Ways to create a good data visualization and cases study

4 min readMar 30, 2022

Definition

Data visualization, as the name suggests, is a technique to interpret data by maps, charts or other graphical elements. With the help of these graphs, the data is easier to understand and can be represented more comprehensively, which means it is simpler to discover insights and relationship in the data.

Posted by Kevin Flerlage on public.tableau.com

Key Principles

Choose the right attributes and graphs to answer the query

As mentioned above, data visualization is a good way to interpret data and make it simple to understand insights. Based on this goal, we should focus on the key elements which are appropriate to answer the query. In addition, different graphs suits different queries. if query is to display distribution, box-and-whisker plots can be used; if query is to analyze statistics, bars and charts can be used; if query is to find out relationship, lines are suitable. Therefore, it is important to choose proper attributes and graphs.

Use proper labels, baselines, color and design

Avoid fancy labels that might compromise clarity; always start at zero when labeling the axes of a graph or chart, unless the data is clustered at unreasonably high values; color is used to draw the audience’s attention to key data; the graph should be simple and easy-to-understand.

Provide context and pay attention to titles

For queries, graphs should include some brief background and add some brief narrative after the visualization to highlight key insights. Besides, helpful, explanatory titles help highlight the point of the presentation, headlines serve as a snapshot of people providing key insights and focus them on the right questions.

Cases study

Hot cases

Here, I find some hot cases in different categories on public.tableau.com:

1.Expenses Audit Dashboard

This is a statistical dashboard about expenses. From the graph at first row, we can easily know the total and average amount of expenses and the number of expenses at the top, which is very intuitive. For the second row bar charts, it is easy to know what is the top one, how much is it, what is the proportion and other detailed statistics information.

However, I cannot know about the exact statistics attributes at first sight for some graphs, for example, I don’t know the total amount graph counts spending for a month or a year. From my perspectives, the author should provide some brief data explanation at each subgraph, which can give audiences a snapshot of the graph.

2. #VizForSocialGood- Self Reflection 2017 to 2020

This is a visualization for social good. At the first glance, it is beautiful and a little bit mysterious, which attracts me to look through. Different from the graph above, it contains some brief context at the left-hand side, which makes it simple to understand the goal. Although text labels in the center of rings are small, audiences can click the dot near each label to get detail. And we may curious about the meaning of each rings, the author also provides the representations and narratives at the right-hand side.

In my opinion, visualization should not only be clear, but also have a sense of design. It is also one of the key principles mentioned in the previous section: use proper color and design in viz.

3. Sunny Street | #VizForSocialGood

This is a graph about healthcare. It is organized, consistent and intuitive and has reasonable layout since it has succinct title, context, subtitles and also emphasizes key areas and elements using bold and large size text. and for typesetting, it is also appropriate because it groups statistics of the same type together.

Other cases

For this case, some necessary data is missing like title, context, explanation of statistics. Audiences can only know the blue circle is the largest. Besides, the text labels are also unsuitable, since in the purple circle, the text is larger than the circle.

For this case, it is chaotic and also missing the key information.

Conclusion

Data visualization is very useful not only at work but also in our daily life, for example we can use some tools like tableau, python, R, etc. to visualize our expenses or plan our trips. How to create a good and intuitive visualization is not easy, but just following key principles and regular practice can help us improve our visualization skills.

REFERENCE