Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Forest Plot


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Forest Plot
Date   Wed, 19 Mar 2008 15:21:01 -0000

Historically in England a forest was a place to hunt, meaning attempt to find and slaughter innocent wild animals, except that only a privileged few, possibly only royalty, had that right. Those who have read or watched versions of the Robin Hood legend will have picked up some of that history. 

So, a forest was not necessarily "forested" in the now standard sense of being mostly 
or entirely covered with trees. It often was, but that was a matter of fact rather than of definition. 

But now the main sense of "forest" in English (British if you will) is indeed a wooded area. However, "forest" and "wood" (meaning, wooded area) are not synonyms: a wood could be quite small (say tens of metres on the side) but a forest could not be. 

Wood as material (timber or lumber) is indeed not the sense meant here. 

I've been involved in forestry applications from time to time. In talking to foresters about statistics I've sometimes found it necessary to explain very carefully what I mean by log or root transformation, given the scope for misunderstanding. 

Nick
n.j.cox@durham.ac.uk 

Steichen, Thomas J.

Two peoples separated by a common language...

The American version of "seeing the wood for the trees" is generally "can't see the forest for the trees."  The interpretation being that the one big thing (the forest) is obscured by the many small things (the trees).

The British form (as per Nick and Lewis & Clark, 2001) almost turns that on its ear... seeing the wood (i.e., the interior of the tree in Americanese) rather than the tree itself.

Now, had Nick said "woods" (with an "s", which would reasonably equate to forest in Americanese) then the simile works better... the plot helps us see the forest despite the trees!

Nick Cox

At the risk of reminding many what they know already, but this is an international list:

It seems that the term 'forest plot' was used informally for some years before
it reached print. Often there is some selling line about seeing the wood for the trees.
That is, at least for native English speakers and perhaps for many others, a conscious play on a proverbial criticism: not being able to see the wood for the trees is missing a broad pattern through undue focus on the details. I guess every other language has some equivalent if not identical saying.

That's standard, I think. But Richard Peto, no less, is credited with a joke attribution of the plot to a cancer researcher called Pat Forrest. Whether for that reason, or for others, or just as a matter of lousy spelling, the spellings "Forest", "forrest" and "Forrest" are probably all to be found somewhere.

My main authority here, after a quick Google, is

Lewis, S. and Clarke, M. 2001.
Forest plots: trying to see the wood and the trees.
British Medical Journal 322: 1479–1480. doi:10.1136/bmj.322.7300.1479. 

Many if not all list members will be able to read that directly on the net.

Nick
n.j.cox@durham.ac.uk

Newson, Roger B

I think

findit forest plot

is the correct spelling. A forest plot is so called because the
bottom-line summary confidence interval is like a forest, and the
individual study confidence intervals are like the individual trees. It
is not named after a "Professor Forrest".


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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