzed data with factorial design more one treatment control. For example I have 3 ^ 5 (3 sources of protein and 5 levels) factorial design one more treatment control (without level) in a randomized design. My doubts are the following:

  • How should consider this treatment in control analysis.
  • If the interaction is significant levels*sources should to make the regression.

So what is the best method of doing this analysis? Below is a small example and various analysis I did, but what is the best ..

DATA fish; 
INPUT TReAT$ sources level      control GP; 
IF level = 1 THEN level = 0; 
cards; 
CONTROL 1       1       1       107.1 
CONTROL 1       1       1       118.2 
CONTROL 1       1       1       69.8 
CONTROL 1       1       1       121.93 
CONTROL 1       1       1       58.2 
2SOJA   2       2       0       106.6846667 
2SOJA   2       2       0       100.1246667 
2SOJA   2       2       0       96.08466667 
2SOJA   2       2       0       92.08466667 
2SOJA   2       2       0       157.6846667 
4SOJA   2       4       0       94.83 
4SOJA   2       4       0       101.63 
4SOJA   2       4       0       80.93 
4SOJA   2       4       0       127.73 
4SOJA   2       4       0       116.53 
4SOJA   2       4       0       71.73 
6SOJA   2       6       0       104.154 
6SOJA   2       6       0       85.544 
6SOJA   2       6       0       64.804 
6SOJA   2       6       0       110.184 
6SOJA   2       6       0       93.174 
8SOJA   2       8       0       69.44933333 
8SOJA   2       8       0       67.68933333 
8SOJA   2       8       0       147.0293333 
8SOJA   2       8       0       60.96933333 
8SOJA   2       8       0       45.64933333 
10SOJA  2       10      0       94.34866667 
10SOJA  2       10      0       99.92866667 
10SOJA  2       10      0       115.4686667 
10SOJA  2       10      0       98.76866667 
10SOJA  2       10      0       80.59866667 
2MILHO  3       2       0       121.5706667 
2MILHO  3       2       0       104.1606667 
2MILHO  3       2       0       87.90066667 
2MILHO  3       2       0       106.7106667 
2MILHO  3       2       0       66.80066667 
4MILHO  3       4       0       121.582 
4MILHO  3       4       0       65.742 
4MILHO  3       4       0       69.182 
4MILHO  3       4       0       115.322 
4MILHO  3       4       0       125.722 
6MILHO  3       6       0       76.92133333 
6MILHO  3       6       0       83.32133333 
6MILHO  3       6       0       179.6213333 
6MILHO  3       6       0       95.72133333 
6MILHO  3       6       0       92.52133333 
8MILHO  3       8       0       89.66333333 
8MILHO  3       8       0       108.0533333 
8MILHO  3       8       0       76.31333333 
8MILHO  3       8       0       126.5133333 
10MILHO 3       10      0       119.628 
10MILHO 3       10      0       88.858 
10MILHO 3       10      0       102.468 
10MILHO 3       10      0       164.128 
10MILHO 3       10      0       93.068 
2LINHAÇA       4       2       0       122.952 
2LINHAÇA       4       2       0       101.622 
2LINHAÇA       4       2       0       88.152 
2LINHAÇA       4       2       0       87.252 
2LINHAÇA       4       2       0       95.242 
4LINHAÇA       4       4       0       76.692 
4LINHAÇA       4       4       0       102.442 
4LINHAÇA       4       4       0       75.162 
4LINHAÇA       4       4       0       85.122 
4LINHAÇA       4       4       0       84.332 
6LINHAÇA       4       6       0       102.662 
6LINHAÇA       4       6       0       120.512 
6LINHAÇA       4       6       0       118.872 
6LINHAÇA       4       6       0       57.362 
6LINHAÇA       4       6       0       73.212 
8LINHAÇA       4       8       0       93.32666667 
8LINHAÇA       4       8       0       132.1766667 
8LINHAÇA       4       8       0       128.9466667 
8LINHAÇA       4       8       0       67.36666667 
8LINHAÇA       4       8       0       101.5266667 
8LINHAÇA       4       8       0       132.9866667 
10LINHAÇA      4       10      0       123.7566667 
10LINHAÇA      4       10      0       128.9566667 
10LINHAÇA      4       10      0       160.5566667 
10LINHAÇA      4       10      0       101.3566667 
10LINHAÇA      4       10      0       121.6566667 
10LINHAÇA      4       10      0       91.15666667 
10LINHAÇA      4       10      0       123.3566667 
10LINHAÇA      4       10      0       78.15666667 
10LINHAÇA      4       10      0       84.35666667 
10LINHAÇA      4       10      0       101.7566667 
10LINHAÇA      4       10      0       93.45666667 
10LINHAÇA      4       10      0       104.4566667 
10LINHAÇA      4       10      0       93.45666667 
10LINHAÇA      4       10      0       118.2566667 
10LINHAÇA      4       10      0       128.4566667 
10LINHAÇA      4       10      0       75.19933333 
10LINHAÇA      4       10      0       60.66933333 
10LINHAÇA      4       10      0       61.66933333 
10LINHAÇA      4       10      0       61.66933333 
10LINHAÇA      4       10      0       84.36933333 
10LINHAÇA      4       10      0       55.86933333 
10LINHAÇA      4       10      0       57.26933333 
10LINHAÇA      4       10      0       33.76933333 
10LINHAÇA      4       10      0       65.96933333 
10LINHAÇA      4       10      0       110.0693333 
10LINHAÇA      4       10      0       95.56933333 
10LINHAÇA      4       10      0       78.06933333 
10LINHAÇA      4       10      0       87.66933333 
10LINHAÇA      4       10      0       99.36933333 
10LINHAÇA      4       10      0       33.66933333 
10LINHAÇA      4       10      0       106.9233333 
10LINHAÇA      4       10      0       120.3933333 
10LINHAÇA      4       10      0       124.3933333 
10LINHAÇA      4       10      0       104.2233333 
10LINHAÇA      4       10      0       113.8233333 
10LINHAÇA      4       10      0       145.1233333 
10LINHAÇA      4       10      0       109.9233333 
10LINHAÇA      4       10      0       144.2233333 
10LINHAÇA      4       10      0       117.6233333 
10LINHAÇA      4       10      0       102.2233333 
10LINHAÇA      4       10      0       171.7233333 
10LINHAÇA      4       10      0       95.62333333 
10LINHAÇA      4       10      0       122.0486667 
10LINHAÇA      4       10      0       91.41866667 
10LINHAÇA      4       10      0       110.4586667 
10LINHAÇA      4       10      0       86.15866667 
10LINHAÇA      4       10      0       98.90866667 
; 
Run; 

Option 1

proc mixed data=fish; 
class Treat sources level ; 
model GP = sources|level/outp=xx2 residuals; 
lsmeans sources*level /adjust=Tukey pdiff slice = level 
slice=sources; 
run;quit; 

Option 2

proc mixed data=fish; 
class control sources level; 
model GP = control sources(control) level(control) sources*level 
(control)/outp=xx1 ddfm=sat residuals; 
lsmeans sources*level (control)/ slice= level(control); 
run;quit; 

Option 3

proc glm data=peixe; 
class TReAT; 
model GP =treat/ss3 clparm; 
means treat; 
means treat/dunnett('CONTROL'); 
contrast ‘control vs other ……… 
…………………… 
run; 
quit; 

Option 4

DATA REGPOL; 
SET fish; 
IF TReAT = 'CONTROL' THEN DELETE; 
NQ = level*level; 
NC = level*level*level; 
NQ2 = level*level*level*level;RUN; 
proc sort data=REGPOL;by sources;RUN; 
proc mixed data=REGPOL method=ml; 
class level; 
model GP =level/residuals outp=GP; 
Contrast 'level linear' level -2 -1 0 2 1; 
Contrast 'level Quadratic' level 2 -1 -2 -1 2; 
Contrast 'level Cubic' level -1 2 0 -2 1; 
Contrast 'level Q' level 1 -4 6 -4 1; 
by sources; 
run; 

Option 5

proc REG data=REGPOL; 
model GP =level NQ NC NQ2; 
BY sources; 
RUN; 
QUIT; 

Thanks for any help.

asked 02 Jul '10, 01:49

SAS-L's gravatar image

SAS-L ♦♦
425720
accept rate: 0%


How was the study designed? Are these fixed effects or random effects?

link

answered 02 Jul '10, 17:26

MarkLamias's gravatar image

MarkLamias ♦
838215
accept rate: 20%

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Tags:

×1
×1

Asked: 02 Jul '10, 01:49

Seen: 1,520 times

Last updated: 16 Jul '11, 14:22

Related questions

powered by OSQA