[R) Bland Altman plot (was: [R] paste ?)

Murray Jorgensen maj at waikato.ac.nz
Wed Apr 26 07:45:22 CEST 2000

The idea of comparing the variances of two variables X and Y by looking at
the correlation of 

	X + Y   with  X - Y
seems to go back to Pitman (1939) A note on normal correlation, Biometrika,
31, pp 9-12.

I havn't seen the paper myself but I presume that the essentials are
summarized in Cox & Hinkley (1974) pp 140, 141. These days, of course, we
are keener on plots than formal tests.

I discovered all this around 1977 when I was interested in the case where X
and Y were measurements of quantities like %fat in cheese obtained by two
different methods or by two different labs.

Murray Jorgensen

At 10:35 PM 25-04-00 +0200, Peter Dalgaard BSA wrote:
>"Christophe Declercq" <cdeclercq at nordnet.fr> writes:
>> > Secondly, I'm curious about the history of this kind of plot.
>> > I've only heard it called a "Tukey mean difference" plot, (and
>> > Trellis graphics has a function, tmd(), that does it, but no one
>> > knows about it...).  Does anyone know who invented it?
>> Martin Bland and Doug Altman introduced it in their very successful 1986
>> paper in The Lancet (which has become a 'Citation Classic') :
>However, as so many other CTs, I don't think they actually invented
>the procedure... Plotting residuals vs. fitted values to avoid
>correlation, and warnings against the fallacy of plotting a difference
>against one of the terms have certainly been around long before '86.
>The Lancet paper is not really an introduction of a new method, rather
>it is a polemic paper against the widespread practise of publishing
>correlation coefficients as measures of agreement between two methods. 
>[Of course, since people in the medical profession treats Lancet
>papers as The Law, and vocal statisticians like Altman as demigods,
>one occasionally meets people trying to use the Bland-Altman plot for
>comparing measurements on completely different scales!]
>> Bland JM, Altman DG. Statistical methods for assessing agreement between
>> two methods of clinical measurement. Lancet 1986; i: 307-310.
>   O__  ---- Peter Dalgaard             Blegdamsvej 3  
>  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
> (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
>~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
>r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
>Send "info", "help", or "[un]subscribe"
>(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
Dr Murray Jorgensen       http://www.stats.waikato.ac.nz/Staff/maj.html 
Department of Statistics, University of Waikato, Hamilton, New Zealand 
*Applications Editor, Australian and New Zealand Journal of Statistics* 
maj at waikato.ac.nz Phone +64-7 838 4773 home phone 856 6705 Fax 838 4155

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list