Analysis Using SAS 

The analysis of the data is performed using PROC GLM of SAS. The SAS commands are given in the sequel.

Data Input:

For performing analysis, input the data in the following format. {Here one can call the replication as rep, main-plot treatments as ms (method of sowing) and sub-plot treatments as mt (manorial treatment). (It may, however, be noted that one can retain the same name or can code in any other fashion).

Prepare a SAS data file using

data splitplot;           /*one can enter any other name for data*/;

input rep  ms  mt yield;

cards;

1          1          1          4940

1          1          2          4810

1          1          3          5150

1          1          4          5090

1          1          5          5130

1          1          6          5140

1          2          1          4900

1          2          2          4920

1          2          3          5070

1          2          4          4890

1          2          5          5150

1          2          6          5070

2          1          1          4830

2          1          2          5110

2          1          3          4920

2          1          4          4900

2          1          5          4880

2          1          6          4930

2          2          1          5020

2          2          2          5110

2          2          3          5230

2          2          4          5120

2          2          5          5160

2          2          6          5200

3          1          1          5080

3          1          2          5160

3          1          3          5180

3          1          4          5190

3          1          5          5160

3          1          6          5280

3          2          1          5090

3          2          2          5130

3          2          3          4980

3          2          4          5200

3          2          5          4920

3          2          6          5250

;

Perform the analysis of the data using the following statements.

proc glm;

class rep ms mt;

model yield=rep ms rep*ms mt ms*mt;

test h= ms e = rep*ms;

means ms /lsd e=rep*ms;

means mt ms*mt/lsd;

lsmeans ms*mt/pdiff;

run;

Note: In the test statement H = numerator for source of variation and

                                            E = denominator source of variation

One can use TUKEY or Scheffe in place of lsd for making all possible pairwise comparisons. Statement  means mt ms*mt/lsd; gives only means and standard deviations for level combinations of ms and mt. For pairwise comparison of level combinations of ms and mt, the statement lsmeans ms*mt/pdiff is used. One may, however, compute only the means from the software and compute the minimum significant differences using the given formulae on the click of mouse.