I also went to the following website and found the full text of the 1986
paper:
http://www.sghms.ac.uk/phs/staff/jmb/ba.htm
......dale glaser
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
On Behalf Of Rich Ulrich
Sent: Wednesday, January 10, 2001 7:29 AM
To: [EMAIL PROTECTED]
Subject: Re: Bland and Altman Test
On 1 Jan 2001 09:00:27 -0800, [EMAIL PROTECTED] wrote:
> Dear list members
>
> I have a reference to the Bland-Altman test.
>
> Bland J.M and Altman D.G (1986) Statistical methods for assessing
agreement
> between two methods of clinical measurement., Lancet, i, 307-10.
>
> I am not able to travel to any university at the present time (slight
health
> problems). If anyone has this article, I would greatly appreciate it if
they
> might be able to send it to me, either e-mail or ordinary mail (snail
mail).
>
Is there actually a test?
I found the "Bland-Altman method" described in Stanton Glantz's Primer
of Biostatistics (4th edition, 1997), and it looks straightforward; I
could have used the reference in the past, just to justify "doing the
obvious."
Glantz describes: for two versions of a continuous measurement, the
Pearson r depends on the variability of your sample, and it is not
the information people ought to care about. So,
a) present the mean difference, as an indicator of bias; and
b) present the standard deviation of the differences, as the measure
of precision; and
c) examine a graph of Mean (of each pair) versus Difference, as a
check for homogeneity of prediction.
If you are happy with what you see as the 2 SD range of precision,
then one rating may substitute for the other. Where you have both,
accept their average as the best estimate.
I've always used a formal test for (a), the paired t-test.
You could use a regression program to ask for tests of (c).
Glantz does not mention either. (If the original authors do, then I
would have to consider the omission a shortfall of this textbook, even
though he calls it a Primer.)
Glantz does give a second reference:
DG Altman and JM Bland, "Measurement in medicine: the analysis of
method comparison studies," Statistician 32:307-317, 1983.
Hope this helps.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================