Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Checking reliability of a measurement device

From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: Checking reliability of a measurement device
Date   Fri, 26 Nov 2004 17:51:02 -0000

At one level this is simple, unless you 
are happy to act under instruction. 

He seems to have things the wrong way around. 
The real question is likely to be 

(a) to check how far the measurements are reliable
(b) for the purposes for which they are to be used. 

He needs to explain (b); you might be able to 
illuminate how to answer (a) in that context. 

In addition, there is a presumption here that
(a) necessarily involves a test, or a summary 
measure, or a model. How about some graphs too? 
There are some suggestions in 

Graphing agreement and disagreement. 
Stata Journal 4(3): 329--349 (2004)

Sometimes you have to dig your heels in. 

I once was called in to give advice to a
younger scientist, now very well known in his field. 
His paper submitted to Nature had been rejected 
because it was based on just eight data points. 
He was indignant and wanted some statistical way of 
compensating for the small sample size, and was even more 
indignant when I sided with the reviewers and told 
him the only remedy was to collect more data. 

[email protected] 

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of n p
> Sent: 26 November 2004 11:52
> To: [email protected]
> Subject: st: Checking reliability of a measurement device
> Hi Statalisters, 
> a Dentistry PhD student did some measurements on 12
> teeth with varying conditions and he asked me how
> could he show that the device used for the
> measurements is reliable. More specifically each one
> of the 12 teeth has been measured by this device by 2
> raters (a and b) X 2 time points (week 1 and week 2) X
> 6 relative positions =   24 measurements. The goal is
> to show that the discrepancies among measurements are
> not statistically significant. 
> My first thought was to use a 2-level model
> (measurements nested within tooth) and test the rater,
> time, position effects and maybe their interractions.
> Something like 
> xi:xtreg length i.rater i.time i.position,i(tooth)
> assuming that a random intercept structure is adequate
> for this experiment. 
> The problem is that this is the first time I deal with
> a problem where the goal is to "prove" that some
> factors have no significant effects, thus I am not
> very confident with the aforementioned method.

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index