An introduction of chart design principles in IT 7113 Data Visualization at Kennesaw State University - updated in 2024.
About Jack Zheng
Faculty of IT at Kennesaw.edu
http://idi.kennesaw.edu/it7113/
Chart Design
General Principles and Best Practices
IT 7113 Data Visualization
J.G. Zheng
Fall 2024
http://idi.Kennesaw.edu/it7113/
Content Overview
This lecture notes discuss some chart design
principles and selected best practices.
• “CASE” design principles
• Example best practices and patterns aligned
with the principles
• Develop and conform to guidelines and
standards
2
https://www.linkedin.com/pulse/data-design-six-must-know-visualization-principles-everyone-eppler/
https://kevinlanning.github.io/DataSciLibArts/principles-of-data-visualization.html#tufte-first-principles
https://medium.com/google-design/redefining-data-visualization-at-google-9bdcf2e447c6
Principles and Best Practices
• Principles are the high-level guidelines that may apply to a broader scope of applications.
– Principles are usually summarized in highly concise terms or sentences
• Best practices are the lessons and design references at a very detailed level for specific cases
and/or specific design elements (such as color, layout, slicers, etc.).
– Best practices extend principles to more specific cases, and they should be related to principles.
– Contains more details and contexts, may be limited in scope.
– Often it is embodied in the form of “patterns”.
• Guidelines and standards are working rules set up for an environment (an organization or an
industry)
•
In practice, people use the terms loosely and generally refer to good practices and/or pitfalls.
• Examples of principles/best practices
– https://www.linkedin.com/pulse/data-design-six-must-know-visualization-principles-everyone-eppler/
– https://kevinlanning.github.io/DataSciLibArts/principles-of-data-visualization.html#tufte-first-principles
– https://medium.com/google-design/redefining-data-visualization-at-google-9bdcf2e447c6
3
Basic Design Principles
• The CASE principles
4
Clarity
The chart delivers the message and makes
the point clearly.
Accuracy
Avoid data visual distortion and
disinformation.
Simplicity
Perceptually easy to locate and identify key
data and other information.
Elegance
Visual quality to attract audience and sustain
that sentiment and interest – Andy Kirk.
https://visme.co/blog/dos-and-donts-chart-making/
https://www.storytellingwithdata.com/blog/2012/10/my-penchant-for-horizontal-bar-graphs
Clarity
• Clarity means being clear and straightforward, causing no confusion, and with sound logic
• Clear message serving the purpose
– Right chart for the purpose.
– The chart delivers the message and makes the point, straightforward and clearly.
– Self explanatory: use proper labels and annotations: title, legend, label, etc.
– Avoid unnecessary implication
• Consistency
– Message consistent with chart: scale consistency, color consistency, unit consistency
• Relevancy
– The visualized data are relevant to the purpose and message.
– Show context that helps understand the message.
• Readability
– Ease to read and find information and data in the visualization.
5
Image from https://visme.co/blog/dos-and-
donts-chart-making/
Case from
https://www.storytellingwithdata.com/blog/201
2/10/my-penchant-for-horizontal-bar-graphs
https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
Clear and Unconfusing Message
• Don’t go against intuition and convention
• Avoid potential unintended implications
6
The convention is
values on the Y axis
goes up, not down
The use of the bright
green implies unnecessary
attention – unless it is
intended.
Image from https://filwd.substack.com/p/clarity-
and-aesthetics-in-data-visualization
https://trinachi.github.io/data-design-builds/ch14.html
https://www.qlik.com/blog/data-visualization-foundations-color-by
https://visme.co/blog/dos-and-donts-chart-making/
Readability
• Clear and easy to see the main
objects, entities and data.
• Non-cluterred: control the number
of items displayed at once.
– Limit data items in certain chart
types like pie chart and data serials
column chart and line chart
– Related to simplicity – simplicity
improves readability
• Highlight the important
• Avoid distracting backgrounds and
colors.
• Readability of data labels and their
associations to data points
– https://trinachi.github.io/data-
design-builds/ch14.html
7
https://www.qlik.com/blog/data-
visualization-foundations-color-by
https://visme.co/blog/dos-and-donts-chart-making/
Not clear. Low readability
of the other lines.
https://www.linkedin.com/pulse/edward-tuftes-six-principles-graphical-integrity-radhika-raghu/
https://viz.wtf/
https://www.reddit.com/r/dataisugly/
https://courses.cs.washington.edu/courses/cse412/21sp/lectures/CSE412-EthicalDeceptive-MichaelCorrell.pdf
https://www.linkedin.com/pulse/edward-tuftes-six-principles-graphical-integrity-radhika-raghu/
Accuracy
•
Accuracy is the correct representation (coding) of data, so
it matches people’s perception.
–
Inaccurate visualizations impact human perceptions and
decisions. Resulting in false perception and incorrect
conclusions
• Hard mistakes: value coded wrong
•
Soft mistakes: gives wrong perception because of user’s visual
limitations or different perspectives
– Relates to visual or graphical integrity
https://www.linkedin.com/pulse/edward-tuftes-six-principles-
graphical-integrity-radhika-raghu/
•
Types
– Disproportion
– Distortion
– Narrative / visual mismatch, or out of context, irrelevant
–
Inconsistency
•
Cause of Inaccuracy
– Design mistakes. Examples:
• Use circle size to represent values: value coded as area or
diameter?
•
Truncated axis
•
3D decoration (pie): view perspectives affect size perception.
– Data reference/calculation mistakes
•
See more collections of bad chart examples
–
https://viz.wtf
–
https://www.reddit.com/r/dataisugly/
8
Ethical and Deceptive Visualization
Michael Correll
https://courses.cs.washington.edu/courses/cse412/21sp/lect
ures/CSE412-EthicalDeceptive-MichaelCorrell.pdf
https://www.linkedin.com/p
ulse/edward-tuftes-six-
principles-graphical-
integrity-radhika-raghu/
Accurate
proportion
Inaccurate
proportion
https://visme.co/blog/dos-and-donts-chart-making/
https://viz.wtf/post/649279158970646528/chilean-president-uses-disproportionate-bar-chart
Disproportion
• Avoid axis truncation, as it gives
wrong impression of
exaggerated differences.
– An exception to always starting
an axis at zero is found in the
case of certain line graphs.
When the data tends to vary
minimally at a quantity far above
zero, it is difficult to read. In this
situation, starting a baseline at a
quantity closer to the data brings
the variations to light.
https://visme.co/blog/dos-and-donts-chart-making/
https://viz.wtf/post/649
279158970646528/chi
lean-president-uses-
disproportionate-bar-
chart
Axis
truncation
leads to
disproportion
https://peltiertech.com/extra-distortion-in-a-pie-chart/
https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
http://www.labo.mathieurella.fr/?p=403
https://www.engadget.com/2008-01-15-live-from-macworld-2008-steve-jobs-keynote.html
Distortion
• Distortion is the
change of shape or
size
• Maybe because of
viewing perspective or
inappropriate
proportion
• Examples
– Disproportionate X/Y
axis
– 3D Pie chart distortion
https://peltiertech.com/
extra-distortion-in-a-
pie-chart/
10
https://filwd.substack.com/p/clarity-
and-aesthetics-in-data-visualization
Jobs’s smart phone chart
http://www.labo.mathieurella.fr/?p=403
Photo from https://www.engadget.com/2008-01-15-live-
from-macworld-2008-steve-jobs-keynote.html
Is the green part
bigger than the
purple part?
Same data and
graph; but this one
seems smoother in
changes.
https://www.climate.gov/news-features/understanding-climate/climate-change-global-temperature
Units Scaling
• Axis scaling also affects perception
11
https://www.climate.gov/news-features/understanding-
climate/climate-change-global-temperature
The “zoom-out” effect
minimizes the differences.
The “zoom-in” effect
maximizes the differences.
Either chart may have a specific message to
deliver, with different interpretations.
Pure Mistake
• ? Hard to explain
– Sometimes it may be
the programming
error
12
Same height for
different values?
https://sports.yahoo.com/olympics/tokyo-2021/medals/
https://dark-star-161610.appspot.com/secured/_book/design-and-integrity.html#ethical-principles
Intentional vs. Unintentional
• Intentional distortion or disinformation: “tempt to use
visual displays that bend the truth to benefit one’s
own interest”
– Bias
– Deception
• Unintentional: out of expectation
– Data source mistake
– Design mistake
– Algorithm (mapping) mistake
13
https://sports.yahoo.com/olympics/tokyo-2021/medals/
(the bars were corrected later)
29 bar is longer than 34 bar. The size (length)
of the part represents the weight of the total
(%) instead of the absolute number. The
inconsistency causes the confusion and wrong
impression if numbers are ignored.
Extended reading
https://dark-star-161610.appspot.com/secured/_book/design-
and-integrity.html#ethical-principles
https://infovis-wiki.net/wiki/Data-Ink_Ratio
Simplicity
• Simplicity means using the simple design to
achieve the outcome
– Start with simple; do not change if there is no
problem.
– Simple does not mean minimalism. It is a
balance between simplicity and functionality. It
should not impact the clarity of the visual.
• Best practices
– Declutter
– Data ink ratio
https://infovis-wiki.net/wiki/Data-Ink_Ratio
14
https://www.darkhorseanalytics.com/blog/too-many-bars
https://www.fluencetech.com/post/lessons-from-ibcs-eliminating-chart-clutter
Declutter
• Use simpler color systems
• Control the number of
objects/elements or data
point
• Use lighter visual marks –
column vs dot/line
15
https://www.darkhorseanalytics.com/blog/too-many-bars
declutter
https://www.fluencetech.com/post/lessons-from-ibcs-
eliminating-chart-clutter
Simple color
https://www.fluencetech.com/post/lessons-from-ibcs-eliminating-chart-clutter
Declutter – Case Study
• https://www.fluencetech.com/pos
t/lessons-from-ibcs-eliminating-
chart-clutter
• Techniques applied to improving
the upper chart:
– The chart clutter which has been
removed or toned down
includes:
– Remove the 3D effect – it never
(ever!) helps…
– Grey backfill on the chart area
– Borders on the columns
– Border on the legend
– Heavy gridlines have been
subdued
– Vertical axis bar
– Axis now displayed in thousands
16
https://infovis-wiki.net/wiki/Data-Ink_Ratio
https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
https://classes.engr.oregonstate.edu/eecs/winter2018/cs519-400/modules/4-data-visualization/1-excellence-integrity/
Data Ink Ratio
• Data ink
–
It basic refers to the salience of
objects and visual properties that
directs represents data, e.g, bars,
lines, dots, etc. (vs. non-data ink like
grid lines, border, background, etc.)
– https://infovis-wiki.net/wiki/Data-
Ink_Ratio
• Some practices
– Be ware of flashy decorative effects
such as textured background,
shadow, 3D, etc., unless they are
absolutely needed or associated
with a particular meaning.
– Avoid heavy (dark and/or bold) grid
lines and borders
17
https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
https://classes.engr.oregonstate.edu/eec
s/winter2018/cs519-400/modules/4-data-
visualization/1-excellence-integrity/
https://www.darkhorseanalytics.com/blog/data-looks-better-naked/
https://blog.hubspot.com/marketing/data-graph-design-powerpoint-tips-ht
Data Ink Case Study
• Case on simplifying charts
– https://www.darkhorseanalytics.com/blog/data-
looks-better-naked/
• A similar case with step-by-step explanation
– https://blog.hubspot.com/marketing/data-graph-
design-powerpoint-tips-ht
18
https://www.darkhorseanalytics.com/blog/too-many-bars
https://nightingaledvs.com/what-makes-a-data-visualisation-elegant/
https://creativemarket.com/blog/10-basic-elements-of-design
Elegance
• Elegance is a visual quality to attract audience and sustain that
sentiment and interest – Andy Kirk
– This means visual appealing and some level of beauty and creativity
– Should not interfere with other three principles
• Example best-practices
– Using predefined color sets instead of random color combinations
– Using themes (or consistent styles)
– Associate color/texture/shape/icon with a particular meaning to boost
understanding and emotion
– Using round corner bars/columns
– Choosing proper font family
– Reasonably vary the use of chart types (which serves the same purpose)
https://www.darkhorseanalytics.com/blog/too-many-bars
• This class does not focus on this principle; you may refer to other
resources like
– Andy Kirk’s book “Data Visualization” https://nightingaledvs.com/what-
makes-a-data-visualisation-elegant/
– https://creativemarket.com/blog/10-basic-elements-of-design
19
https://www.datarevelations.com/balancing-accu-ement-and-tone/
http://www.perceptualedge.com/blog/?p=2121
http://www.thefunctionalart.com/2015/09/stephen-few-asked-me-what-i-thoughts.html
Elegance Example
•
Balancing Accuracy, Engagement,
and Tone
–
https://www.datarevelations.com/bal
ancing-accu-ement-and-tone/
•
Original post by Steven Few
–
http://www.perceptualedge.com/blog
/?p=2121
•
Alberto Cairo’s reply
–
“Tufte’s ‘data-ink-ratio’ rule leads to
charts that lack elegance.”
–
And his redesign
http://www.thefunctionalart.com/201
5/09/stephen-few-asked-me-what-i-
thoughts.html
•
Stephen Few’s comments upon
seeing the redesign:
–
“Alberto, You’re the man! I love your
improvements to the graphic. You
described your version as middle
ground between my position and
that of the embellishers, but I don’t
see it that way. I’m an advocate of
the kinds of embellishments that you
added to the graphic for journalistic
purposes, for they don’t detract from
the information in any way. I’ve
always said that journalistic
infographics can be both informative
and beautiful without compromising
either. ...”
https://excelcharts.com/data-visualization-elegant-not-beautiful/
https://www.datarevelations.com/balancing-accu-ement-and-tone/
https://www.darkhorseanalytics.com/blog/data-looks-better-naked/
https://blog.hubspot.com/marketing/data-graph-design-powerpoint-tips-ht
https://www.storytellingwithdata.com/blog/2021/1/10/lets-improve-this-graph-yt9xj
Chart Redesign Cases
• https://excelcharts.com/data-visualization-
elegant-not-beautiful/
• https://www.datarevelations.com/balancing-
accu-ement-and-tone/
• https://www.darkhorseanalytics.com/blog/data-
looks-better-naked/
• https://blog.hubspot.com/marketing/data-graph-
design-powerpoint-tips-ht
• https://www.storytellingwithdata.com/blog/2021/
1/10/lets-improve-this-graph-yt9xj
21
https://www.ibcs.com/standards/
https://xdgov.github.io/data-design-standards/
https://www.eea.europa.eu/data-and-maps/daviz/learn-more/chart-dos-and-donts
https://material.io/design/communication/data-visualization.html
Guideline and Standard
• Establish a general guideline or standard that everyone can follow
– Guidelines and standards embodies principles and best practices, and provides clear
expectations
– Promotes consistency across stakeholders, projects, and units.
– Facilitates communication and understanding
• Many organizations realize the importance of data visualization and set up
guidelines or standards.
–
It is difficult and to make and require standards at a larger scale in the visualization and
design industry.
– Many of them are more like guidelines, even called standards.
– A bit different from standards from engineering, food processing, or manufacturing.
– They are also in various forms with varied complexity and details.
• Exemplar guidelines/standards
–
IBCS https://www.ibcs.com/standards/ is a comprehensive guidebook
– US Census Bureau xd.gov https://xdgov.github.io/data-design-standards/ focuses more on
individual chart type and chart component
– European Environment Agency (eea.europa.eu) has a set of usability guidelines for
improving visualisations https://www.eea.europa.eu/data-and-maps/daviz/learn-more/chart-
dos-and-donts - these are more like a simple list (used to be a 25-point list; recently
organized into categories)
– Google data visualization guidelines https://material.io/design/communication/data-
visualization.html
22
https://www.ibcs.com/ibcs-standards-1-2/
https://ibcs.konveio.site/ibcs-standards-12
https://www.youtube.com/watch?v=VkCyNOioUQQ
https://www.ibcs.com/resource/rolf-hichert-about-ibcs/
https://ibcs.konveio.site/ibcs-standards-12#page=81
https://ibcs.konveio.site/ibcs-standards-12#page=82
https://ibcs.konveio.site/ibcs-standards-12#page=127
https://ibcs.konveio.site/ibcs-standards-12#page=120
https://ibcs.konveio.site/ibcs-standards-12#page=149
IBCS
•
International business communication standard (ibcs.com) defines a set of
standards for business communication, including the visual communication part.
– The International Business Communication Standards (IBCS®) are practical proposals for
the design of reports, presentations, dashboards and the diagrams and tables contained
therein.
• Current version 1.2
– https://www.ibcs.com/ibcs-standards-1-2/
– https://ibcs.konveio.site/ibcs-standards-12
•
IBCS explained
– Promotional short video: https://www.youtube.com/watch?v=VkCyNOioUQQ
– https://www.ibcs.com/resource/rolf-hichert-about-ibcs/
•
The “Perceptual Rules” section define some standard chart types and their
usage (https://ibcs.konveio.site/ibcs-standards-12#page=81)
– “The perceptual rules from the EXPRESS (choose proper visualization), SIMPLIFY (avoid
clutter), CONDENSE (increase information density), and CHECK (ensure visual integrity)
sections help to clearly relay content by using an appropriate visual design.”
– Express 1.1 chart types (page 68-86) https://ibcs.konveio.site/ibcs-standards-12#page=82
– Simplify (page 113-119) https://ibcs.konveio.site/ibcs-standards-12#page=127
– Condense https://ibcs.konveio.site/ibcs-standards-12#page=120
– Check (page 135 to 142) https://ibcs.konveio.site/ibcs-standards-12#page=149
23
CASE compared to IBCS
• C Express,
• A Integrity
• S Simplify
• E?
• Condense?
24
https://www.ibcs.com/resource_category/chart-templates/
https://www.ibcs.com/software/
https://www.truechart.com/software/truechart4powerbi/
https://zebrabi.com/ibcs/
https://appsource.microsoft.com/en-us/product/power-bi-visuals/WA200002681
https://appsource.microsoft.com/en-us/product/office/WA200004249
https://public.tableau.com/app/profile/ceterisag/viz/ibcs/IBCS
IBCS Templates and Tools
• Chart templates
– https://www.ibcs.com/resource_category/chart-templates/
• Some BI tools have adopted IBC
– https://www.ibcs.com/software/
• Power BI - Charts through third party addons
– https://www.truechart.com/software/truechart4powerbi/
– https://zebrabi.com/ibcs/
– https://appsource.microsoft.com/en-us/product/power-bi-
visuals/WA200002681
– https://appsource.microsoft.com/en-
us/product/office/WA200004249
• Tableau has not adopted IBCS completely but can be
customized to comply the standard
– https://public.tableau.com/app/profile/ceterisag/viz/ibcs/IBCS
25
Specific Best Practices
• This course does not focus on design details
for each type of charts.
• We discuss some selected best practices
through selected special topics in module 3
to 5.
• Use references like the standards mentioned
earlier.
• Please do some research yourself and apply
based on your own study and experience.
26
https://www.linkedin.com/pulse/data-design-six-must-know-visualization-principles-everyone-eppler/
https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
https://www.linkedin.com/pulse/edward-tuftes-six-principles-graphical-integrity-radhika-raghu/
https://blog.hubspot.com/marketing/data-graph-design-powerpoint-tips-ht
https://excelcharts.com/data-visualization-elegant-not-beautiful/
https://www.datarevelations.com/balancing-accu-ement-and-tone/
https://ibcs.konveio.site/ibcs-standards-12#page=81
https://ibcs.konveio.site/ibcs-standards-12#page=82
https://ibcs.konveio.site/ibcs-standards-12#page=127
https://ibcs.konveio.site/ibcs-standards-12#page=149
Key Readings
• Example principles
https://www.linkedin.com/pulse/data-design-six-must-know-visualization-
principles-everyone-eppler/
• Clarity: https://filwd.substack.com/p/clarity-and-aesthetics-in-data-visualization
• Accuracy/integrity:
https://www.linkedin.com/pulse/edward-tuftes-six-principles-graphical-integrity-
radhika-raghu/
• Simplicity:
https://blog.hubspot.com/marketing/data-graph-design-powerpoint-tips-ht
• Elegance
– https://excelcharts.com/data-visualization-elegant-not-beautiful/
– https://www.datarevelations.com/balancing-accu-ement-and-tone/
•
IBCS Standards version 1.2 Perceptual Rules Section
https://ibcs.konveio.site/ibcs-standards-12#page=81 – particularly the following
sections. Other sections optional.
– Express (page 68-86) https://ibcs.konveio.site/ibcs-standards-12#page=82
– Simplify (page 113-119) https://ibcs.konveio.site/ibcs-standards-12#page=127
– Check (page 135 to 142) https://ibcs.konveio.site/ibcs-standards-12#page=149
27
https://dark-star-161610.appspot.com/secured/_book/design-and-integrity.html#ethical-principles
https://nightingaledvs.com/what-makes-a-data-visualisation-elegant/
https://visme.co/blog/dos-and-donts-chart-making/
https://www.datarevelations.com/accurate-vs-emotional-comparisons-sometimes-pies-bubbles-and-waffles-are-the-better-choice.html
https://www.darkhorseanalytics.com/blog/too-many-bars
https://trinachi.github.io/data-design-builds/ch14.html
https://www.geckoboard.com/best-practice/data-visualization-tips/
https://www.youtube.com/c/IBCS-Institute/featured
https://xdgov.github.io/data-design-standards/
https://www.eea.europa.eu/data-and-maps/daviz/learn-more/chart-dos-and-donts
https://material.io/design/communication/data-visualization.html
https://kevinlanning.github.io/DataSciLibArts/principles-of-data-visualization.html#tufte-first-principles
https://medium.com/google-design/redefining-data-visualization-at-google-9bdcf2e447c6
https://viz.wtf/
https://twitter.com/WTFViz
https://www.reddit.com/r/dataisugly/
Additional Good Resources
•
https://dark-star-161610.appspot.com/secured/_book/design-and-integrity.html#ethical-principles
• What Makes a Data Visualisation Elegant? https://nightingaledvs.com/what-makes-a-data-visualisation-
elegant/
•
Selected chart design best practices
–
https://visme.co/blog/dos-and-donts-chart-making/
– Accurate vs. Emotional Comparisons – Sometimes Pies, Bubbles, and Waffles are the Better Choice
https://www.datarevelations.com/accurate-vs-emotional-comparisons-sometimes-pies-bubbles-and-waffles-are-the-better-
choice.html
–
Too many bars: https://www.darkhorseanalytics.com/blog/too-many-bars
–
https://trinachi.github.io/data-design-builds/ch14.html
–
https://www.geckoboard.com/best-practice/data-visualization-tips/
•
https://www.youtube.com/c/IBCS-Institute/featured
• Other principles/guidelines/standards
– US Census Bureau xd.gov https://xdgov.github.io/data-design-standards/ focuses more on individual chart type and chart
component
– European Environment Agency (eea.europa.eu) has a set of usability guidelines for improving visualisations
https://www.eea.europa.eu/data-and-maps/daviz/learn-more/chart-dos-and-donts - these are more like a simple list (used to
be a 25-point list; recently organized into categories)
– Google data visualization guidelines https://material.io/design/communication/data-visualization.html
–
https://kevinlanning.github.io/DataSciLibArts/principles-of-data-visualization.html#tufte-first-principles
–
https://medium.com/google-design/redefining-data-visualization-at-google-9bdcf2e447c6
•
Problematic chart collection
– Visualizations that make no sense https://viz.wtf and https://twitter.com/WTFViz
–
https://www.reddit.com/r/dataisugly/
28