st: R: which statistics to use

["Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>]

Dear b,
Though I am not sure if I am reading your data correctly (due to plain text
format), I would suggest an OLS with clustered standard error:
................begin example...................................
set obs 12
input concentration sample condition time id
.2 1 1 0 1
.3 1 1 1 2
.3 1 1 2 3
.2 2 1 0 4
.5 2 1 1 5
.9 2 1 2 6
.7 1 2 0 1
.7 1 2 1 2
.9 1 2 2 3
.7 2 2 0 4
1.02 2 2 1 5
1.03 2 2 2 6
end
regress concentration sample  condition time, robust cluster(id) 
Dr. Carlo Lazzaro

Studio di economia sanitaria

Via Stefanardo da Vimercate, 19

20128 Milano

tel/fax: 02 26000516

portatile: 335 6786741

e-mail: carlo.lazzaro@tin.it

            carlo.lazzaro@tiscalinet.it

-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di b. water
Inviato: lunedì 22 marzo 2010 3.03
A: statalist@hsphsun2.harvard.edu
Oggetto: st: which statistics to use

Dear all,
 
Stata 8.2,
 
I have data consisting of serum concentrations measured at regular intervals
but under two different conditions (but a few missing reading here and
there) like so:

 
  condition_1  condition_2
  t0 t1 t2 t0 t1 t2
sample_1 0.2 0.3 0.3 0.2 0.5 0.9
sample_2 0.7 0.7 0.9 0.7 1.2 1.3
.
.
.
.
.
.
sample_n 0.3 0.5 0.6 0.3 0.8 1.1

Just 'eye-balling' the data I can see that condition_2 did affect the
concentrations. My questions:
 
1. How may I show that in condition_1 concentration at t1 and t2 _are_
statistically different from t0 (if indeed they are). Had it been just t0
with one other time interval, I would have thought of paired t-test off the
bat but don't what to do in this circumstances. It looks like ANOVA but
again would appreciate advice.
 
2. How may I show (statistically that is - and therefore what statistical
analyses I should deploy) that w.r.t. t1 and t2 that condition_2 IS
different from condition_1.
 
Thank you for advice/help.
 
bw  		 	   		  
_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
http://clk.atdmt.com/GBL/go/210850553/direct/01/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/