Graphics Reference
In-Depth Information
hese comments relate to presentation graphics, that is, graphics drawn for the
purpose of illustrating and explaining results. Computing advances have benefitted
exploratory graphics, that is, graphics drawn to support exploring data, far more.
Notjust the quality of graphic representation hasimproved but also the quantity. Itis
nowtrivial todrawmany differentdisplaysofthe same dataortori
ethrough many
different versions interactively to look for information in data. hese capabilities are
only gradually becoming appreciated and capitalized on.
he importance of sotware availability and popularity in determining what anal-
yses are carried out and how they are presented will be an interesting research topic
forfuturehistoriansofscience.Inthebusinessworld,nooneseemstobeableto
do without the spreadsheet Excel. If Excel does not offer a particular graphic form,
then that form will not be used. (In fact Excel offers many graphic forms, though
not all that a statistician would want.) Many scientists, who only rarely need access
to computational power, also rely on Excel and its options. In the world of statistics
itself, the packages SAS and SPSS were long dominant. In the last
years, first S and
S-plus and now R have emerged as important competitors. None of these packages
currently provide effective interactive tools forexploratory graphics, though they are
all movingslowlyinthat direction aswellasextending therange andflexibility of the
presentation graphics they offer.
Data visualization is a new term. It expresses the idea that it involves more than
just representing data in a graphical form (instead of using a table). he information
behind the data should also be revealed in a good display; the graphic should aid
readers or viewers in seeing the structure in the data. he term data visualization is
related to the new field of information visualization. his includes visualization of
all kinds of information, not just of data, and is closely associated with research by
computer scientists. Up till now the work in this area has tended to concentrate just
onpresenting information, rather than onwhatmaybededucedfromit.Statisticians
tendtobeconcerned morewithvariability andtoemphasizethestatistical properties
ofresults.hecloserlinkingofgraphicswithstatisticalmodellingcanmakethismore
explicit and is a promising research direction that is facilitated by the flexible nature
of current computing sotware. Statisticians have an important role to play here.
The Chapters
1.2
Needless to say, each Handbook chapter uses a lot of graphic displays. Figure
.
is
a scatterplot of the number of figures against the number of pages. here is an ap-
proximate linear relationship with a couple of papers having somewhat more figures
per page and one somewhat less. he scales have been chosen to maximize the data-
ink ratio. An alternative version with equal scales makes clearer that the number of
figures per page is almost always less than one.
he Handbook has been divided into three sections: Principles, Methodology,
and Applications. Needless to say, the sections overlap. Figure
.
is a binary matrix
visualization using Jaccard coe
cients for both chapters (rows) and index entries
