BITS Pilani

Pilani Campus

Course No: S2-23_DSECLZG555

Data Visualization and Interpretation

 Session 2: CH 3 – 4
Data Visualization and Interpretation
(Course : SEZG555)

BITS Pilani Jaideep Butta, Guest Faculty

Pilani Campus BITS Pilani, Pilani Campus
Agenda
1
Recap Session#1
2
Choosing an effective Visual
3
Cluttering
4

Pre-attentive Attributes
5

6
Design Concepts

7 Dissecting Model Visuals

Wrap Up / Next Class - Topics

3 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Course - Module Summary
No. Module Content of Module (Contact Hours) No. Module Content of Module (Contact Hours)
Name Name
 Introduction, Exploratory vs. explanatory Merits of Matplotlib
Data •
analysis Matplotlib visuals basics
M1 Visualizations Data •
and Practices • Cluttering, Pre-attentive Attributes • Getting Started with Matplotlib
• Taxonomy of Data Visualization Methods Visualization
(CS 1-5) • Decorate Graphs with Plot Styles and Types
• Data Visualization Tools with Python
• Advanced Matplotlib
M4 –1
• Creating Visual Analytics with Tableau • Plotting data from a database, CSV file, the
(Matplotlib)
Desktop Web
Data • Connecting to Your Data (CS 12-13) • Plotting extrapolated data using curve
Visualization • Building Your Visualizations fitting
M2 • Plotting geographical data
with Tableau • Creating Calculations to Enhance Your Data
(CS 6-8) • Using Maps to Improve Insight • Discussion Visual Analysis for Everyone
• Developing an Ad Hoc Analysis • Data Preparation with Tableau Desktop
Environment Data
features
Visualization
• Dashboard-Introduction with Python • Exploring Cards , Shelves on Visual Interface
• Variations in Dashboard Uses and Data –2 • Data Exploration with Tableau Features
• Common Mistakes in Dashboard Design M5 • Exploring Filtering, Groups, Hierarchies in
Effective (Seaborn
• Power of Visual Perception detail
Dashboard and Bokeh)
• Characteristics of a Well‐Designed • Exploring Tableau Built-in Functions
M3 Design • Story-boarding data story
Dashboard (CS 14-16)
(CS 9-11) • Putting Compelling Data Visualizations into
• Effective Dashboard Display Media
• Dashboards design for Usability Persuasive Business Presentations
• Building Your First Advanced Dashboard
• Tableau Stories

4 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Announcement

• Course slides / PPT - To be uploaded at MS

Teams

• Course Handout – Will share once I receive

5 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Poll Result – From our last class

6 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Few queries from last class….

• Tableau – Any data limitation? – No known limitation related

to number of rows or columns

• Heatmap URL (Moneycontrol)-

https://www.moneycontrol.com/stocksmarketsindia/heat-
map-advance-decline-ratio-nse-bse

7 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Recap

 Data Visualization Overview (Best Practices, Use cases)

 Exploiting the Digital age (Data accessibility, Interactive nature, multiple

tools available, global collaboration)

 Visualization as a Discovery tool (Unveiling hidden insights,

Explore/Inquire, enhanced collaboration/communication)

 Exploratory vs. Explanatory analysis

 Understanding the context for data presentations

8 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

How Digital Age is transforming Data Visualization…
1. Data Accessibility and Abundance: - The digital age provides access to vast
quantities of data from diverse sources, previously unthinkable. (e.g. Social Media
platforms, Crowdsourcing platforms - Kaggle, Wearable devices)

2. Advanced Processing and Analytics: - Powerful computational tools and

algorithms can analyze data and identify patterns too subtle for human eyes.

3. Interactive and Immersive Experiences: - Interactive visualizations &

Immersive technologies like VR and AR

4. New Visualization Techniques and Tools: - Advanced software and platforms

offer intuitive interfaces and features that empower anyone to create effective
visualizations, democratizing the process. (e.g. Tableau, PowerBI, Google Data
Studio)
5. Global Collaboration and Dissemination: - Online platforms and tools allow for
real-time collaboration on data visualization projects across geographical
boundaries.

9 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Visualization as a Discovery Tool – Empowers us to…

 See beyond numbers and uncover hidden patterns and relationships (e.g.
Grocery Store Sales - Seasonality trends, Website Traffic)

 Ask better questions and generate new hypotheses (e.g. Tracking Fitness Data -
Noticing a plateau in step count despite increased exertion, you ask: "Am I overtraining?)

 Deepen our understanding of complex data and make informed decisions

(e.g. A financial analyst tracks historical stock prices, trading volume, and market news (using
time series) or A hospital analyzes records of readmitted patients by department, diagnosis,
and treatment methods (using scatter plot)

 Communicate insights effectively and inspire action (e.g. Secure project funding
from investors using data for Market analysis, projected revenue gains, and competitor
comparison. Presenting a dynamic infographic blending data charts with compelling visuals
and a clear narrative about the project's potential)

10 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Exploratory vs. Explanatory analysis

Exploratory Analysis: Explanatory Analysis:

• Goal: To explore and understand the • Goal: To confirm or refute initial
data, uncover patterns, trends, and
hypotheses identified in exploratory
outliers, and generate initial
analysis, explain underlying
hypotheses.
relationships and draw conclusions
from the data.
• Visualization: Often uses interactive
and flexible charts like scatter plots, • Visualization: Uses clear and concise
histograms, boxplots, and parallel
visualizations like bar charts, line
coordinates. These allow for dynamic
graphs, heatmaps, and pie charts.
exploration, filtering, and
These focus on presenting well-
manipulation of the data to reveal
defined insights and conveying
potential relationships and anomalies.
findings to a broader audience.
• Audience: Primarily the analyst • Audience: Can be broader, including
themselves, or a small group of
stakeholders, decision-makers, or the
collaborators involved in exploring
general public who need to
the data.
understand the meaning and
11 implications of the data.
DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
 BITS Pilani
Pilani Campus

Understanding the “Context” for data presentations

BITS Pilani, Pilani Campus
Understanding the context for data presentations

“Understanding the context for data

presentations means considering various factors
surrounding the presentation to craft the most
impactful and relevant delivery. It extends
beyond the data itself and dives into the
environment in which it will be communicated”

13 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Understanding the context for data presentations – Key Aspects
Audience:

• Who are you presenting to? What is their level of data literacy and
background knowledge? Tailoring your language, complexity, and visual
choices to their understanding is crucial.

• What are their expectations and interests? Understanding their goals and
pain points helps you focus on the data that matters most to them.

Purpose:

• What is the intended outcome of your presentation? Do you want to

inform, persuade, or inspire action? Aligning your data and visuals with
your purpose ensures a focused and impactful message.

• What kind of decision-making process is this data influencing?

Understanding the broader context of the decision empowers you to
present the data in a way that directly guides their conclusions.
14 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Understanding the context for data presentations – Key Aspects
Platform & Delivery:

• Is this a formal presentation, a casual meeting, or an online report? The

format and tone should adapt to the platform and delivery style.

• Are there any time constraints or specific presentation software

requirements? Planning your visualization choices and flow within these
limitations keeps your audience engaged.

External Factors:

• Are there any relevant current events, trends, or industry benchmarks

that your data relates to? Contextualizing your findings within a broader
narrative adds depth and relevance.

• Are there any cultural or ethical considerations in communicating your

data? Being mindful of sensitivities and biases ensures responsible and
inclusive data presentation.
15 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Understanding the context for data presentations – How it Helps

 Optimize your communication: Tailor your message

and visuals to resonate with your specific audience
and purpose.

 Increase engagement: Create a compelling narrative

that captures your audience's attention and promotes
understanding.

 Drive action: Present your data in a way that informs

decisions, solves problems, and inspires action.

16 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

What is context?
 Success in data visualization does not start with data visualization

 Attention and time should be paid to understanding the context

for the need to communicate
Who
(Who am I communicating to? –
Know your audience)

Context What
(What do I want my audience to
know or do?)

How
(How can I use data to help make
my point?)

17 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Who, What and How (Examples)
 Who (Be Specific)
 Bad Example - Students
 Good Example – University Students
 Better Example – University Students studying Data Science

 What do I want my audience to know or do?

 Action - Not always easy. Encourage discussion in case coming up with action
is tough. Suggesting possible next steps can be a great way to get the
conversation going (Examples – Start, Join, Remember, Promote etc…)
 Mechanism - How will you communicate to your audience? Amount of control,
Level of detail (Refer next slide)
 Tone - What tone do you want your communication to set? - Are you
celebrating a success? Trying to light a fire to drive action? Is the topic light
hearted or serious?

 How can I use data to help make my point?

 Data becomes supporting evidence of the story you will build and tell

18 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Today’s Agenda
 M1 - Data Visualizations and Practices

 Choosing an effective Visual

 Cluttering
 Pre-attentive Attributes
 Design concepts
 Dissecting model visuals

19 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

2 Choosing an Effective Visuals

BITS Pilani, Pilani Campus
2.1 Understand Your Goal and Audience

• Clearly define the message you want to convey.

• Identify your target audience and their level of data literacy.

• Tailor the visualization to their needs and preferences.

21 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.2 Match the Chart Type to Your Data
• Line charts: Ideal for showing trends over time (e.g., sales growth,
website traffic)
• Bar charts: Effective for comparing values between categories (e.g.,
product sales, customer demographics)
• Pie charts: Useful for showing part-to-whole relationships (e.g., market
share, budget allocation)
• Scatter plots: Reveal correlations between two variables (e.g., price vs.
demand, test scores vs. study hours)
• Maps: Visualize geographical data (e.g., sales distribution, population
density)
• Histograms: Show the distribution of data (e.g., customer ages, product
prices)
•22 Area charts: Emphasize
DSECLZG555cumulative
- Data Visualizationtotals (e.g., total sales over
and Interpretation time)
BITS Pilani, Pilani Campus
How to choose the best chart?

23 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

How to choose the best chart?

24 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Tableau – Different Chart Types

Simple Tableau hands on……

25 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.3 Keep It Simple and Clear

• Avoid clutter and visual complexity.

• Use clear labels, titles, and legends.

• Highlight key insights through color, size, or font emphasis.

26 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.4. Use Visual Cues Strategically
• Employ color to differentiate categories, highlight trends, or convey
meaning (e.g., red for negative, green for positive).

• Vary shapes and sizes to represent different data values.

• Use spatial positioning to compare elements (i.e. arranging elements within a

visual space in such a way that their relative positions help the viewer to easily compare their
values, categories, or other attributes)

27 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.5 - Consider Interactivity

• Allow users to explore data through filtering, zooming, or panning.

• Facilitate deeper understanding through interactive elements.

28 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.6 - Test and Refine

• Gather feedback from users to ensure clarity and effectiveness.

• Iterate on your visualizations to improve their impact.

29 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

2.7 - Additional Tips
• Maintain visual consistency across multiple visualizations.

• Consider accessibility for users with visual impairments (i.e. add description of
visualization, keyboard navigation, interactive features)

• Use storytelling to engage your audience and make data more

memorable.

30 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

3 Cluttering
BITS Pilani, Pilani Campus
What is Clutter?

Any guesses….?

32 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

3.1 Clutter in Data Visualization: Enemy of Clarity

• Data visualization aims to illuminate insights hidden within data, but

sometimes, instead of shedding light, visualizations can become shrouded
in clutter.

• This clutter, like an overgrown garden, obscures the data's message and
hinders effective communication.

33 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

3.2 What is Clutter?
Clutter refers to any visual element that doesn't contribute to understanding
the data or detracts from the key message. It can manifest in various forms,
• Excessive
including: chart elements:
Unnecessary gridlines,
shadows, borders, or
decorative icons.

• Too many data points:

Cramming too much
information into a single
chart, overwhelming the
34 viewer and making it difficult
DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
3.3 Impact of Clutter
• Reduced comprehension: It hinders viewers' ability to understand the data
and derive insights.

• Cognitive overload: It strains the viewer's cognitive resources, making it

difficult to process the information.

• Misinterpretation: It can lead to misinterpretations of the data due to

visual confusion.

• Loss of trust: Poorly designed visualizations can undermine the credibility

of the data and the presenter.

35 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Other types of Visual Clutter
1. Lack of Visual Order

1. Disorganized 1. Placement and formatting of elements

2. Uncomfortable to look at have been modified
3. Various components are haphazardly 2. There is order
put there without regard for the 3. components are better arranged
structure of the overall page 4. greater attention to alignment and white
space

36 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Other types of Visual Clutter
2. Alignment

1. Centre‐aligned 1. Left justified aligned

2. Diagonal elements 2. Changed to horizontal orientation

37 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Other types of Visual Clutter
3. White Space

1. Avoid stretching visuals to take up the 1. White space in visual communication is as

available space important as pauses in public speaking

2. The lack of white space is like the lack 2. Margins should remain free of text and
of pauses in a spoken presentation visuals

38 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Other types of Visual Clutter
4. Non‐strategic use of
contrast

1. Negative values that complicated the 1. Horizontal bar chart to depict the information
visualization challenge 2. Rescaled all the numbers to be on a positive
2. Categories were previously along the scale
horizontal x‐axis 3. Within each category, the length of the bar
shows the summary metric across “Our
business” (blue) and the various competitors
39 DSECLZG555 - Data Visualization and(grey)
Interpretation BITS Pilani, Pilani Campus
3.4 Combating Clutter
• Focus on simplicity: Prioritize clarity over embellishment. Use only the
essential elements needed to convey the message.

• Tame the data: Filter out irrelevant data points and avoid overloading the
chart.

• Emphasize hierarchy: Use visual cues like size, color, and placement to
guide viewers' attention towards the most important information.

• Maintain consistency: Apply consistent design principles across all charts

for a polished and professional look.

• Less is more: White space is your friend. Don't be afraid to leave room for
the data to breathe.
40 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
 BITS Pilani
Pilani Campus

Reduce Decluttering – Example (Step by Step)

BITS Pilani, Pilani Campus
Approach to reduce clutter
Your team receives tickets, or technical
issues, from employees.

You plot the monthly trend of incoming

tickets and those processed over the
past calendar year.

You see that there is some evidence

your team’s productivity is suffering
from being short‐staffed and now want
to turn the quick‐and‐dirty visual you
created into the basis for your hiring
request

42 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Approach to reduce clutter
1. Remove chart
border

2. Remove Gridline

1. Remove chart border

2. Remove gridlines
3. Remove data markers
4. Clean up axis labels
5. Label data directly
6. Leverage consistent color
43 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Approach to reduce clutter
5. Label Data
Correctly

6. Leverage
consistent colour

4. Clean up axis
labels

44 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Final Comparison

Befor After
e

Approach
1. Remove chart border 4. Clean up axis labels
2. Remove gridlines 5. Label data directly
3. Remove data markers 6. Leverage consistent colour

45 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

4 Pre-attentive Attributes
BITS Pilani, Pilani Campus
Focus our audience’s attention
 Earlier we learned about clutter and work to eliminate distractions.

 Now, we further examine how people see and how we can use that to our
advantage when crafting visuals.

 Pre-attentive attributes are visual properties that we notice without using

conscious effort to do so

 Pre-attentive processing occurs without our consciousness at an extremely high

speed. It is tuned to detect a specific set of visual attributes called Pre-attentive
Attributes.

 Pre-attentive attributes are information we can process visually almost

immediately, before sending the information to the attention processing parts of
our brain
47 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Pre-Attentive attributes…..
 These can be leveraged to help direct your audience’s attention to where we
want them to focus it.

 Second, they can be used to create a visual hierarchy* of elements to lead

your audience through the information you want to communicate in the way you
want them to process it

 Once we understand how our audience sees and processes information, we put
ourselves in a better position to be able to communicate effectively.

* pattern in the visual field wherein some elements tend to "stand out," or attract attention, more strongly than other elements,
suggesting a hierarchy of importance

48 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Pre-attentive attributes signal where to look
Let us demonstrate power of pre-attentive attributes. Taking note of how you process
the information and how long it takes, quickly count the number of 3s that appear in
the sequence

if we use pre-attentive attributes strategically, they can help us enable our audience
to see what we want them to see before they even know they’re seeing it!

49 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Various Pre-attentive attributes

Our eye is drawn to the one element within each group that is different from the rest we don’t
have to look for it. When used sparingly, pre-attentive attributes can be extremely useful for
doing two things: (1) drawing your audience’s attention quickly to where you want them to
look, and (2) creating a visual hierarchy of information

50 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Why are Pre-attentive Attributes Important?
They empower data visualization by:

• Guiding attention: We can direct viewers' focus to key insights by using

specific pre-attentive attributes. For example, using a larger bubble in a
scatter plot highlights an outlier.

• Prioritizing information: By strategically manipulating pre-attentive

attributes, we can create a visual hierarchy, emphasizing the most
important data points.

• Reducing cognitive load: Pre-attentive processing happens quickly and

efficiently, saving viewers mental resources for deeper analysis.

• Enhancing aesthetics: Pre-attentive attributes can add visual interest and

make data visualizations more engaging.

51 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

How to Leverage Pre-attentive Attributes:
They empower data visualization by:

• Size: Use larger size to represent higher values or key data points.

• Color: Apply bright or contrasting colors to highlight important elements.

• Position: Place significant data points in the center or at intersections.

• Orientation: Tilt or angle elements to draw attention.

• Shape: Use unique shapes for outliers or key categories.

• Movement: Animate charts to emphasize changes or trends.

52 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

Examples…
BITS Pilani, Pilani Campus
(Bad) Example ….

54 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Example # 1

We automatically compare the lengths. Even without looking By highlighting (Color) Delivery and Price, we have drawn
at the numbers on the axis, you know Delivery scored the our viewer’s attention to these 2 top performers. Notice how
highest. Hence length is a very good pre-attentive attribute the other 3 categories have been pushed to the background?
for quantitative information

Suddenly, our viewer’s attention shifts to the

bottom of the chart, the worst performers. We
Source -
have set the stage for a discussion on what’s
http://daydreamingnumbers.com/blog/preattentiv going wrong in these areas.
e-attributes-example/

55 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Example # 2

The chart above uses a different color for each bar and orders the The chart above uses a single color across all bars and orders the
bars alphabetically by product bars descending by sales amount.

Notice how your eyes jump back and forth between the colors in the chart
on the left. When you look at the right side chart, your eyes more easily
follow the length of the bars down and across the categories from largest
to smallest

Source - https://datasavvy.me/2017/11/30/design-concepts-for-better-power-bi-reports-part-2-preattentive-
attributes/
56 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Example # 3

Fig- Fig-
1 2

Fig – 1 : The vertical bar chart on the top has a dark black border. It’s
probably the first thing you notice about the chart.

Fig – 2 : Once we remove the border, as shown in the chart in the

middle, we notice that the chart category labels have a diagonal
orientation. They stand out because nothing else in the chart is
diagonal. It’s a bit distracting and difficult to read.

Fig – 3 : The horizontal bar chart shows the same information in the
same order, but allows the category labels to remain horizontal. Now
our eyes focus on the information encoded by the bars

Fig-
3
Source - https://datasavvy.me/2017/11/30/design-concepts-for-better-power-bi-reports-part-2-preattentive-
attributes/
57 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Tableau Example…

58 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

5 Design Concepts
BITS Pilani, Pilani Campus
5.1 Design Concepts for Captivating Data Visualization
Data visualization aims to translate complex information into clear and
engaging visuals. Beyond choosing the right chart type, effective design
concepts elevate your creations from mere charts to captivating stories.

Here are some key concepts to consider:

1. Storytelling:

• Narrative arc: Craft a narrative with a beginning, middle, and end, guiding
viewers through the data's journey.
• Focus on the why: Don't just show the what, explain the so what. Highlight
the insights and implications of the data.
• Use data as evidence: Support your arguments and claims with the power
of visualized data.

60 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

5.1 Design Concepts for Captivating Data Visualization
2. Visual Hierarchy:

• Prioritize information: Emphasize key data points using size, color, and
position.
• Create visual layers: Layer information progressively, starting with the
most important.
• Group related data: Organize data into logical groups using color, lines, or
proximity.

3. Minimalism and Clarity:

• Less is more: Avoid clutter. Focus on clean and uncluttered designs.

• White space is your friend: Let the data breathe. Empty space enhances
focus and readability.
• Consistent and cohesive: Maintain consistent font styles, colors, and
design elements across your visuals.

61 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

5.1 Design Concepts for Captivating Data Visualization
4. Engaging Color Palettes:

• Color is language: Use color deliberately to represent categories, trends,

or emotions.
• Accessibility considerations: Ensure color choices are accessible for
viewers with color vision deficiencies.
• Create contrast: Use contrasting colors to differentiate key elements.

5. Interaction and Animation:

• Make it interactive: Allow viewers to explore the data through filtering,

zooming, or hovering.
• Animate effectively: Use animation to highlight trends, transitions, or
changes over time.
• Interactive storytelling: Combine animation with narration to create a
dynamic and engaging experience.

62 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

5.1 Design Concepts for Captivating Data Visualization
6. Emotional Resonance:

• Data has a heart: Use visual metaphors, comparisons, or storytelling to

connect with viewers emotionally.
• Evoke curiosity: Spark interest and encourage exploration of the data.
• Memorable visuals: Leave a lasting impression with unique and impactful
design elements.

Additional concepts

• Typography matters: Choose clear and readable fonts that complement

your overall design.
• Data textures and patterns: Experiment with textures and patterns to add
visual interest and depth.
• Unexpected perspectives: Break the mold and present data from unique
angles or views.

63 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Design Concepts - Conclusion

“By utilizing these concepts and exploring

your creative potential, you can transform
data visualization from a technical exercise
into a compelling storytelling experience”

64 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

6 Dissecting model visuals

BITS Pilani, Pilani Campus
6.1 Dissecting Model Visuals in Data Visualization
 Everything is titled and labeled, so
there’s no question about what we
are looking at.

 Graph title, vertical axis title, and

horizontal axis title are present.

 The various lines in the graph are

labeled directly (so there’s no work
going back and forth between a
legend and the data to decipher what
is being graphed)

 Good use of text makes this visual

accessible

 Pre-attentive attributes used: “Progress to date” trend: color, thickness of line,

presence of data marker and label on the final point, and the size of the corresponding text

 Other Considerations: The goal of $50,000 is drawn on the graph for reference, and is
pushed to the background by use of a thin line (so that it doesn't compete for attention

66 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

6.1 Dissecting Model Visuals in Data Visualization
 Visual cues to draw a distinction
between the actual (Solid Line) and
forecast data (dotted line)

 Clear labeling of Actual and Forecast

under the x‐axis helps reinforce this,
with the forecast portion set apart
visually via light background shading

 In this visual, everything has been

pushed to the background with the use
of grey font and elements except the
graph title, dates within the text boxes,
data (line), select data markers, and
numeric data labels from 2014 forward

 The $108 numeric label is bold. This is

emphasized intentionally, since it is the
last point of actual data and the anchor
for the forecast

67 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

6.1 Dissecting Model Visuals in Data Visualization
 The graph title, legend, and vertical y‐axis
title are all aligned in the upper‐left‐most
position. This means audience encounters
how to read the graph before they get to
the data

 On the left‐hand side, the graph title,

legend, y‐axis title, and footnote are all
aligned, creating a clean line on the left
side of the visual

 Red is used as the single attention‐

grabbing color

 The $108 numeric label is bold. This is

emphasized intentionally, since it is the last
point of actual data and the anchor for the
forecast

68 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

6.1 Dissecting Model Visuals in Data Visualization

69 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

6.1 Dissecting Model Visuals in Data Visualization
 Lot of information, but due to
strategic emphasizing and de‐
emphasizing of components, it
does not become visually
overwhelming

 Stacked bars make sense here

given the nature of what is being
graphed

 The top three priorities are

emphasized specifically through
the use of color

 The numeric data labels within

the bars have been preserved,
but de‐emphasized with smaller
text (oriented to the left, which
creates a clean line as you scan
down the data labels for the
“Most important,”

 The x‐axis was eliminated

altogether
70 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus
Summary
• We touched on the choice of graph type and ordering of data.

• We considered where our eyes are drawn and in what order due to
strategies employed to emphasize and de‐emphasize components
through the use of color, thickness, and size.

• We discussed the alignment and positioning of elements.

• We considered the appropriate use of text that makes the visuals

accessible through clear titling, labeling, and annotation

71 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Reference

Video - 10 Steps to User-Friendly Design for

Tableau
https://www.youtube.com/watch?v=-OgFdOhXtL
A

72 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Next Class
 Taxonomy of Data Visualisation Methods

 Lessons in Storytelling
 Comparing Categories of Plots
 Dot Plot
 Bar Chart
 Floating Bar

73 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

Appendix
BITS Pilani, Pilani Campus
Cognitive load
 Cognitive load refers to the total amount of information your working memory
can handle

 When we ask a computer to do work, we are relying on the computer’s

processing power. When we ask our audience to do work, we are leveraging their
mental processing power. This is cognitive load

 Humans’ brains have a finite amount of this mental processing power. As

designers of information, we want to be smart about how we use our audience’s
brain power.

75 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Data - Ink ratio
The data-ink ratio is the proportion of Ink that is used to present actual data compared to the
total amount of ink (or pixels) used in the entire display. (Ratio of Data-Ink to non-Data-
Ink). Good graphics should include only data-Ink. Non-Data-Ink is to be deleted everywhere
where possible

76 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus

Additional Tips
• View the visual from different perspectives: Consider how different
audiences might interpret it.

• Compare it to other visualizations of similar data: Identify alternative

approaches and design choices.

• Seek feedback from others: Get different viewpoints on its effectiveness.

77 DSECLZG555 - Data Visualization and Interpretation BITS Pilani, Pilani Campus