Two way anova for clinical trial in SAS

[Barkamih]

Dear All,

I have clinical trial data, where I have control and treatment variables. The treatment variables have three readings for each.

How can I conduct this analysis and how to import this data properly.

Example of my data: TD is control. TN and PW are treatments.

I want to test TD against TN and  PW:

                TD                            TN                           PW

24H     20.3   21.2   20.6        30.1  31      32         40  40.1 42

48H     19.5    20      21          18   18.1   18.3        40  41   42

Thanks,

[StatsMan]

You can use a DATA Step to read in your data.  With the small amount of data, using a DATA Step will be the quickest way to get your data into a SAS data set.  Try something like

data test;
input trt $ y;
datalines;
TD 1.6
TN 2.4
;
run;

You will need 1 line for each observation, so just keep adding lines below the datalines; command and above the semicolon and run; lines.  Once you have the data typed in, submit that code and then

proc print data=test;
run;

to make sure the data is in the data set correctly.

Once your data is in a SAS data set, you can use PROC GLM to analyze your experiment.  Something like

proc glm data=test;

class trt;

model y=trt;

lsmeans trt / adjust=dunnett pdiff=control('TD');

run;

quit;

should do it.  The LSMEANS statement says to compare each of the treatment levels of the variable TRT to the control level (as specified through the PDIFF= sub-option).  

You can find documentation on PROC GLM here:

GLM Documentation