Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: RE: reliability with ordinal data-Kendall's w?
From
Nick Cox <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
RE: st: RE: reliability with ordinal data-Kendall's w?
Date
Tue, 14 Dec 2010 12:01:55 +0000
I agree broadly with Ronán.
The misunderstanding that Bland and Altman invented this plot sometimes goes with the use of the name Bland-Altman plot by others.
On the question of appropriate analysis, it is often arguable whether what they recommend goes far enough in terms of modelling the data generation process.
However, jittering a scatter plot is a poor way to show (dis)agreement on a few-point ordered scale when all the frequencies (percents) can be shown directly.
Nick
[email protected]
Ronán Conroy
A scatterplot with a large jitter can also be interesting, especially were there are more than two raters.
As a matter of historical trivia, Bland and Altman didn't claim to have invented the plot, nor did they give it a name in their original paper. The paper, however, became a citation classic (I believe it is the most cited paper the Lancet ever published) and more or less single-handedly eradicated inappropriate analysis of agreement data in biomedicine, something for which we should all be grateful as we get older.
On 13 Dec 2010, at 20:12, Nick Cox wrote:
> I can't see much point to plots of the form re-invented by Bland and Altman in this case. Nor does the limits of agreement approach transfer other than queasily to a 4-point graded scale. -tabplot- from SSC offers one of various graphical alternatives. Diagonal agreement and off-diagonal disagreement will be pretty clear.
*
* 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/