What is data visualization? Definition, Importance and Chart Types
You must have seen a lot of data visualization examples. But what do you mean by data visualization? A graph? A chart? Or a dashboard? Today, I collected everything you want to know about data visualization, including the definition, importance, basic types. Also, I will introduce the top 16 types of chart in data visualization and analyze their application scenarios to help you quickly select the type of chart. In the end, you will have an idea of how to design good data visualization.
- What is Data Visualization?
- Why is Data Visualization Important?
- Five data Visualization Types Commonly Used
- Top 16 Chart Types in Data Visualization
- How to Design Good Data Visualization?
1. What is Data Visualization?
In short, data visualization is presenting structured or unstructured data graphically to present information hidden in the data directly to people.
But there is a catch:
It is not merely using data visualization tools to turn data into graphs. Instead, it is looking at the world from a data point of view. In other words, the object of data visualization is data, and what we really want is using data as a tool, using visualization as a means to explore the world.
2. Why is Data Visualization Important?
Let’s start with a game.
The largest importance of data visualization is that it helps people understand data faster. Finding connections between mountains of information aren’t easy, but graphs and charts can transform the invisible information into the visible graph symbol, express it directly and clearly, help you discover key points quickly.
That’s not all…..
Other importance of data visualization
Researches show that people remember around 80 percent of things they see but only 20 percent of what they have read. And the brain can remember images a million times faster than abstract words.
Therefore, visualizing data can deepen people’s memory of information.
The ability to display big data is another importance of data visualization. For instance, the dashboards built by FineReport can integrate big data from different resources, reflect real-time data, and display it on the large screen. Therefore, people can build connections between big data from different departments and monitor business performance. It opens up new avenues for business. You might find something unexpected in big data through visualizing data.
3. Five Data Visualization Types Commonly Used
3.1. Data visualization by area & size
Differentiate the length, height or area of the same type of graphics (such as columns, rings, spiders, etc.) to clearly express the contrast between the index values corresponding to different indicators. This approach will allow viewers to see the data and the comparison between them at a glance. When making such data visualization graphics, mathematical formulas are used to express accurate scales.
a: Check-in for City Class
This Check-in City Class histogram clearly shows the proportion of students in different regions. From the above picture, we can strongly perceive the absolute proportion of students in Beijing, Guangzhou, and Shanghai at first glance.
b: Federal Budget Map
As shown in the above figure, in the US federal budget map, the flow of funds is clearly expressed in different currency flows, and the proportion of each amount.
c: Enterprise Ability Model Spider Diagram
As shown above, through the spider map, you can see that the company’s ability in profitability and risk control capability is nearly 100 points, which can be said to be outstanding.
3.2 Data visualization by color
It is a common method of data visualization design to express the strength and size of the index value by the depth of the color. The user can see at a glance which part of the indicator data value is more prominent.
a: Click Heatmap
As shown below, via tracking mouse movements and creating the heat map from the mouse log, the user behavior is visualized by it.
b. Earthquake heatmap
The heat map below shows the seismic intensity of each place on the map and the distribution of the seismic intensity.
3.3 Data visualization by image
By using images and icons that have real meaning, can display data and charts more realistically, and can easily convey the meaning of the data.
For example, the graph below shows the proportions of each with a male and female icon as a background. At a glance, you can identify the men or women
3.4 Data visualization by the concept
By translating abstract indicator data into familiar, easy-to-perceive data, which is easier for users to understand the meaning of the graphics.
a: What is Unstructured Data?
Everyone knows that the iceberg suspended in the sea is just the tip of the iceberg. The iceberg below the sea is the vast majority of the iceberg. Explaining the amount of data of unstructured data and structural data and describing the characteristics of unstructured data through the form of conceptual transformation is very vivid and makes it easier to understand unknown and difficult concepts.
Infographics are the most extensive representation of concept visualization. If you are interested in making infographics, you can refer to 7 Data Visualization Tools to Create Infographics.
3.5 Data visualization by graphs or charts
When we design indicators and data, using graphics with corresponding actual meanings to combine presentations will make the data charts more vividly displayed, making it easier for users to understand the topics to be expressed by the charts. It is the most commonly used data visualization type.
So, let’s see the top 16 types of chart in data visualization and analyze their application scenarios to help you quickly select the type of chart that shows the characteristics of your data.
4. Top 16 Types of Chart in Data Visualization
4.1. Column Chart
Column charts use vertical columns to show numerical comparisons between categories, and the number of columns should not be too large (the labels of the axis may appear incomplete if there are too many columns).
The column chart takes advantage of the height of the column to reflect the difference in the data, and the human eye is sensitive to height differences. The limitation is that it is only suitable for small and medium-sized data sets.
Application Scenario: comparison of classified data
4.2. Bar Chart
Bar charts are similar to column charts, but the number of bars can be relatively large. Compared with the column chart, the positions of its two axes are changed.
Application Scenario: comparison of data (the category name can be longer because there is more space on the Y axis)
4.3. Line Chart
A line chart is used to show the change of data over a continuous time interval or time span. It is characterized by a tendency to reflect things as they change over time or ordered categories.
It should be noted that the number of data records of the line graph should be greater than 2, which can be used for trend comparison of large data volume. And it is better not to exceed 5 polylines on the same graph.
Application Scenario: the trend of data volume over time, comparison of series trends
4.4. Area Chart
The area chart is formed on the basis of the line chart. It fills the area between the polyline and the axis in the line chart with color. The filling of the color can better highlight the trend information.
The fill color of the area chart should have certain transparency. The transparency can help the user to observe the overlapping relationship between different series. The area without transparency will cause different series to cover each other.
Application Scenario: series ratio, time trend ratio
4.5. Pie Chart
Pie charts are widely used in various fields to represent the proportion of different classifications and to compare various classifications by the arc.
The pie chart is not suitable for multiple series of data, because as the series increase, each slice becomes smaller, and finally the size distinction is not obvious.
A pie chart can also be made into a multi-layer pie chart, showing the proportion of different categorical data, while also reflecting the hierarchical relationship.
Application Scenario: series ratio, series size comparison (rose diagram)
4.6. Scatter Plot
The scatter plot shows two variables in the form of points on a rectangular coordinate system. The position of the point is determined by the value of the variable. By observing the distribution of the data points, we can infer the correlation between the variables.
Making a scatter plot requires a lot of data, otherwise, the correlation is not obvious.
Application Scenario: correlation analysis, data distribution
4.7. Bubble Chart
A bubble chart is a multivariate chart that is a variant of a scatter plot. Except for the values of the variables represented by the X and Y axes, the area of each bubble represents the third value.
We should note that the size of the bubble is limited, and too many bubbles will make the chart difficult to read.
Application Scenario: comparison of classified data, correlation analysis
A gauge in data visualization is a kind of materialized chart. The scale represents the metric, the pointer represents the dimension, and the pointer angle represents the value. It can visually represent the progress or actual situation of an indicator.
The gauge is suitable for comparison between intervals.
It can also be made into a ring or a tube type, indicating the ratio.
Application Scenario: clock, ratio display
4.9. Radar Chart
Radar charts are used to compare multiple quantized variables, such as seeing which variables have similar values, or if there are extreme values. They also help to observe which variables in the data set have higher or lower values. Radar charts are suitable for demonstrating job performance.
The radar chart also has a stacked column style that can be used for two-way comparison between classification and series, while also representing the proportion.
Application Scenario: dimension analysis, series comparison, series weight analysis
4.10. Frame Diagram
The frame diagram is a visual means of presenting the hierarchy in the form of a tree structure, which clearly shows the hierarchical relationship.
Application Scenario: hierarchy display, process display
4.11. Rectangular Tree Diagram
The rectangular tree diagramis suitable for presenting data with hierarchical relationships, which can visually reflect the comparison between the same levels. Compared with the traditional tree structure diagram, the rectangular tree diagram makes more efficient use of space and has the function of showing the proportion.
Rectangular tree diagrams are suitable for showing the hierarchy with weight relationships. If it is not necessary to reflect the proportion, the frame diagram may be clearer.
Application Scenario: weighted tree data, proportion of tree data
4.12. Funnel Chart
The funnel chart shows the proportion of each stage and visually reflects the size of each module. It’s suitable for comparing rankings.
At the same time, the funnel chart can also be used for comparison. We arrange multiple funnel charts horizontally and the data contrast is also very clear.
Application Scenario: data ranking, ratio, standard value comparison
4.13. Word Cloud Chart
The word cloud is a visual representation of text data. It is a cloud-like color graphic composed of vocabulary. It is used to display a large amount of text data and can quickly help users to perceive the most prominent text.
The word cloud chart requires a large amount of data, and the degree of discrimination of the data is relatively large, otherwise the effect is not obvious. And it is not suitable for accurate analysis.
Application Scenario: keyword search
4.14. Gantt Chart
The Gantt chart visually shows the timing of the mission, the actual progress and the comparison with the requirements. So managers can easily understand the progress of a task (project).
Application Scenario: project progress, state changes over time, project process
The map is divided into three types: regional map, point map, and flow map.
(1) Regional Map
A regional map is a map that uses color to represent the distribution of a certain range of values on a map partition.
Application Scenario: comparison and distribution of data
(2) Point Map
A point map is a method of representing the geographical distribution of data by plotting points of the same size on a geographical background.
The distribution of points makes it easy to grasp the overall distribution of data, but it is not suitable when you need to observe a single specific data.
Application Scenario: distribution of data
But if you replace the point with the bubble, then the point map can not only show the distribution but also roughly compared the size of the data in each region.
(3) Flow Map
The flow map displays the interaction data between the outflow area and the inflow area. It is usually expressed by the line connecting the geometric centers of gravity of the spatial elements. The width or color of the line indicates the flow value.
Flow maps help to illustrate the distribution of geographic migration, and the use of dynamic flow lines reduces visual clutter.
Application Scenario: flow, distribution and comparison of data
The heatmap is used to indicate the weight of each point in the geographic area. In addition to the map as the background layer, you can also use other images. And Color in a heatmap usually refers to density.
Application Scenario: regional visits, heat distribution, distribution of various things
5. How to design good data visualization?
Now, you know the importance of visualizing data.
But how to actually use it?
For beginners, you can follow the design guide step by step. But no matter you’re a beginner or an expert, there are some tips to keep in mind when using it.
Know your audience
Your audiences usually have different backgrounds. If your data visualization aimed at professional audiences, you can interpret the data in more appropriate ways and technical terms. On the other hand, the general audience may need a clearer interpretation of the same data.
It is also important to know the audience’s expectations of the data. What are the key points they want? You need to be clearly present in the data.
Understand your data
After knowing your audience, you need to ensure the correctness of your data and understand your data clearly. If you do not understand your data, you cannot deliver your information to your audience
Storytelling with data
Your design should also convey a story, not the data itself but the information behind it. Using a story often means that the audience gets more insight from the data. It helps the audience deeply understand new information.
In fact, data visualization is a great storytelling tool because pictures tell a thousand stories and you should use it as your advantage. Telling a story through a data set is not difficult because you can use colors, fonts, and statements as part of your storytelling approach.
Keep it simple
There are many data visualization techniques, but it doesn’t mean you need to include too many different techniques for it. You need to keep your data visualization techniques simple. Firstly, compare the data visualization tool and choose one based on your needs. After that, pick the right data visualization type and the perfect color combination for your design.
Also, keep your visualization effects simple. Too many elements can actually corrupt the design and skew the data.
You need to remember:
The benefit of data visualization is to present large amounts of data intuitively. If your displays seem difficult, you’ll need to go back and see if you’re using the wrong data presentation or including too much verbatim information.