1

I have collected data from two groups (age 25-30, age 30+). In each group I have 10 patients. I performed several medical tests (i.e., R.B.C, Liver test, etc.) on the patients, before and after treatment.

The data look like this:

Group 1
                    Test1           ...          Test 10   
               Before    After                  Before    After

patient 1       10        16                     13  
patient 2       12        11                     19

It's the same for Group 2.

Question

How can I test the effect of the treatment in both groups?

Jeromy Anglim
  • 44,984
Bioinformatics
  • 287
  • 1
    See also this question on best practice for examining differences between groups in a pre-post design: http://stats.stackexchange.com/questions/3466/best-practice-when-analysing-pre-post-treatment-control-designs – Jeromy Anglim Jul 23 '12 at 07:08
  • Can't you simply do a mutliway ANOVA? – Michael R. Chernick Jul 23 '12 at 11:14
  • @Michael: I thought that multiway ANOVA will work perfectly if there are not dependent variables (i.e., Before and After). Do multiway ANOVA consider the dependent variables? – Bioinformatics Jul 23 '12 at 16:42
  • Aren't the groups what form the dependent variable? Regression, analysis of variance and analysis of covariance all fall under the category of the general linear model. You can formulate your problem as a general linear model. I think maybe you meant that you have independent or predictor variables. If that is the case analysis of covariance is possible. – Michael R. Chernick Jul 23 '12 at 20:19

0 Answers0