先看数据和程序
<br />
Data Fengluan;<br />
input Trt Blk sprich;<br />
datalines;<br />
1 1 0<br />
1 2 3<br />
1 3 2<br />
1 4 4<br />
2 1 5<br />
2 2 6<br />
2 3 8<br />
2 4 3<br />
3 1 5<br />
3 2 5<br />
3 3 7<br />
3 4 6<br />
;<br />
proc format;<br />
value Trtshow 1='1RMon' 2='2RC' 3='3RS';<br />
run;</p>
<p>proc format;<br />
value Blkshow 1='A' 2='B' 3='C' 4='D';<br />
run;</p>
<p>proc sort data=Fengluan;<br />
by Trt;<br />
run;</p>
<p>PROC PRINT DATA=Fengluan;<br />
Title 'DATA Fengluan';<br />
format Trt Trtshow.;<br />
format Blk Blkshow.;<br />
run;</p>
<p>Proc means data=Fengluan n mean stderr;<br />
var sprich;<br />
by Trt;<br />
run;</p>
<p>Proc report data = Fengluan nowindows headline missing;<br />
Column Trt sprich,(n Mean Stderr);<br />
Define Trt / Group;<br />
format Trt Trtshow.;<br />
Run;</p>
<p>Proc glm data = Fengluan;<br />
Class Trt Blk;<br />
Model sprich = Trt Blk;<br />
Lsmeans Trt / adjust = tukey;<br />
Output out = res p = pred r = resid;</p>
<p>*Test for normality;<br />
Proc univariate data = res plot normal;<br />
var resid;</p>
<p>Proc plot data = res;<br />
Plot resid * (Trt Blk pred);<br />
run;<br />
quit;<br />
</p>
运行上述代码后，发现有处理效应：
<br />
Source DF Type I SS Mean Square F Value Pr > F<br />
Trt 2 30.50000000 15.25000000 5.66 0.0416<br />
Blk 3 8.33333333 2.77777778 1.03 0.4433<br />
</p>
但是，Tukey检验结果却是：
<br />
Least Squares Means for effect Trt<br />
Pr > |t| for H0: LSMean(i)=LSMean(j)</p>
<p> Dependent Variable: sprich</p>
<p> i/j 1 2 3</p>
<p> 1 0.0698 0.0534<br />
2 0.0698 0.9749<br />
3 0.0534 0.9749<br />
</p>
这是为什么呢？？？？？？？请高手指教,不甚感激！
