Misleading Data Visualization Examples will be discussed in this article. Everything around us in the modern digital world is made up of data, but not all of it is reliable. Although we don’t often see this data in its raw state, we use it to determine whether something is true or false.
15 Misleading Data Visualization Examples In 2023
In this article, you can know about 15 Misleading Data Visualization Examples In 2023 here are the details below;
Consider how difficult it would be to understand all the rows and columns of numbers; this is why data visualization is necessary. Data visualization aids in making meaning of the data at our disposal by presenting patterns. To convey trends, data is transformed into visuals like charts and graphs. As useful as data visualization is for understanding data, it can also be used to distort reality and show false trends.
1. Cherry Picking
The term “cherry-picking” was developed from the idea that, if you only select the ripest and healthiest fruit, anyone who witnesses you doing so might be led to think that all of the other fruits on the tree are also in good health, even though this is untrue. Also check strategies for instagram marketing
When testing the responses of some animals, like cats and dogs, this phenomenon is very prevalent in the veterinary field, where veterinarians are more likely to report on successful trials. When they have received funding from pharmaceutical firms, this mostly occurs. This is also a common occurrence in commercial drug studies for human medications like antidepressants. When you run the same drugs through a government-funded study, the findings don’t always line up.
Cherry-picking is risky because it’s clear that the researcher could be off. There may be some hidden complexities in the research that can only be discovered with more data. It is frequently simple to identify cherry-picking. To begin with, there will be a limited number of participants in the study, and the data collected from them will only be partially made public. Additionally, the researcher frequently uses generalizations and repetitions of specific examples, which only serves to demonstrate the study’s lack of saturation.
2. Cumulative vs. Annual Data
To make sure that the graph only increases after each input, cumulative data is what you add to the data model’s subsequent inputs. Annual data, however, will display the data for each year. It is possible for the specific years to be rising or falling, which depicts the reality more accurately. You have undoubtedly seen the Worldometer COVID-19 histogram or other charts like it. They have been in great supply throughout the epidemic. The total number of patients in the region under review is frequently shown in these graphs.
Companies frequently use cumulative data visualization to make their sales appear higher than they actually are. This came under fire for Apple in 2013 when CEO Tim Cook gave a presentation that only included the total number of iPad purchases.
Many people believed that it was done on purpose to stop the declining sales of iPads. While the advantage of displaying changes in growth and total number is a benefit of cumulative data, you must dig deeper before you can understand some very significant changes. Calculus may be easier for you to solve.
3. Misleading pie chart
Politicians frequently manipulate data to present a certain group or individual in a more favorable light than they actually are. A visualization tool used to accomplish this is the pie chart. You’ve probably seen a lot of pie charts used to illustrate various proportions, such as which political party voters are most likely to support. But what if poll respondents were given the option of selecting more than one party? In that scenario, you might receive more than 100%. Pie charts make it difficult to present this because they are designed to display proportions of a whole where each group is unique.
A pie chart will give the impression that a sizable portion of voters supported that specific contender alone. Your best option is to use a Venn diagram if you want to correctly represent such data with a visualization element like a pie chart. The Venn diagram will display the vote totals and areas of overlap for each contender. This is another misleading data visualization examples.
4. Omitting the baseline
Bar Graphs have a reputation for falsifying data by changing the y-axis’s size. Politicians frequently use this particular visualization tool to overstate how certain things have fared under their government. For instance, a politician could start the graph’s y-axis at 50% rather than 0% in order to overstate how the high school graduation rate has grown during his administration.
The difference between 65% and 70% is then more important than if he had started at zero. In data visualization, this is referred to as truncating the graph. Although not always with the intent to deceive readers, news organizations also employ this tactic to mislead readers and viewers. Sometimes they try to be original or creative with a graph by doing this, but they might find themselves in difficulty in the process. In the tech sector, the music industry, and pretty much everywhere else where people are attempting to support their claims, you will also find this kind of misleading data visualization. Also check self managed teams
5. Manipulating the Y-axis+
This is somewhat opposite of our prior example but very similar to it. The diagrams in this instance include the baseline and the axis, but they have been altered to the point where their original significance has been lost.
Data manipulators accomplish this by altering the graph’s scale to either exaggerate or conceal a change. Axis shifting is a very popular technique used to spread false information on social media in the data visualization field. For instance, to make the line of a graph about global warming as flat as feasible, temperatures from -10 capacities to over 100 degrees may be included.
This is frequently used to advance erroneous claims that global warming is exaggerated or unreal. The majority of the time, this kind of misleading data is not done accidentally. Even though the manipulators frequently are aware of what they are doing, they still opt to promote the false story. This is another misleading data visualization examples.
Sometimes all it takes is one alteration to totally change a story that could end up costing you a lot of money. Your credit rating may appear very good or very poor depending on the scale a credit bureau chooses to use, for instance. A 634 out of 700 is significantly different from a 634 out of 850. Be aware that not all misleading data is displayed in line or bar diagrams. Pie plots are another way to show them.
6. Using the Wrong graph
We’ve discussed how people purposefully misrepresent data visualization in the examples above to further their own goals, but it can also happen due to incompetence. People frequently choose the incorrect visualization tool inadvertently, which results in inaccurate representation of the data. Pie charts frequently exhibit this, but that does not mean that the pie chart is to blame. Not at all. Usually, the individual who chose the pie chart in the first place is at fault.
This pie chart from the NFL selection serves as an illustration of what we mean: Can you tell what it’s trying to convey at first glance? Heck, even if you spend the entire day studying it, it’s possible that you still won’t be able to comprehend it. For whatever data they are attempting to represent there, a bar graph would have made a lot more sense. This is another misleading data visualization examples.
This type of data misrepresentation is common, particularly when brands attempt to be overly inventive with their graphs and charts. Even if you want your data to seem less dull, you must come up with other solutions rather than compromising its truth.
7. Going against convention
Going against long-held conventions or associations is typically a bad notion when it comes to data visualization. Imagine a graph where red represents profit and green represents loss because we are so used to using the hues red and green to denote loss and profit, respectively. It would be complete mayhem.
What happens if a zealous election reporter chooses to depict Democrats with red and Republicans with blue? Many individuals would find that to be perplexing. Take a look at the image below, which displays the prevalence of STIs in America. Contrary to what one might anticipate, the lighter hue depicts low levels while the darker color indicates high levels. This is another misleading data visualization examples.
This defies logic, but it was probably done on purpose to deceive the public. Given how clear the creator’s purpose was, it is one of the multiple misleading graphs I have ever seen. In reality, the line is the incorrect way around from how it ought to be. Although the number of gun fatalities in Florida was actually increasing, it was made to appear that way. It is clear that this was done to advance a political goal by cynical data manipulators.
8. Overloading readers with data
This is a typical error. One too many times, I’ve seen. Including too much data on a single graphic will confuse an audience rather than inform them, whether it’s on purpose or not.
The most current COVID-19 graph shared by the White House is a good example. If you can’t make out anything on the line, don’t be hard on yourself. To be honest, you’d need to be a genius to comprehend this without a magnifying lens. Not only are there too many trend lines for me to comprehend, but the legends on the side also make it difficult. The chart would be readable if the creator isolated the various states or created different graphs for each state.
9. Omitting data
Some data analyzers believe that withholding information is preferable to lying about it. They were correct about one thing: it’s bad to lie about data, but it’s also bad to leave out data. By omitting data, you leave room for others to infer trends that don’t exist and, in a similar manner, you risk missing some important insights. When data is omitted, it leaves room for interpretation and all kinds of inferences can be made from it. In spite of this, data manipulators purposefully remove information to deceive readers. On the other hand, data can be omitted due to creator indifference, making their job simpler by omitting some data points like dips and surges.
10. Number doesn’t add up
A pie chart or stacked bar chart must have figures that add up to 100% according to statistical convention. Even though it ranks high on the list of the most ridiculous errors, it is still committed far too frequently. Take a look at our prior illustration, which is a presidential election chart from Fox News. Have you yet detected the error? You did, of course. The three candidates’ percentages do not sum up to 100%. Rather, it is a staggering 193%. This is another misleading data visualization examples.
A Venn diagram would have been a better choice to depict the concept, but Fox chose for a pie chart, as we have previously stated, the poll that led to this result must have allowed for more than one answer. The size of the pies also tells a different tale from the numbers not adding up. Contrary to what is actually the case, it gives the idea that each candidate holds close to a third of the total number.
11. Not using annotations
The use of annotations for visuals should be at your option, so it might be overstating the case to call this a mistake, but it is good practice to include them each time you create a chart. There will be times when visuals alone are insufficient because your charts will probably be viewed by a variety of groups. Only works with qualifying text and figures in those circumstances would make sense to the perplexed readers. Take a peek at the graph below as an illustration. The axes are clearly labeled and it appears good, right? But wouldn’t you be interested in learning what occurred in 2015 to account for the decline in sales? In this kind of circumstance, a decent annotation will be very helpful. Isn’t this the way it looks better?
12. Improper bubble sizes
Each component of a graphic has a purpose, and bubble charts are no different. They are employed to show two-dimensional representations of three-dimensional data. This is another misleading data visualization examples.
Many people frequently change the radius rather than the area to show the data when trying to represent data with bubbles. To see what I mean, take a look at the bubble diagram below.
I can already tell from looking at the map that there are a few problems, but let’s focus on the first two bubbles from the left in order to emphasize this point. Do you not think it is strange that one equals $0.92 billion and the other $1.84 billion? If you were to judge the size of the bubbles, you would assume that the bigger bubble is at least four times larger than the smaller one, even though it is actually only twice as big. Therefore, if the texts weren’t present, it would almost definitely have been misinterpreted. Any time you change radius rather than area, that is what is inevitably going to happen.
13. Hard to compare
Comparing market shares across various nations is a very common practice for company owners with global reach. Here, data visualization is helpful, but in some cases, it can make comparison more challenging. This is another misleading data visualization examples.
Look at these two illustrations:
- Which would be simpler for you to read?
- It baffles me why anyone would like to use a pie chart to display this sort of data, but many businesses do, so undoubtedly the bar charts.
- Examine your writing from the viewpoint of the reader to prevent such errors. Or, even better, have a friend review the charts before you post.
14. Correlating causation
This is another misleading data visualization examples. In the era of the internet, it can be difficult to remember that correlation doesn’t always imply causality. Unfortunately, this has permeated the data business, and more and more researchers are starting to draw inferences based on correlation rather than causation. The graph below demonstrates how misleading correlation-cause visualizations can be.
15. Wrong audience
Finally, it’s crucial to remember that no audience can be completely satisfied by a visualization. Consider your target audience and the type of data they can absorb when developing a visual. Imagine showing a group of third graders a convoluted graph detailing how countries with high pollution rates are the most affected by climate change. That is far from the truth. This is another misleading data visualization examples.
Although there are many methods to manipulate data, the ones noted above are some of the multiple popular. You can make better decisions and be more selective about the data you entertain now that you have this information.