TI2110

Analisis dan Visualisasi Data

Dasar Visualisasi Data
Overview
This chapter will explain how to use Tableau to
illustrate the data by implementing data
visualization.
Objectives
After completing this unit, you’ll be able to
understand the foundation of data visualization and
how to use the data to make a visualization.
Contents
• Introduction to Data Visualization
• From Data to Visualization
Introduction to Data Visualization
Introduction
• Data visualization is part art and part science.
• A data visualization first and foremost has to
accurately convey the data.
• At the same time, a data visualization should be
aesthetically pleasing.
Introduction Data Visualization
When you help people visualize the meaning of data,
you add tremendous value to any organization.
When it comes to gaining valuable insight in a
company setting, the use of data visualization is
critical. Companies are desperate to view and learn
from their Big Data. Data visualization, however, is a
growing field with a critical shortage of true experts.
Big Data refers to the voluminous amounts of
information that can be collected from social media
data as well as internal company data. Analyzing and
extracting insights from it is the goal.
Ugly, Bad, and Wrong Figures
To provide a simple visual guideline of which
examples should be emulated and which should be
avoided, we are going to labeling problematic figures
as “ugly,” “bad,” or “wrong”
• Ugly : A figure that has
aesthetic problems but
otherwise is clear and
informative
• Bad : A figure that has
problems related to
perception; it may be unclear,
confusing, overly
complicated, or deceiving
• Wrong : A figure that has
problems related to
mathematics; it is objectively
incorrect
Understanding Data Visualization
Here’s a simple definition of data visualization: It’s
the study of how to represent data by using a visual
or artistic approach rather than the traditional
reporting method.
Two of the most popular types of data visualizations
are dashboards and infographics, both of which use a
combination of charts, text, and images to
communicate the message of the data. The practice
of transforming data into meaningful and useful
information via some form of visualization or report
is called Business Intelligence (BI).
Importance of Data Visualization
Data visualizations (you can call them data viz for short)
are widely used in companies of all sizes to communicate
their data stories. This practice, known as BI, is a
multibillion-dollar industry. It continues to grow
exponentially as more companies seek ways to use their
big data to gain valuable insight into past, current, and
future events.
With the recent popularity of social media and mobile
apps, the amount of data that’s generated on a moment-
to-moment basis is astounding. For this reason, many
companies find that making sense of that data requires
the use of some form of data visualization. It’s virtually
impossible to view 1 million rows of data and try to make
sense of it.
Who Use Data Visualization
Data visualizations are for everybody. All of us use them,
whether or not we realize it.
Eg : If you use apps on your smartphone, for example,
chances are that you depend on data visualizations to make
critical decisions on an almost daily basis. Do you ever use a
weather app to determine how to dress for that day? If you
open the app and see a cloud with lightning at the top of the
app, you have a good idea that it’s going to be a stormy, rainy
day without having to read any data about temperature,
barometric pressure, and humidity.
This example shows you how a simple visual helps you gain
quick insight and make a quick decision (in this case, to wear a
raincoat and carry an umbrella). Believe it or not, you just
consumed a good data visualization.
Traits of Good Data Visualization
Good data visualizations come in all shapes and sizes,
but all of them have certain traits.
Table below shows the key items discussed during
the presentation.
Traits of a Good Data Visualization
Trait Description
People use it on a regular basis and can make relevant
Useful decisions by viewing all the information they need in one
place.
It’s not only easy to use but also pleasurable to use.
Desirable

People who use it can accomplish their goals quickly and

Usable
easily.
Point of Excellence
Before you move on to the basics of building your
data visualization, you should have some idea of
what criteria make a data visualization excellent. An
excellent data visualization has the following
qualities:
• Visual Appealing
• Scalable
• Give the Right Information
• Accessible
Data Visualization VS Infographic
To simplify the process of understanding visualizations, we focus on
the two most popular types: data visualizations and infographics.
Because the use of graphical data visualizations is growing quickly,
there is a bit of disagreement about how to define a data visualization
versus an infographic.
The following three criteria to determine whether to call a graphic a
data visualization or an infographic:
• Method of Generation
• Quantity of data represented
• Degree of Aesthetic
Data Visualizations VS Infographics
Data Visualization Infographic
Method of Generation More numbers used Original images created
Quantity of Data More data Less data, more conclusions
Degree of Aesthetic Less artful, more focused More artful
Treatment on information itself
Interactive VS Static Interactive (data changes) Static (data remains fixed)
Infographic
Infographics have gained great favor of late. If done well, they
can illuminate a problem and tell an interesting story.
Infographics have generated great interest on the Internet
because of their ability to entertain as well as enlighten.
Infographics use design rules to artfully display text, numbers,
metaphors, and other data types. You should use a few key
guidelines when you’re planning your own infographic or
evaluating one:
• Make it easy to understand
• Make it accurate
• Provide your sources
• Choose complementary colors
• Make it worthwhile
Types of Infographic
Several types of infographics are currently popular. The
following list can help you choose the right type for the
information you’re trying to illustrate:
• Case Study
• Chronology
• Comparison
• Compilation
• Expert Advice
• How-to Information
Dashboard
Dashboards allow staff to see their Key Performance
Indicators (KPIs) and important alerts on one screen. They
have become increasingly popular because of the growing
amount of data companies need to harness. Dashboards
enable companies to put measures from different
departments in one graphic. Following are some benefits of
dashboards:
• They remove the need to update manual calculations
• They focus on the measures that are most important to the
audience
• They alert stakeholders to actions that must be taken
• They increase productivity by showing the most important
data on one screen so that users don’t have to go searching
for it.
Data Visualization Design Process
Steps to start data visualization process:
• Determine Goal and Supporting Data
• Choosing Appropriate Visual Encodings
Determine Goal and Supporting Data
The goal of your visualization is going to be informed by your own
goals and motivations as well as the needs of your reader. A
visualization’s goal is usually to satisfy a need for specific information
on the part of your reader. But the goal may also be to change the
reader’s opinions or behavior in some way.
To begin defining the goal of your visualization, ask yourself, what
information need am I attempting to satisfy with this visualization?
Related questions include:
• What values or data dimensions are relevant in this context?
• Which of these dimensions matter; matter most; and matter least?
• What are the key relationships that need to be communicated?
• What properties or values would make some individual data points
more interesting than the rest?
• What actions might be taken once this information need is
satisfied, and what values will justify that action?
Choosing Appropriate Visual Encodings
1. Natural Ordering
Whether a visual property has a natural ordering is determined
by whether the mechanics of our visual system and the
“software” in our brains automatically—unintentionally— assign
an order, or ranking, to different values of that property.
For example, position has a natural ordering; shape doesn’t.
2. Distinct Values
The second main factor to consider when choosing a visual
property is how many distinct values it has that your reader will
be able to perceive, differentiate, and possibly remember. For
example, there are a lot of colors in the world, but we can’t tell
them apart if they’re too similar. We can more easily
differentiate a large number of shapes, a huge number of
positions, and an infinite number of numbers. When choosing a
visual property, select one that has a number of useful
differentiable values and an ordering similar to that of your data.
Choosing Appropriate Visual Encodings (Cont.)
• Redundant Encoding

• Defaults versus Innovative Formats

Apply Encodings Well
• Color : In the defense of color, it can be an excellent property for
labeling categorical data, or non-ordered categories for
differentiation purpose. (Examples of non-ordered categories
include operating system, gender, region, conference track, and
genre.)

• Size : Size can be used to great advantage to represent the

relative importance of entities. Consider using size to draw your
reader to central, key, or fundamentally important entities.
Apply Encodings Well
• Text and Typography : Consider which words and
numbers are present, how they interact with the
visual elements, and what they do or don’t say.
• Fonts and Hierarchies : Your goal is to make things
clear and easy to navigate for your reader. So avoid
using fancy or trendy fonts just because you can.
Stay away from gothic fonts, fantasy fonts, and
script fonts.
• Avoid Drop Shadows
From Data to Visualization
Adding Title and Caption
Adding as much information as is possible to the title
and caption that describe or summarize some of the
important points of the visualization helps readers
understand the visualization better. If you have
applied filters, Tableau will automatically describe
the filters in the caption area, but as a data designer,
you will have to describe any other important
aspects of the visualization.
Adding Title and Caption
Using file Sample Coffee Chain (Access). Open a new
worksheet.
Once the data is loaded on the worksheet, perform
the following steps to add a tittle and a caption:
• Drag and Drop Measure and/or Dimension from
Data Pane to Shelf
• From the main Menu toolbar, select Show Title
Adding Title and Caption
Once the data is loaded on the worksheet, perform
the following steps to add a tittle and a caption
(Cont.):
• From the main menu toolbar, select Show Caption
• Double-click anywhere in the area next to Title and
in the Edit Title box enter (name it based on your
chart), and hit OK
• Double-click anywhere in the area next to Caption
and in the Edit Caption box enter (write on your
caption which describe the chart)
• Move the Caption box below the Title box
Modifying Font Sizes and Colors
Tableau provides options to modify font sizes and
colors for the whole worksheet or individual
components of the worksheet, such as the pane,
headers, tooltip, and grand total. Although the
default scheme is good enough to be used in
production-quality material, there might be instances
where you would want to customize these options.
Modifying Font Sizes and Colors
To customize the font and color, make sure your
chart is ready (for our practice, continue the
visualization which created in Adding Title and
Caption).
Once you have created the graphic, perform the
following steps to customize the font and color:
• From the main menu toolbar, click on Format and
then select Font
• Make sure that the Format Font button, which has
the letter A in its icon, is highlighted
Modifying Font Sizes and Colors
Once you have created the graphic, perform the
following steps to customize the font and color
(Cont.):
• To modify all the fonts on the worksheet, select a
different font size and font from the Worksheet
dropdown
• To modify the font color of the header, select a
different font color from the Header dropdown
Apply Various Marks
Tableau provides various ways to encode data using
different marks including Square, Circle, and Shape.
The Square and Circle marks will show data points
using a square or circle shape. By using the Shape
mark, however, we can assign some attribute values
to various shapes and help the reader distinguish
data points by those shapes.
Apply Various Marks
Using file Sample Coffee Chain (Access). Open a new
worksheet.
Once the data is loaded on the worksheet, perform
the following steps to use various marks to denote
the data points:
• While holding the Ctrl key, click on Market from
Dimensions and Profit from Measures
• Click on the Show Me button to bring the Show Me
toolbar on the screen
• Click on circle views on the Show Me toolbar
Apply Various Marks
Once the data is loaded on the worksheet, perform the following steps to
use various marks to denote the data points (Cont.):
• To change the shape, from the Marks pane, select Square or Circle from
the dropdown
• To use a different shape for every Market value, select Shape from the
dropdown in the Marks pane
• Drag-and-drop Market from Color into the Shape box to generate a
chart
Apply Various Marks
Although the Square, Circle, and Shape options may
appear similar, they provide different ways to view
and distinguish various data points. That is especially
true in the case of the Shape option: as every
attribute value is given a shape, it becomes easy to
identify and note the different data points. In the
case of many data points or many categories,
however, identification and distinction of data points
even with different shapes is challenging. To assist
the readers, you should limit shapes to three to four
attribute values.
Adding Colors
Depending on the type of chart you created, Tableau
may or may not color code any data. Using the Color
box in the Marks pane, however, you could easily add
colors to your graphs.
• Using file Sample – Superstore Sales Open a new
worksheet.
Once the data is loaded on the worksheet, perform
the following steps to add colors to encode the data:
• Drag-and-drop Profit from Measures into the
Columns shelf
• Drag-and-drop Sub-Category from Dimensions into
the Rows shelf
Adding Colors
Once the data is loaded on the worksheet, perform
the following steps to add colors to encode the data
(Cont.):
• To see the Profit values by Ship Mode, drag-and-
drop Ship Mode in the Color box in the Marks pane
to generate a chart
• To see the Profit values by Customer Segment,
drag-and-drop Customer Segment in the Color box
in the Marks pane
• To compare Profit and Shipping Cost values, drag-
and-drop Shipping Cost in the Color box in the
Marks pane to generate a chart.
Adding Colors
When you drag a dimension to the Color box,
Tableau automatically selects a color palette to draw
colors from, and these colors are very distinctive and
use different hues. When you drag a measure to the
Color box, however, Tableau selects a gradient of a
single color, and bigger data points (larger values) are
encoded by a darker gradient, and smaller data
points (smaller values) are encoded by a lighter
gradient. It is important to note, however, that the
underlying value of the dimension or measure also
dictates what type of color palette would be chosen:
a dimension of a continuous type of data can
generate a gradient palette, and a discrete measure
can generate a discrete color palette.
Adding Labels
Although adding labels to data points is sometimes
redundant, Tableau makes it very easy to add labels
to your graphs. With the help of data labels, readers
of the graph are able to read the exact value of the
data point instead of speculating about the values by
gauging the heights of bars or sizes of shapes.
However, as a data designer of a visualization, you
must ask this question to yourself:
if data labels are important to your graph, can you
replace the graph with a simple table to provide all
the details?
Adding Labels
Using file Sample – Superstore Sales Open a new
worksheet.
Once the data is loaded on the worksheet, perform
the following steps to add the data labels to your
graph:
• Drag-and-drop State from Dimensions into the
Rows shelf
• Drag-and-drop Profit from Measures into the
Columns shelf
• Drag-and-drop Profit from Measures again into the
Label box in the Marks pane to display data labels
next to the bars
Changing Marks Sizes
Since we can tell differences in sizes easily compared
to differences in colors, encoding data in various
sizes of marks will increase the effectiveness of a
graph. If the differences in data points are hard to
observe, then a different type of visualization might
be needed.
Using file Sample – Superstore Sales Open a new
worksheet.
Once the data is loaded, perform the following steps
to show marks sizes encoded by the values of an
attribute:
• Drag-and-drop Sub-Category from Dimensions into
the Rows shelf
Changing Marks Sizes
Once the data is loaded, perform the following steps
to show marks sizes encoded by the values of an
attribute (Cont.):
• Drag-and-drop Profit from Measures into the
Columns shelf
• Drag-and-drop Shipping Cost from Measures into
the Size box in the Marks pane
• Change the mark type to Shape to see the mark
size vary by Shipping Cost
• Click on the Size box to increase the size of shapes
by dragging the visible slider
Adding Reference Lines
By adding reference lines, we can compare data
points with either any constant value or any
statistical computation such as average of the
measure values.
Using file Sample – Coffee Chain (Access) Open a
new worksheet.
Once the data is loaded on the worksheet, perform
the following steps to add various reference lines:
• Drag-and-drop Product Type from Dimensions into
the Columns shelf
• Drag-and-drop Profit from Measures into the
Columns shelf
Adding Reference Lines
Once the data is loaded on the worksheet, perform the following
steps to add various reference lines (Cont.):
• Drag-and-drop Market from Measures into the Rows shelf
• Change the mark type to Shape
• To compare Profit values for each Product Type and Market value
with the average Profit value for all markets and product types,
right-click on the Profit axis, select Add Reference Line, keep the
Scope option value to Per Pane, accept all the default values,
and hit OK. The graph should look like the one in the following
screenshot:
Adding Reference Lines
Once the data is loaded on the worksheet, perform the following
steps to add various reference lines (Cont.):
• To compare the Profit values to a fixed value, right-click on the
Profit axis, select Edit Reference Line, keep the Scope value at Per
Pane, under the Line selections, change the drop-down value to
Constant from Average (as shown in the following screenshot), in
the Value box enter 25.000, and hit OK:

• The chart with a constant reference line should look similar to the
one shown in the following screenshot:
Using Multiple Measures in a View
1. Using Measure Values and Measure Names in a
View
2. Combined or Shared Axis Charts
3. Creating Dual Axis Charts
Using Measure Values and Measure
Names in a View
• Measure values and measure names are Tableau
generated fields that serve as containers for more
than one measure.
• You can see Measure Names at the bottom of the
list of Dimensions and Measure Values at the
bottom of the Measures list in the data pane.
• When you create a combined axis or dual axis view,
these fields appear in the view automatically, as
does a Measure Values card that shows what fields
are included.
Using Measure Values and Measure
Names in a View (Example View)
Combined or Shared Axis Charts
Combined (or “shared”) axis charts have more than one
measure (multiple measure) on the same axis and
highlights the relationship between measures. Which is
useful for viewing multiple measures on the same scale.
Using File Sample – Coffee Chain (Access)
• Drag a Measure to Rows
• Drag a second measure on to the vertical axis, dropping
it when you see the light green double-bar icon
• Drag a dimension to Columns
• To create a view similar to the example, drag an
additional instance of Measure Names from Columns to
Color on the Marks card, and edit the colors as required
Practice : Combined Axis Chart
Using File Combined_Axis_Charts_Starter.twbx,
Create a bar chart broken down by Segment and
Category that shows Profit and Sales on the same
axis. Use your chart to compare measures within
dimensions.
Note: Rename the “Value” axis to “Dollars”
Creating Dual Axis Charts
Dual axis charts are useful for showing how to
different measures compare to each other and
quickly spot important correlations between the two.
Build a Dual Axis Charts:
• Drag a measure onto Rows or Columns.
• Drag another measure onto the view to the same
axis
• Choose Dual Axis in one of those field
• If you have two measures that have the same
measurement units but you want the scales on the
axis to be the same, right-click the axis of the
second measure you added, and choose
Synchronize Axis
Practice: Dual Axis Chart
Using File Dual_Axis_Charts_Starter.twbx, Create a
dual axis chart with synchronized axes in order to
compare sales and profit using different mark types.
Use your chart to analyze the measures over time.
Note: Edit the colors so Sales is shown as light gray
bars instead of orange
Questions and Answers