Information Graphics: A Comprehensive Illustrated Reference : Visual Tools for Analyzing, Managing, and Communicating

450 Pages·1996·39.03 MB·English

by  Robert L. Harris

Information
Graphic s
A Comprehensive
Illustrated Reference
Visual Tools for Analyzing, Managing, and Communicating
Robert L. Harris
Robert L. Harris
-Z.-
C-
Information Graphics
A Comprehensive Illustrated Reference
Visual Tools for Analyzing, Managing, and Communicating
Management Graphics * Atlanta, Georgia * U.S.A.
Information Graphics
A Comprehensive Illustrated Reference
Written and illustrated by Robert L. Harris
Published by Management Graphics
P.O. Box 78581
Atlanta, Georgia 30357-2581
U.S.A.
Telephone [phone]
FAX [phone]
Copy editor Faye Goolrick
Acknowledgements Acknowledgement and thanks are extended to Glenda Jo Fox Hughes, C. Dwight
Tabor, Jr., and Merwyn L. Elliott for reviewing portions of the material in this book.
Copyright Copyright © 1996 by Robert L. Harris
All rights reserved. No part of this book may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopying,
recording or by any information storage and retrieval system without written
permission from the publisher, except for the inclusion of brief quotations in a
review.
Printed in the United States of America
Second printing
Cataloging information Harris, Robert L.
Information graphics: a comprehensive illustrated reference / Robert L. Harris
448 pages: illustrated; page height 28 centimeters
Bibliography: pages 445-448
1. Graphic methods-Encyclopedias I. Title.
2. Charts, diagrams, etc.-Encyclopedias
3. Computer graphics
Library of Congress classification number (LCCN) ....... QA90.H37 1996
Dewey decimal classification number ............. 001.4'226-dc2O
Library of Congress Catalog Card Number (LC) ...... 95-77855
International Standard Book Number (ISBN) ........ 0-9646925-0-3
Notices This publication is designed to provide accurate, illustrative, and authoritative
information in regard to the subject matter covered. It is sold with the
understanding that neither the author nor the publisher is engaged in rendering
legal, accounting, investment, or other professional services. If expert
assistance is required, the services of a competent professional person should
be sought.
Neither the author nor the publisher assumes any responsibility for the uses
made of the material in this book or for decisions based on its use and makes
no warranties regarding the contents of this product or its fitness for any
particular purpose.
All data used in the examples in this book are fictitious. Numeric values in the
examples may or may not be similar to those occurring in actual applications.
Major focus of book
This book addresses charts, graphs, maps, diagrams, and tables used in all areas;
however, its major focus is on their uses for operational purposes.
To many people, information graphics are the images frequently used in presentations at
formal meetings or the stylized charts and graphs used in newspapers and magazines.
Many are used for these purposes; however, for every chart, graph, map, diagram, or
table used in a presentation or publication, there are thousands that are utilized for what
are called operational purposes. Information graphics for operational purposes are used
by millions of people on a daily basis for such things as improving their efficiency and
effectiveness, improving quality, solving problems, planning, teaching, training,
monitoring processes, studying the geographic distribution of data, looking for trends and
relationships, reviewing the status of projects, developing ideas, writing reports,
analyzing census data, studying sales results, and tracking home finances. With the need
to cope with increased amounts of data and at the same time increase quality and
productivity, charts, graphs, and maps are being used more and more in operational
situations. Fortunately, as a result of developments in computer equipment and software,
most of the popular charts and graphs used on a daily basis can be generated rapidly,
easily, and with little or no special training.
This single volume is the first book to focus on the needs of the millions of individuals
who currently use or would like to use information graphics for their operational needs. It
is of particular value to those who might generate their charts and graphs on computers.
The book brings together the necessary information on charts, graphs, maps, diagrams,
and tables so individuals can rapidly discover the wide range of information graphics that
is available, how they are being used, how to construct them, and how to interpret them.
Organization and major contents
The contents of this book are organized for ease of use. Towards this goal:
- Entries are alphabetized using the letter by letter system in which spaces, hyphens,
commas, etc., between words in the major headings are ignored.
- Although the major headings of Chart, Graph, Map, Diagram, and Table are each listed in
their proper alphabetical location, the individual graphics and features that make up these
major categories also have their own headings in the master alphabetical listing.
- Specific applications are described and illustrated to relate the theoretical to the practical.
- Entries are accompanied by one or more graphic examples (frequently annotated) that
complement the written descriptions.
- Terminology applicable to specific graphics are explained in the text, shown on the
example, or both.
- General terminology such as Variable, Fill, Legend, Matrix, Polygon, Plane, Coordinate,
etc., are discussed under their individual headings.
- For the convenience of the reader, some information is repeated under multiple headings.
The following additional features have been incorporated.
_. uV UacLrKe Y CIlluojuli-n s. tLaryr^pauQb oUfI L ouwU.UaQ5Q5--rwc.iacfrausi.Loniunlirisaor, (cid:1)(cid:1) 01
When an information graphic has a single name but might be classified
3W W
several different ways, the major write-up is included under one of the
headings and cross-referenced under the others. For example, the main
write-up for polar graph (right) is under Polar Graph but it is referenced
Polar graph
under Graph, Circular Graph, and Point Graph.
20o 1
When an information graphic is commonly referred to by several different 210-VS
Sometimes referred to as:
names, the major write-up is shown under only one of the headings. The other
Pie chart
names are included at their proper alphabetical locations along with a short
Cake chart
description, an example, and a reference as to where the major write-up is Circle diagram
Circle graph
listed. For example, various users refer to the chart at the right by nine Circular percentage chart
different names. The major description is included under Pie Chart, with only Divide circle
Sectogram
thumbnail descriptions under the other eight headings. Sector chart
Segmented chart
- In cases where the same name is used to describe entirely different information graphics, each situation is handled on an
individual basis. An example is shown below in which each of the graphics is sometimes referred to as a bar chart. In this
particular case, Ul infuormation graphics adcre 20 -t
different enough that each is given its own 5 c State A
write-up. Each is cross-referenced B -tate
appropriately, although all are not referenced State C
to one another. Superscripts are assigned in A State D
cases WHVIC Ut lu VAMSLe aateeaImxaie i IS Ume UUC -'91 92 `93 ''94 ''9 0 20 40 as s0 to '92 '93 94 95
most frequently used for multiple graphics. Column graph Bar graph Bar chart' Bar chart2
- When a topic is related to an entry or it would be beneficial for the reader to be aware of a related topic, the related topic is
noted in the write-up.
IVwfhfliflsf tfsl tiujO U..u .a.h.n.ta JLmA ssiuL *sEsJu sRFJ S. jSJ.
With so many different information graphics used in such diverse
applications, it is sometimes helpful to group them into families or
categories. For example, it is useful to know which graphs are used to study
data distribution or to look for correlations. It is helpful to know which
graphs present percent-of-the-whole data most efficiently or what types of
graphs are used to determine probabilities. In the area of maps, it is useful to
know that most maps that are used as charts fall into six major categories, the
four major ones being statistical, descriptive, flow, and topographic. The
sections that discuss the various groupings of information graphics are in
addition to the sections that discuss the specifics of the various graphics.
AaK- o. v %-rza-3n.l .ca^tu:r-sL n tinn Uru IUAJI U-H1av^t:uo|9ilc"AM
Most information graphics software programs have construction details designed 4 4 4 - - 4-
into them. For instance, the initial decisions are made largely by the software 3 3 3 3
manufacturer regarding line size, tick marks, grid lines, scales, type of text, etc. 2- 2 2- 2- -
Many programs give the operator the option of changing these, and as people 1- 11 1- 1- -
become more proficient they often generate their own unique graphics. If the
0 1 2
graphic is being made by hand, all the construction decisions must be made by 0E xter1n al to2 Intemal to Ac0r oss 1t he 2 Co0m bin1a tion2s
the person generating the information graphic. Sections of this book are devoted tthhee bgroadpyh o f tohf eth beo dy ainxtiesr n(haal lfa nd oaft iothnes vsahrlo-wn
graph half external) at the left
to a discussion of each of the major construction features. Examples of tick mark locations
4
Organization and major contents
(continued)
- '7' I
-lichen "-Ulr-'-: ll-l_ _ I-tvPae-t-lU_t_ll_lr eoV c-a 4-tt h. a-t _m_l1-iR-g ;1h1 t mlclP58-0I t&[-hliIe R- lvViAue VwW Ve1r
Certain methods of presenting data have been found to frequently mislead the
viewer. Many times these methods are used because the person making the graphic
is unaware of the hazard or does not know an alternative. These misleading design
Features such
features, such as broken scales and perspective views, are discussed under their k as perspective
individual headings such as Scales and Perspective Projection, as well as under views that might
mislead the viewer
certain specific types of graphs and maps. are pointed out.
The most up-to-date developments in information graphics
70 -
Many advances have been made in the area of information graphics, both as a
result of creative individuals such as W.S. Cleveland, E.R. Tufte, and J.W. s4oi0l Fl U . Ai j X Framed
Tukey as well as many excellent software developers. In some cases an rectangles
entirely new information graphic has been invented, such as the box graph. In for
20- encoding
other cases it might be a component, such as a framed rectangle symbol, or a quantitative
information
concept, such as the data-ink ratio. Because previously there has been no
into maps
vehicle to bring these developments to the attention of the vast majority of Grouped box graph for displaying
the distribution of sets of data
users, many of the new designs and techniques are largely underutilized.
Examples of some of the most recent
developments in information graphics
Information graphics available as a result of new software
There are a number of information graphics that have been around for
many years, but because there was no efficient way to generate them,
they have not been widely used. With the development of powerful
desktop computer software, these graphical tools are now
economically available to anyone interested. These charts are
discussed in the context of all of the other charts with no special
category assigned to them. Three examples are shown here.
Time --
Candlestick chart used for
recording the price of stocks,
PERT chart used for planning and tracking major programs commodities, etc.
Information graphics used in many different fields
Pareto chart for analyzing
I quality situations
.15
.5
In some cases information graphics developed in one field can be
City A
directly applied in other fields. In other cases, a slight modification EL city 1
might make the graphic useful, or in still other cases, a specific
information graphic may not work but the idea of how the chart Tyipeo r cause of reject City D 41
elements are used might trigger a completely new chart design. One City E 2 1
of the purposes for including application-specific information .d 31 15
h
graphics is to serve as a catalyst in the transfer of graphic ideas from Brnd A Brd B B.,d C Br.d D Bra16 One-half
one field to another. bidirectional
Ctailteyfo rdata
N in which row and
Brnd F B,.nd B.,d H Br-d I Bra:M J column headings are
Icon comparison display useidt o the same
compare three or more
characteristics for multiple eentities
Interrelationships of complex information graphics
In most cases a brief explanation plus an example is all that is
required for readers to understand how a chart or graph is
constructed and functions. In a few cases it is not obvious how
a particular graph or map is generated or how two or more
graphs or maps relate. In these cases a more detailed i4.9s
explanation is sometimes given, as shown at the right. Taking
the time to study these more detailed explanations is not
necessary for an understanding of the basic graphs or maps.
Such explanations can be skipped without detracting from the Illustration of hc
main content of the section. graphs In a dra -
relate to a th
(al
5
6 Information Graphics: A Comprehensive Illustrated Reference
.Abscissa The distance a data point is located from zero along the horizontal axis of a graph.
-Abscissa Axis A technical name for the horizontal or X-axis on a two-dimensional graph.
Abstract Graph Sometimes referred to as a diagram. An abstract Break-even graph I
Profit -
graph typically has no scales, tick marks, or grid
lines and is used primarily to investigate, Break even
illustrate, and convey ideas and concepts rather 012a pont HIVolume that Vcoasrtisable
than specific values. A break-even graph, shown 1IO must be sold
i to make a
at right, is one of the better-known examples of profit
an abstract graph. This graph illustrates some of Loss/ cFixed
the basic concepts associated with sales, costs, / I
and profits without using numeric values. Units sold (volume)
Example of an abstract graph
Abstract Map A map that conveys a focused but limited
amount of specific information such as
location, direction, relationships, time, or
information by area with little concern for
the presence or accuracy of information
outside its focus. For instance, in the
example at the right, the distances on the
map show equal driving times between
locations. Most of the actual ground
distances are distorted. Abstract maps as
generally convey one or at most a few sets
An abstract map indicating points of equal
of facts, ideas, concepts, etc. Many distorted
driving time from a central location, taking
maps, strip maps, cartograms, and into consideration type of roads, terrain,
congestion, etc. Most of the actual ground
diagrammatic maps qualify as abstract maps.
distances are distorted.
75 * -
70. 74
65. 09
Age and Sex Pyramid Also referred to as a population pyramid. A
65.6
specific application of a pyramid graph or two- 4,50.54
464 0.4
way histogram for studying a population by age
36- 3.
s0.34
and sex. Age intervals are plotted on the vertical 25.29
2 .24
axis and the number of males and females in each 15.19
age interval on the horizontal axis. See 11.15
Population Pyramid. 10 8 6 4 2 0 2 4 6 8 10
Number of females Number of males
Age and sex pyramid
Alignment Graph Sometimes referred to as a nomograph, nomogram, or calculation 101 K20 10
1118 -9
graph. An alignment graph is designed to solve an equation -V7 - ' - 164 -78
involving three or more variables. Such graphs consist of three or 6- -. -6
more scales arranged so that a straight line crossing all the scales
4 8 -
3 6 3
intersects the scales at values that satisfy the equation. The graph at 2 - 4 2
1 2 1
the right satisfies the equation A + B = C (e.g., 8 on scale A plus 5 0 A c0 BL
on scale B equals 13 on scale C). See Nomograph. Alignment graph
Alignment of Text A term used to describe how text is formatted. There are five major types of alignment: the
four shown below plus alignment on the decimal point. See Text.
Aligned flush left Aligned flush right Centered Justified
Sometimes referred to as Sometimes referred to as Sometimes referred to as Sometimes referred to as
justified left and/or ragged justified right and/or ragged right and left. flush right and left. Text
right. When text is aligned ragged left. When text is Uniform spacings are that is justified normally
left, the left edges of all aligned rght, the right used between words and has the right and left
lines are even. The right edges of all lines are even letters. Excess space is edges aligned. This is
edge of each line is The lef edge of each line distributed equally at the accomplished by varying
determined by determined determined by the full two ends of each line. the space sizes between
words, partialords,,parti aords, partial words, and Generally used only with words and letters and the
spacincings int le l. spacings int he line a few lines of text. frequent use of hyphens.
Amount Scale Sometimes referred to as a quantitative, numeric, value, or
interval scale. An amount scale consists of numbers organized 0 1 2 3 4 5
in an ordered sequence with meaningful and uniform spacing EExample of a linear amount
Scale. Other types of amount
between them. Quantitative variables are typically measured Sscales include logarithmic,
along amount scales. See Scale. Pprobability, and power.
7
Analytical Chart/Graph The term refers to the way a chart or graph is used rather than a specific format. As the
name implies, analytical charts are used to analyze information rather than for such
purposes as reference, monitoring, presenting, entertainment, or advertising. An analytical
chart might be a simple scatter graph or a complex conceptual diagram. Such charts are
often helpful when studying information for relationships, similarities, correlations,
patterns, trends, etc.
Analytical Table Sometimes referred to as a statistical or Number of students with given combination of scores
summary table. This type of table ISu sed T10o0ta=l 3 8 13 22 21 20 | 14 10 4 31 1 1119I
primarily for analyzing information, as 80%-- 2 4 1 1 1
opposed to referencing, scheduling, 70 = 3 4-3 4 1 = 16
860% _ _1 4 6 6 4 2 1 22_
advertising, etc. There are always two or 0% _ 1 6 6 4 2 1 - -i2- 0
more variables and the data in the body are 240 4 2 23 21 19
normally numeric. Information is generally
2 1 -3 232 112
110% 1 12
arranged, ranked, or sorted to make 0% 2 - . 4 -% % % To=l-
relationships, trends, comparisons, Score on test B
distributions, or anomalies stand out. Example of an analytical table
Ancillary Map Sometimes referred to as an inset or supplementary
map. If viewers are not familiar with the area shown
on a detailed map, a small ancillary map is often used
to help orient them. For example, the larger map at
right shows a hypothetical area in northern Georgia. If
shown only the detailed map, most people would not
know what state it was located in, whether it was in
the northern, southern, eastern, or western portion of
the state, and what was close to it that they might
recognize. An ancillary map, shown in the upper
ngnt-nand corner uI Uie ind11 Inlp, Suppies dnswC[s Ancillary map used in conjunction
to most of these questions. with a larger map
Angular (Polar) Scale Angular scales are used with circular
graphs. They typically are arranged with
equal distances and equal numbers of
degrees between the major tick marks
F
(major values). The number of degrees is
normally an even fraction of 360 degrees,
such as 1/24 (150), 1/12 (300), 1/6 (600),
or 1/4 (900). The labels might be shown as
degrees or radians. See Polar Graph. 90m
Example of an angular or polar scale
Area Band Graph Sometimes referred to as a band, range, or silhouette graph. See Range Symbols and Graphs.
Area Bar Graph In an area bar graph, the widths of the bars are proportional to some measure or character-
100/ , istic of the data element(s) represented by the bars. For
example, if the bars are Total . es
91 80%- displaying the profitability equals
y sales
0 of various product lines, the L4250
width of the bars might
indicate what percentage of 200 oduct A
E 40%-
0 the total sales the various r 150
0
products represent. Bar
CL2 0%- oduct B
widths can be displayed D 10c
Mo.I alnnor the vertical axis in
% 5%1 6% 15% 20% %- -5 a0----- --- oduct C
Percent profitability terms of percents (left) or
Simple area bar graph showing units (right). The concept product D
meiop roi_I_d_u- llity o1rw eA--a-{ u+n:I np Arlu-n ucl c^ an be uHsee d we ith a si-mpl-e O 09% 20% 40% 60% 80% 100%
on the horizontal axis with the
width of the bars proportional bar graph (left) or a stacked Percent domestic, export,
to the percent the product line bar graph (right). See Bar and Intercompany sales
represents of total sales. Graph. Unit scale on vertical axis
8
Area Chart Sometimes referred to as a proportional area chart. A variation of a proportional chart used
for communicating differences in size, quantity, value, etc., by drawing data graphics in the
same proportions as the things they represent. For example, if the sales of A are twice as
big as the sales of B, the area of the data graphic for A would be twice as big as the area of
the data graphic for B. Area
charts may be used to display
actual values (left) or
I
percents-of-the-whole (right).
Pie charts are the most
.. :v.--1- . ..- : I-
The area of The area of circle A is widely usea vacation ror Column Pie chart
column A is twice twice that of circle B presenting percent-of-the- chart
that of column B Two variations of area
Two variations of area charts showing whole data. Area charts can charts used to present
the relative size or value of two things also be used to compare the percent-of-the-whole data
-_m thinn-a t -illirli n-n ntsi m- * QRrin__ _ i*
3a111% tiliiiia aLt I1UILIjFI.. Ji111 III Llilt. kLI611tJ. - OLIIlA.. IL la
difficult to accurately estimate values using area charts,
the actual values are generally shown on or adjacent to
)
the data graphics. * The areas of data graphics, which can
be any shape (right and below), are not meant to convey
exact data but simply to give the viewer some visual 1990
indication of the Area charts used to depict changes
relative sizes of the in the value of something, such as
items they represent. coffee consumption, over time.
Area charts are used almost exclusively for commun-
Aication
28 i purposes. They are seldom used for analytical
Irregularly shaped area charts purposes. See Proportional Chart and Pie Chart.
Area Column Graph A variation of column graphs in which the width of the columns have significance. In the
area column graph, the widths of the columns are proportional to some measure or
characteristic of the data elements represented by the columns. For example, if the columns
are displaying the profitability of various product lines, the width of the columns might
25 . I indicate what percent- -,,oov/
age of total sales the X8
20 various products Eo%
al 15 represent. Column d 60%
widths can be display- = 40%
& 10
ed in terms of percents
5 (left) or units (right). . 20%
There may be a scale OE 0%
0
on the horizonta-l ax-is ~Units5 0 100 s1o5ld0 200 I250
percent or total saues or a scale may be Total units sold of A,B , C, &D equals 250
Simple area column graph with a
percent scale on horizontal axis. provided in a legend. Stacked area column graph with a
In still other cases, the unit scale on the horizontal axis
values might be noted directly on the graph. If the values along the horizontal axis are
cumulative, the columns are generally joined so there are no unaccounted-for spaces. For
a) example, if all of the columns add up to a certain
fin - value, such. as .10H0- %, -or if t...h e sum of- a.l. l of- t.he
| e values along the horizontal axis is important, no
" R UM
i 240 L <a Hn I| 11 I1- 1I.o.-.l..1 . .l. . m usfpeaecds b ebtew%twereee el enf ttt hhheep twcclop lPnunmm -rnnn -sc ioiffmt htnhese. vS-aknlwul rePse saq9cr-ae- n-nn ohtp
o 5 10 cumulative, the total is unimportant, and a scale is
a2: Cu Legend Cmltvtettli nmotnadasaei
part of the legend (example at left). Sometimes
0(aU
or the widths and as
it
N I u A r| oup t. uroup U heights of the
Z Group B columns are 2
An area column graph with spaces between D
columns. Int his type of graph the values related such that c1o
along the horizontal axis are not cumulative, as one varies, the v,/
the total is unimportant, and the scale is other varies also. U
bi l-uOm i I 1n1 I ac g II V-Y--anl -y1U5UA).
For example, in a
histogram, if the width of a column is increased to
2
encompass a broader class interval, the height of the column
Class interval
is adjusted accordingly (example at right). In these cases the The histogram is probably the
area of the column is a truer measure of the thing being most widely used application
of an area column graph. In a
represented than either the height or width. The concept of
histogram, as the width of a
using the width of columns to convey additional information column is expanded to cover a
is generally applied only to simple and stacked column broader class interval, the
height is adjusted accordingly.
graphs, including 100% stacked graphs. See Column Graph.
9