"Our greatest responsibility is to be good ancestors."

-Jonas Salk

Sunday, January 27, 2008

How Not to Communicate Quantitatively

OK, UNEP doesn't want me to livelink to their map pages without their permission, and I'm going to have to be scrupulous about such things. But I think this partial clipping is fair use in an act of criticism.

The page on which this image is found is a spectacularly bad presentation because it titles both the article and the graphic with the unfortunate text "Temperature increases 2001-2005". If this were an accurate label this would be a map of truly terrifying proportions. In fact I am relieved to report that it is merely (one hopes) a very severe and disap
pointing error, as the caption below the map, in small print and somewhat opaque language essentially acknowledges, as follows:

Increases in annual temperatures for a recent five-year period relative to 1951–1980. Warming is widespread, generally greater over land than over oceans, and... (etc.)

Will every reader notice that "relative to 1951-1980" Will everyone who notices it be able to confidently explain what it means? Does, say, your aunt understand "five year period relative to three decade period?"

The title is misleadingly wrong. It should be captioned "temperature changes since mid-century" or something like that. Sowing such confusion is hardly a good idea.

ohmmmm, yeah, that URL? http://maps.grida.no/go/graphic/increases-in-temperature-2001-2005

(rolls eyes)

If anyone has any information on whom to contact at UNEP to correct this problem I would be much obliged.

Update: I submitted this in their feedback form:
The title of the page, the legend of the map and the URL are all seriously misleading in the same way. One could easily be misled into thinking that the image shows the increase in temperature over a five year period. The caption makes this explicit, but not so clearly as to outweigh the potential damage done by the title, legend and URL. A much better title would be "2001-2005 temperatures compared to 1951-1980". I strongly urge you to repair the misleading title in all its manifestations.
Feel free to join in. The feedback link is right on the map page.

Update: It's fixed! See comments.

1 comment:

Unknown said...

Hi Michael, my name is Hugo Ahlenius, and I am the map/web-master of the UNEP/GRID-Arendal maps and graphics library, as well as the cartographer behind that map. Thanks for the note, the title and legend have been revised to make this clearer and to reduce confusion, please check back to http://maps.grida.no/go/graphic/increases-in-annual-temperatures-for-a-recent-five-year-period-relative-to-1951-1980 (note that the URL is changed to reflect this as well).

This map was originally prepared for the UNEP Global Outlook for Ice and Snow, a report launched in July 2007, as supporting material for an overview chapter on feedback effects and the cryosphere (authored by James E. Overland of NOAA and others) and in that report the title is part of the caption, which makes it a bit low key. The original map is from the James Hansen et al article Global temperature change published in PNAS 2006, and you can see it on the PNAS website.