Tufte has written several books on presenting graphical information, all
beautiful and informative. One of his pet peeves (and mine, too) is taking
up lots of white space to present a few numbers that could be in a small
table, or even in text. Many simple histograms fall into this category. He
even uses a graph by a friend of mine as an example (of bad graphing), but
the graph is one type that is commonly used in psych.
How about the silliness of presenting three dimensional histograms that make
it harder to see the relationship than if it were the old fashioned flat
ones. (But we can do it so easily!)
Donald McBurney
Stephen Black wrote:
> Just venting here, although it could well be used as an example of how
> not to present information graphically.
>
> The winner is:
>
> Flor, H. et al (1995). Phantom-limb pain as a perceptual correlate of
> cortical reorganization following arm amputation. Nature, 375, 482--
>
> They present a remarkable finding of a correlation of r = 0.93 between
> cortical reorganization and amount of phantom limb pain. I'm reviewing
> this for a lecture this afternoon. But that's not the point here.
>
> Consider their tiny Figure 2, in which the legend takes up more space
> than the graph. Picture it. A three-dimensional graph, with two
> parallel rows of mountains (26 peaks in all), each ascending higher
> and higher. Imagine how long it took me to figure out what they were
> trying to show.
>
> It's actually a scatterplot of pain against reorganization. I
> re-plotted it (with difficulty, because it's not easy to pick off the
> values on a tiny 3-dimensional plot) and ended up with a simple graph
> with r = 0.87. No mountains.
>
> Give someone a 3-dimensional graph programme and he'll find a way to
> use it. But where were the referees on this one? And in _Nature_,
> of all places.
>
> -Stephen
>
> ------------------------------------------------------------------------
> Stephen Black, Ph.D. tel: (819) 822-9600 ext 2470
> Department of Psychology fax: (819) 822-9661
> Bishop's University e-mail: [EMAIL PROTECTED]
> Lennoxville, QC
> J1M 1Z7
> Canada Department web page at http://www.ubishops.ca/ccc/div/soc/psy
> ------------------------------------------------------------------------
