Graphics Reference
In-Depth Information
ticular examples. It would be invidious, and perhaps unfair, to single out egregious
examples here. Readers should be able to find plenty for themselves without having
to look far.
The Media and Graphics
2.2.3
Graphical displays of data appear in the press very oten. hey are a subset of info-
graphics, graphical displaysforconveying information ofany kind,usually discussed
underthe heading information visualization (Spence,
).Many impressive exam-
plescanbefoundintheNewYorkTimes.Whilethereareguidelineswhichapply
to all infographics, this chapter is restricted to the construction of data visualiza-
tions.
Data displays in the media are used to present summary information, such as the
results of political surveys (what proportion of the people support which party), the
developmentoffinancial measuresovertime(acountry's tradebalance orstockmar-
ket indices) or comparisons between population groups (average education levels of
different sections of the community). here are many other examples. hese topics
only require fairly basic displays, so it is not surprising that in the media they are
commonly embellished with all manner of decoration and ornamentation, some-
times effectively drawing attention both to the graphic and to its subject, sometimes
just making it more di
cult to interpret the information being presented. What is
surprising is that the graphics are oten misleading or flawed.
Presentation
(What to Whom, How and Why)
2.3
Howis it possible tomake amess of presenting simple statistical information? Surely
there is little that can go wrong. It is astonishing just what distortion can be intro-
duced: misleading scales maybe employed;
-Ddisplaysof
-Ddata make it di
cult
to make fair comparisons; areas which are apparently intended to be proportional
to values are not; so much information is crammed into a small space that noth-
ing can be distinguished. While these are some of the technical problems that can
arise, there are additional semantic ones. A graphic may be linked to three pieces of
text: its caption, a headline and an article it accompanies. Ideally, all three should be
consistent and complement each other. In extreme cases all four can tell a different
story! Astatistician cannot domuch about headlines (possibly added oramended by
a subeditor at the last minute) or about accompanying articles if he or she is not the
firstauthor(inthepressthejournalist choosesthegraphicandmayhavelittle timeto
find something appropriate), but the caption and the graphic itself should be “good”.
Some displays in the media highlight a news item or provide an illustration to
lighten the text. hese are oten prepared by independent companies at short notice
and sold to the media as finished products. Fitting the graphic to its context may be
