Dear All, I work as a statistical analyst mainly in the area of finance. I am a SAS programmer work maily on time series analysis. This query is to solve a problem in analyzing medical data. I will be greateful if somebody in this list guide me how to do it. The drug is a variant of Viagra. Two randomly selected groups undergone treatment with placibo and the drug. Data were collected at the end of 2 weeks 4 weeks. For example, if placibo is given to the first group, the second group was given the drug and after 2 weeks, the drug is given to first group and placibo to the second. the response of the patients are at 5 leveles (like always -5, somtimes-3 etc) the data look like Baseline (W0) PlaciboW2 DrugW2 Placibow4 Drugw4 ---------------------------------------------------------PNO>qno1(score) qno1(score) w0- baseline, w2-end week2, w4-end of week4,qno1-quistion no -1 (there are 16 quistion to analyse) I think I could convey the data information. Now we need to analyze this to find out the effect of the drug and placibo based on the baseline. secondly , to find out the carry over effect. Please let me know what technique best suit for this and how to design the data analysis. How to carry out Chi square and ANOVA on such data. What are the other useful techniques to explain the effects better ? This result is very important for the company. Biometrics is an exciting area to me but first time I am getting a chance to try out something. Since I want to do it perfect I decided to get the help of the online community.. thanks in advance, Jeomoan Kurian ________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%