Question

(use TUKEY for the multiple comparisons)

Create a SAS data set by reading these data. As an interesting exercise, try reading three lines like this: 7.2 Two cholesterol-lowering medications (statins) and a placebo were given to each of 10 volunteers with total cholesterol readings of 240 or higher. After 6 weeks, the following total cholesterol values were recorded: Statin A: 220 190 180 185 210 170 178 200 177 189 Statin B: 160 168 178 200 172 155 159 167 185 199 Placebo: 240 220 246 244 198 238 277 255 190 188 220 190 180 185 210 170 178 200 177 189 160 168 178 200 172 155 159 167 185 199 240 220 246 244 198 238 277 255 190 188 That is, have the program create the treatment variable (call it TREAT) with values of A','B'.and Placebo. Next run a oneway ANOVA followed by a Student-Newman-Keuls

Answer 1

Solution:

SAS CODE:

data cholesterol;
infile cards;
input treat $ cholvalues;
cards;
A 220
A 190
A 180
A 185
A 210
A 170
A 178
A 200
A 177
A 189
B 160
B 168
B 178
B 200
B 172
B 155
B 159
B 167
B 185
B 199
placebo 240
placebo 220
placebo 246
placebo 244
placebo 198
placebo 238
placebo 277
placebo 255
placebo 190
placebo 188
;
run;

proc anova data=cholesterol;
class treat;
model cholvalues = treat;
means treat / tukey cldiff ;
run;

output:

The ANOVA Procedure
Dependent Variable: cholvalues
Source	DF	Sum of Squares	Mean Square	F Value	Pr > F
Model	2	16258.47	8129.233	17.58	<.0001
Error	27	12483.4	462.3482
Corrected Total	29	28741.87

From ANOVA table

F(2,27)=17.58

p<0.0001

p<0.05

Reject H0

there is sufficient statistical evidence at 5% level of significance to conclude that all the 3 group means are different.

The ANOVA Procedure
Tukey's Studentized Range (HSD) Test for cholvalues
Note:This test controls the Type I experimentwise error rate.
Alpha	0.05
Error Degrees of Freedom	27
Error Mean Square	462.3481
Critical Value of Studentized Range	3.50633
Minimum Significant Difference	23.842
Comparisons significant at the 0.05 level are indicated by ***.
treat	Difference	Simultaneous 95% Confidence Limits
Comparison	Between
	Means
placebo - A	39.7	15.858	63.542	***
placebo - B	55.3	31.458	79.142	***
A - placebo	-39.7	-63.542	-15.858	***
A - B	15.6	-8.242	39.442
B - placebo	-55.3	-79.142	-31.458	***
B - A	-15.6	-39.442	8.242

from tukey paiwise comparsions

A - placebo

B - placebo

are different