In the process of making charts, we will inevitably encounter various problems. What data visualization mistakes should we avoid to make a proper chart?

People who like data visualization, it is really difficult to make a good chart, and it is not easy to make a chart without making mistakes. The purpose of the chart is to provide an easier and faster understanding of shortcuts. Under this purpose, a lot of information is simplified and hidden, but this simplified process may not be able to completely reproduce the information. The original intention is completely opposite and misleads the reader’s thinking.

By digging deep into our customer reviews, we carefully analyzed the problem of underperforming charts and found that they usually have the following three characteristics:

Misleading

Confusion

Cannot explain the problem

If we analyze it in more detail, we will find the cause of an “imperfect” chart is various. It may be due to the mismatch of data and chart types, or the improper setting of parameters such as color matching … We will introduce 5 common data visualization mistakes, let’s check now!

Don’t Use 3D Charts Casually

Although you may have seen many 3D charts, and even some tools support the production of 3D charts, 3D charts are not very applicable in most situations, especially in application scenarios that emphasize data comparison and analysis, 3D charts are for readers. While bringing cool visual effects, it often distorts the authenticity of the data itself, thereby affecting the audience’s judgment on the results.

Take the following pie chart as an example, we can see how the 3D  chart distorts the authenticity of the data.

From Visually

It can be seen from the 2D pie chart that three colors occupy the same part, but in the 3D chart, you will feel that the white part is larger than the other two parts.

This principle is actually very simple. When an object becomes three-dimensional, it will be smaller visually in the far place and larger in the near.

It can be seen more clearly in the histogram below.

From Google

Similarly, for chart viewers, it is important to be vigilant when seeing similar 3D charts.

Misleading by Map Area

Maps are an important part of data visualization. A key point of geographic visualization is the division of geographic regions, which makes it easier for readers to view the distribution of data in different geographic regions. Maps usually color different areas (such as gradient colors) to show population or election data. For example, the different results of voting are represented by red and blue. The problem is that the size of the area may not be related to the topic of the chart, and it may even be misleading.

For example, the coloring of the ruling party, usually we are more concerned about how many people are affected under the scope of the political party, but the population is not necessarily positively related to the land area.

By the way, here is a good example

From FineReport

Reverse Causality Diagram Explanation is not Advisable

Bloomberg BusinessWeek published a chart in 2011 illustrating the misuse of correlation and causality, the most classic of which is the match between a mountain range and the murder rate in New York: this is amazing! The murder rate in New York is the same as that of the mountain! So can the next movement of the mountain predict changes in crime rates? The answer is of course no.

From https://www.bloomberg.com/news/articles/2011-12-01/correlation-or-causation

When making a chart, you may have some interesting ideas and discoveries. For example, try to overlap the population with annual income? It seems that the more people the higher the income. As long as you are willing, you may always find the consistency of the trends in some two events. This is a good attempt, but don’t forget: you may find something, but it doesn’t mean you prove anything.

Parameter Settings are Also Important

In order to make the chart more beautiful, the detailed setting of the chart also requires us to take some effort. For example, if the interval between each column of the histogram is too wide or too narrow, it is not good-looking. You should set the interval width according to the column width. In terms of color matching, gradient colors are good-looking and simple color matching. Don’t use garbled and non-realistic color matching in the visualization, which will divert the reader’s attention to the data.

As an easy-to-use visualization tool, FineReport perfectly avoids many problems that users may encounter in drawing. For example, in terms of chart types, FineReport supports 19 categories and 50+ types of charts, which can generate cool charts without programming. In terms of data, each chart has sample data and introductions. For existing data, it can also “One-click visualization” greatly improves the efficiency of drawing; in terms of color matching, you can completely set the color matching you want, and at the same time, you can choose multiple color matching schemes …

From FineReport

Convey the Right Information

Pictures are the best way to convey information, and the very popular infographics in recent years have proven it well. But in many cases, people mistakenly believe that the more data displayed, the better. This is actually a wrong idea. When selecting data, we need to pay attention to the logic and relationships behind the data and display them in pictures. The dashboard is a good choice, if you want to learn how to make it, you can refer to our guides :

https://www.finereport.com/en/data-visualization/a-beginners-guide-to-business-dashboards.html

From FineReport

Conclusion

As we process data, we need to figure out what is the best way to represent the data, and sometimes even data visualization is superfluous. Maximizing the value of data is our ultimate goal.