Bioequivalence and Bioavailability Forum

Handling of dividing subject in two groups [General Sta­tis­tics]

posted by Pankaj Bhangale – India, 2011-09-03 11:33 (5399 d 05:49 ago) – Posting: # 7331
Views: 3,884

Dear BEman,

❝ in a 2x2-crossover BE-Study with 80 subject the subjects are divided in two groups with 40 Subject due to clinical capacity.

❝ d) Way proposed form a non-statistican (of our sponsor)

Following is 2 x 2-ANOVA Model with multiple groups

Proc GLM Data=work.Dummi;
  class Sequence Subject Group Formulation Period;
  model LogAUCinf LogAUCT LogCmax = Sequence Group Subject(Sequence*Group) Formulation Period/ss1 ss3;
  test h=Sequence   e=Subject(Sequence*Group);
  test h=Group   e=Subject(Sequence*Group);
  lsmeans Formulation Period/stderr pdiff;
  estimate 'R vs T' formulation -1 1;
  run;quit;

After execution of above SAS code, there is no zero degrees of freedom for the factor group.


Best Regards,
Pankaj Bhangale

Complete thread:

• Handling of dividing subject in two groups beman 2011-09-01 16:24 
  • Handling of dividing subject in two groups Helmut 2011-09-01 17:24
  • Handling of dividing subject in two groupsPankaj Bhangale 2011-09-03 09:33