Introduction

to Data
Visuzaliation

A fundamental skill of any Data Analyst

01 02
INTRODUCTION METHODOLOGY
What is data visualization? And
Basic method for data visualization
why we need it?

03 04
FUNDAMENTAL CHARTS USE CASES
Introduction to common plots Data viz in real life
 01.
Introduction
What is data visualization? And
why we need it?
The role of visualization systems is to
provide visual representations of
datasets that help people carry out
tasks more effectively.

–Tamara Munzner
-SOMEONE FAMOUS
A Visualization should
1 2 3 4

Include only Encodes

Have a CLEAR
Save TIME the RELEVANT DATA/INFO
PURPOSE
CONTENT appropriately
 Why we need
visualization?

The role of visualisation systems is

to provide visual representations
of datasets that help people carry
out tasks more effectively.
 Easier

With a visualization we want to

reduce the difficulty of a
processing data task into “sense”.
Better communication

Graphics in general are better communication method for

demonstrating terminologies such as statistical summaries.
 Anscombe’s
Quartet

If we only read the statistics from the table

above, we should conclude 4 datasets are
similar. However, the 4 sets of data are different.
 02.
Methodology
Method to visualize data
The process of data visualization THINKING

The process of data visualization includes 4 main steps:

- Understanding the requirements (KYC);
- Understanding the nature of data;
- Choosing appropriate representation; and
- Encode information optimally.
O
b
j
e Insight Error
c
t
Every visualization aims to Throughout the process of ”seeing”
i illustrate a single objective and decoding the information
and provide sufficient within the visualization, people can
v information for the audience translate it in a wrong way
e
Whoa!!!

What is thissss???
Basic components of a visualization

01 02 03

SPACE AXES GRAPHICS

The space where we The fixed reference lines The geometry
visualize data help us position our representations of the data
visualizations in question
Graphical representation of data

Shape Size Color

The shape used to The area within the The hue and saturation we
demonstrate the data defined shape. used to illustrate data.
(round, rectangle,…)

Position Length Angle

The position of data points The length of the line The angle between 2 lines.
associating with axes. within visualization
Method to transform data into chart
01 03
Identify the
Adjustment
space 02 04
Map the data Finalizing
Modify the initial
Define the space and setup (color, length,
axes needed to shape, size,…) to
Choose appropriate optimize the
represent data Add supplemental
properties of
visualization information into the
graphical
chart (title, label, axis
representation to
name,…)
illustrate the data
Example 1
Chart X
What is X? 20%
How many properties illustrated
within X?
What type of space the chart is put in?
What is the chart axis? 50%
15%
Salary Venue
Transport COGS 15%
03.
Fundamentall
visualization
 Bar charts
Bar charts contain columns (or
bars) that represent the
quantitative scale of a
continuous/discrete variable (e.g.
sales number, discount
percentage, number of new
customers,…)
Properties of bar chart

OBJECTIVE DATA TYPE

Used for comparing the quantitative Quantitative metrics/variables measured
difference; OR ranking the data by the height of the columns/bars

AXES ADDITIONAL INFO

The scale can be We can add more properties to bar
quantitative or qualitative chart, usually color, in order to add more
information
Example 2
BAR CHART
The bar chart below illustrates the sales of a candy company throughout the year:
- X axis: the months of the year (qualitative).
- Y axis: the sales number (quantitative - continuous).
- Graphic: bars that represent sales number associate with Y axis.
- Color: no meaning.
Variations of bar chart

Colorful bar chart. Dot chart. Horizontal bar chart.

Stacked bar chart. Group bar chart. NOT a bar chart.

 Line charts
A line chart is a graphical
representation of an quantitative
measurement that connects a
series of data points by a
continuous line.
Properties of line chart

OBJECTIVE DATA TYPE

Used to show the fluctuation Qquantitative metrics/variables
throughout the time OR compare the measured by the difference from data
fluctuation of different groups. points to X-axis.

AXES ADDITIONAL INFO

In common, X-axis shows the time while Line charts can show more information
Y-axis measures quantitative variable. by using different data point shape, size
and color.
Example 3
LINE CHART
The bar chart below illustrates the sales of a candy company throughout the year:
- X axis: the months of the year (qualitative).
- Y axis: the sales number (quantitative - continuous).
- Graphic: data points connected by a continuous line.
- Color: no meaning.
Why don’t we just use the bar chart in example 2?
Variations of line chart

Make color. Smooth line chart.

Variations of line chart

Different measurements in the same chart Different data point SHAPE & LINE FORMAT.
 Pie charts
A pie chart is a circular
statistical graphic, which is
divided into slices to
illustrate numerical
proportion. In a pie chart,
the arc length of each slice,
is proportional to the
quantity it represents.
Properties of pie chart

OBJECTIVE DATA TYPE

Illustrate the contribution of each group Quantitative metrics/variables measured
into the whole AND compare the size by the angle of the pie.
between groups.

AXES ADDITIONAL INFO

Only 1 axis for measuring the angle. Hard to add more information to the pie.
Variations of pie chart

Smaller pie chart explain the “Others” slice of bigger one. Multiple layers pie chart.
 Area charts
An area chart or area graph
displays graphically
quantitative data. It is based
on the line chart. The area
between axis and line are
commonly emphasized with
colors, textures and
hatchings.
Properties of area chart

OBJECTIVE DATA TYPE

Show the difference in size between Quantitative metrics/variables measured
groups OR the contribution of each by the difference between the lines and
group into the whole. Y-axis.

AXES ADDITIONAL INFO

X-axis usually shows the time while Y- Hard to add more information.
axis measures quantitative variable.
Variations of area chart

Các đường chồng lên nhau. Phần diện tích nằm đè lên nhau. Phần diện tích cộng lại thành 100%.
 Scatter plots
A scatter plot is a type of plot
or mathematical diagram
using Cartesian coordinates
to display values for typically
two variables for a set of
data. If the points are coded,
one additional variable can
be displayed.
Các tính chất của biểu đồ phân tán

OBJECTIVE DATA TYPE

Used to evaluate the relationship AND The data points plotted into the chart
the correlation between 2 variables that may or may not represent any variable.
represented in 2 axes.

AXES ADDITIONAL INFO

Both X and Y axis are quantitative scales. There are various ways to add more
information into scatter plot using
shape, size, color, etc.
Example 3

SCATTER PLOT
Scatter plot shows the relationship between
the list price and the cost:
- X-axis: indicates the standard cost of the
product (numerical – continuous).
- Y-axis: measures the list price of the
product (numerical – continuous).
- Graphical representation: each data point
represents a product.
- Additional information: none.
Variations of scatter plot

Add color code for an additional info. Add color and shape for 2 additional info.
Chart type selection guide

http://datavizcatalogue.com/blog/chart-selection-guide/
04.
Use Cases
 GIAO TIẾP PHÂN TÍCH KIỂM SOÁT
Biểu đồ và hình ảnh giúp Cũng nhờ thế, việc vẽ Biểu đồ cũng giúp chúng ta
việc tiếp nhận thông tin dễ biểu đồ trong quá theo dõi được tình hình
dàng hơn và vì thế tạo điều trình phân tích dữ liệu kinh doanh một cách dễ
kiện thuân lợi trong việc cũng giúp đẩy nhanh dàng và nhanh chóng hơn.
giao tiếp. tiến độ.
Keep track on
Cost

In business, the act of follow

and benchmark key
performance indicators such as
Sales, Cost, Orders, etc is
critical to the coporation.
Therefore, there is a need of
visualizing those key metrics
onto a dashboard and closely
monitoring them.
 Data Analysis

There are statistical graphs that help

a Data Analyst better understand
the dataset in interest. Moreover,
the insights provided by those
statistical charts also used in
inference statistics and thus leads to
conclusions and recommendations.
Communicate with business partners

In fact, there are many statistical terms that

is hard to explain by pure text. In this case, a
visualization can help the Data Analyst to
easily communicate with other stakeholders.
Q&A