Bad data visualizations can lead to costly errors. If the target audience can’t read and understand them, they won’t be able to make informed, data-backed decisions. Even small confusions can have massive consequences for the business.
What is Data Visualization?
To understand what bad data visualization is, we first need to discuss what data visualization is the graphical representation of information and data. It involves using visual elements like charts, graphs, and maps to convey complex data in an accessible and understandable way. Effective data visualization helps users quickly identify patterns, trends, and outliers within the data.
Bad data visualization
Bad Data Visualization
Data visualization is a powerful communication tool that has become increasingly important in our data-driven world. When done well, data visualizations can make complex information accessible, revealing insights and patterns that would otherwise be difficult to discern. Effective visualizations can quickly convey dense datasets clearly and compellingly.
Poorly designed data visualizations can lead to significant misunderstandings. Charts and graphs that distort, obscure, or misrepresent the data can paint a misleading picture, causing viewers to draw faulty conclusions. This can have real consequences when these visualizations are used to inform important decisions in business, policy, and other high-stakes domains.
Common Mistakes in Data Visualization
One of the most insidious ways that data visualizations can mislead is by distorting the underlying data. This can happen in a variety of ways, but often involves manipulating the visual representation to exaggerate or obscure certain trends and relationships.
Distorting the Data
Improper scaling of axes is a common culprit. By failing to start the y-axis at 0, or compressing the scale, a visualization can make small differences appear much more dramatic than they really are. Truncated or compressed axes that don’t show the full range of the data can create the illusion of large changes, even when the actual differences are minor.
Another way visualizations can distort the data is through the use of 3D effects. While 3D charts may look visually striking, they can actually skew the perception of quantities and make it harder to accurately compare values. The foreshortening and angular distortion inherent in 3D graphics can cause bars or slices to appear larger or smaller than their true size.
Cluttered and Unclear Displays
Another common pitfall of bad data visualization is the creation of overly cluttered and unclear displays. It can be tempting to try and cram as much information as possible into a single graphic, but this often results in visualizations that are busy, confusing, and difficult to interpret.
Too much information crammed into a single graphic can make it challenging for the viewer to quickly identify the key insights and takeaways. Excessive use of colors, fonts, gridlines, and other decorative elements can also contribute to a sense of visual noise, distracting from the core message.
The result of these cluttered and unclear displays is that the data visualization fails in its primary purpose – to communicate complex information in an accessible and meaningful way. Instead of highlighting the key insights, overly busy graphics can overwhelm the viewer and make it difficult to discern the real message.
Misleading Choices of Chart Types
In addition to distorting the data and creating cluttered displays, data visualizations can also mislead through the inappropriate selection of chart types. Choosing the wrong visual representation for the data and message at hand can significantly impact how the information is perceived and interpreted.
For example, using pie charts to compare quantities is often problematic, as the visual area of the slices does not necessarily correspond linearly with the underlying values. Pie charts are better suited for showing part-to-whole relationships, rather than making precise comparisons.
Similarly, the use of 3D charts, while visually striking, can skew the viewer’s perception of the data in the same way as 3D effects discussed earlier. The foreshortening and distortion inherent in 3D visuals make it harder to accurately judge and compare the relative sizes of chart elements.
Another common misstep is the failure to start bar charts with a y-axis that begins at 0. When the y-axis is scaled to highlight small differences, it can exaggerate the visual impact of the data, leading to an inaccurate interpretation of the magnitude of change.
The key is to match the chart type to the specific data and the message you’re trying to convey. Using the wrong visual representation, even if it looks visually appealing, can inadvertently distort the meaning and mislead the viewer.
Data Visualization for Decoration, Not Communication
In some cases, data visualizations are created primarily to look aesthetically pleasing rather than to effectively communicate information. While visually striking graphics can be engaging, if they fail to align with the underlying message and data, they become mere decorations rather than useful tools for conveying insights.
In bad data visualization, charts that are used mainly to enhance the appearance of a report or presentation, without a clear purpose or connection to the content, are often ineffective. The visual elements may be carefully crafted, but if they don’t match the data and the key story you’re trying to tell, they can end up distracting the viewer rather than enhancing understanding.
Similarly, visualizations that try to be too creative or unique in their design, at the expense of clarity and simplicity, often fail to convey the intended insights. When form takes precedence over function, the end result may be an engaging graphic that doesn’t actually help the viewer comprehend the data.
The most effective data visualizations are those that are carefully constructed to align with and amplify the message. They enhance understanding by presenting the information in a clear, accessible way, not by focusing solely on visual appeal. Achieving this balance between aesthetics and communication is essential for data visualizations that truly drive meaningful insights.
Lack of Context and Framing
One of the most common problems with data visualizations is the lack of proper context and framing. When critical details about the data are omitted, it becomes very difficult for viewers to accurately interpret what they are seeing.
Important contextual information that is often missing includes the data source, the time period covered, the units of measurement, and any other relevant metadata. Without this basic information, the meaning and significance of the visual representation can be severely diminished.
Similarly, failing to provide necessary reference points for interpreting the data can leave viewers adrift. Visualizations that don’t include clear labels, legends, or benchmarks against which to compare the values being shown can force the viewer to make unsupported inferences.
The result of this decontextualized presentation is that the charts and graphs lose much of their meaning. Viewers are left to fill in the gaps themselves, which often leads to misinterpretations and flawed conclusions. What may have been an insightful data visualization ends up becoming a decontextualized image that fails to communicate effectively.
Effective data visualization requires providing the proper framing and context so that the viewer can quickly understand the significance of what they are seeing. By including the right supporting details, the visual representation can fulfill its purpose of enhancing understanding, rather than creating confusion.
Principles of Effective Data Visualization
- Focus on Clarity and Simplicity
The primary goal of data visualization should be to communicate information as clearly and simply as possible. Avoid cluttering the display with excessive visual elements that distract from the core message. - Match the Visualization to the Data and Message
Choose chart types and design elements that are well-suited to the specific data you’re working with and the key insights you want to convey. The visual representation should amplify and enhance the underlying meaning. - Provide Proper Context and Framing
Include critical details like data sources, time periods, units of measurement, and any other relevant metadata. Provide necessary reference points to help viewers properly interpret the information. - Avoid Distortions and Misleading Elements
Be vigilant about potential ways the data could be misrepresented, whether through improper axis scaling, 3D effects, or other visual tricks. Ensure the visualization accurately reflects the true meaning of the data.
By adhering to these core principles, data visualizations can effectively communicate complex information in a clear, accessible, and impactful way. The goal should be to enhance understanding, not obfuscate or distort the underlying data.
Example of Bad Data Visualization
Let’s explore some examples of bad data visualization to understand common mistakes and how they can mislead viewers:
- Lack of Context: Consider this bar graph showing forecasted temperatures for each day of the week in Fahrenheit. Without context, it’s unclear whether this graph represents temperature, wind speed, or something else. !Temperature Graph
- Inconsistent Scales: In this visualization of electricity price changes in Spain, the horizontal scale switches from yearly to quarterly after 2012. This inconsistency can confuse viewers and distort the data. !Electricity Price Change
- Cluttered and Confusing: Some visualizations cram too much data into one graph, making it complex and hard to interpret. Clear labeling and simplicity are crucial for effective communication.
Remember, good data visualization should enhance understanding, not confuse or mislead.
Real life examples of bad data visualization
The real-world Bad Data Visualization examples illustrate how data visualizations can go wrong when the principles of effective visual communication are not applied. By being mindful of these pitfalls, we can create data graphics that accurately represent the information and truly enhance understanding. Here are some real-life worst data visualization examples:
- Misleading 3D pie charts:
A common problem is the use of 3D pie charts, which can distort the visual perception of the slice sizes. This example from the UK government shows 3D pie charts that make it difficult to accurately compare the relative proportions. - Inappropriate use of bar charts:
In this example from a news article, the y-axis on the bar chart does not start at 0, which exaggerates the visual differences between the values. This can lead to an inaccurate interpretation of the magnitude of the change. - Cluttered and confusing dashboards:
This dashboard from a financial services company crams too much information into a small space, making it difficult to quickly identify the most important insights. The overuse of different chart types and colors creates a cluttered, overwhelming visualization. - Decontextualized charts:
This chart about smartphone sales lacks critical details like the time period, units of measurement, and data sources. Without this context, it’s challenging for the viewer to properly interpret the meaning and significance of the information presented. -
Misleading data aggregation:
In this case, the data visualization averages together very different underlying metrics, masking important nuances and potentially leading to incorrect conclusions about the trends.
Good vs Bad Data Visualization
Bad data visualization
