On 17 Nov 2003 19:10:04 -0800, [EMAIL PROTECTED] (Thiyagarajan
Jayabaskar) wrote:

> Dear List members,
>   I am new to this list. 
> I have a qn on selection of statistical methods.
> The following is the description of the data.
> Drug A applied on to a sample and various parameters are assessed in fixed time 
> intervals.
> I need to show that there is a significant cahnge in the parameters over the time. 

If you are looking for *change*, you can to a paired t-test.
Or compute a linear trend component and compare it to
zero.  Or something similar.

If you are looking to show that the Drug was responsible,
you must have randomized an initial sample into two groups, 
so that you could see if Change (or trend) was greater 
in the Drug-A  group than in the sample that was otherwise
treated and tested in the identical way.


> I did the anova and post hoc to look in to the changes. 
> My qn is : is Anova the right procedure for this type of analyis?
> Or should I be using repeated measures analysis?

If you have 'repeated measures', then a repeated 
measures ANOVA  is the sort of ANOVA to perform.

> In that case how to do repeated measures in SPSS

Read about it and then ask questions, which you 
might post to the SPSS Usenet group.

-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
"Taxes are the price we pay for civilization." 
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to