Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: correlated data


From   Richard Hiscock <richardhiscock@me.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: correlated data
Date   Tue, 01 Sep 2009 23:23:51 +1000

Nick,
To my mind the scanned raw data signal is the variable (as per BP or  
heart rate)
The instrument that performs the measurement (software) is the  
measurement technique under examination and measures the raw data with  
error In this case the old standard vs 3 new techniques
I don't see the correlation problem, isn't it one of assessing  
agreement on a continuous scale & which technique is least biased?
You could perform analyses of old standard vs each of the new  
techniques.
I believe that OLS regression is not appropriate as  both y (old  
technique) & x (new technique) variables are both measured with error
Approaches suggested include least products regression or the Bland  
Altman method of differences

this reference may be helpful: Ludbrook, J.  Statistical techniques  
for comparing measures and methods of measurement: a critical review.
Clinical and Experimental Pharmacology and Physiology. 2002;29:527 -  
536.

Cheers Richard Hiscock

On 1 Sep 2009, at 13:33, Nikolaos Pandis wrote:

> Hi to all.
>
> I would like to ask the following question.
>
> We have a set of 3-D images constructed from cat scans, and we are  
> measuring volumes defined by certain anatomical points on the 3-D  
> images.
>
> The reconstruction/measuring technique is performed using 3 new  
> types of software and their results will be compared with the  
> results of validated/reference technique.
>
> The same reconstructions/cat scans are used for all techniques.
>
> The objective is to see how close (do they differ significantly?)  
> the volume values recorded by each technique are to the values  
> recorded by the reference technique.
>
> I was thinking along the lines of regression with the  
> volume(continuous) variable as the dependent variable and technique  
> as the categorical dependent variable with 4 levels. The reference  
> level would be the the standard/validated method.
>
> However, how would I account for the fact that the data is  
> correlated since all measurements for the 4 methods are taken from  
> the same reconstructions/scans?
>
> Any suggestions would be greatly appreciated.
>
> Many thanks,
>
> Nick
>
>
>
>
> *
> *   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index