Data visualization is one of the easiest forms of data representation to understand because our eyes are drawn to colors and patterns.
For data visualization, charts, graphs, and maps are mostly used. In fact, it is ideal when interpreting big data. However, there are good and bad data visualizations. For a data visualization to be fair, it should follow basic principles.
Most interpreters ignore these principles which lead to bad data visualization – such that it’s difficult and impossible to comprehend.
Here are 15 bad data visualization examples.
Bad Data Visualization Examples
1. ESPN CricInfo Cities with the best batting talent
In 2019, ESPN CricInfo published an article on Which Top Cricket City Would Win the World Cup. Featured in it was the above data visualization that represented the cricket cities with the best batting averages.
Looking at the graphics, it’s difficult to pick up any meaning from the data. To get an insight into what this visualization is all about, you’ll have to read the written texts – before the graphics and the ones associated with it.
One of the things wrong with this data visualization is the color. Such a color combination of red and blue is least friendly to the eyes.
You cannot categorize this visualization to any visual data visualization type. The closest it gets is to a bar chart, but instead of bars, the designer used different shapes representing each city.
Looking at a bar chart, you can grasp the difference between each bar by the height difference. In this data visualization by ESPN CricInfo, the shapes’ size, height, or width tells no difference. The lowest average is Mumbai at 35.51, but it is the tallest.
You Might Want To Read: Best Tableau Courses
2. Most Wickets in Death Overs in Odis
The above data visualization is a relatively simple one. However, it is bad because of one major thing; the color.
The bar chart represents the wickets and batting averages of cricket players. Both wicket and average are represented in orange. Although the average is of a lighter shade, you can easily miss the color differences. When making such a chart, a good color to use in contrasting orange is blue.
Furthermore, the bar chart is stacked with numbers that have different units. Wickets are measured in length using mainly inches and centimeters. On the other hand, averages have no units.
The stacked up data also disorganizes the chart’s subject. This data visualization was meant to display most wickets in death and not most wickets in death + averages.
Looking at it, the player you would easily say has the most wickets is the first player, Jasprit Bumrah – which is correct. The player you would easily say has the second most wickets is the sixth player, Bhuvneshwar Kumar – which is wrong.
The second player has the second most wickets, but the sixth player has a more extended bar than him.
3. BBC Avocado Toast Index
The above data visualization doesn’t tell you anything important. There are up to 5 of them, and the interpreted data was shared between all five graphics. Therefore, to comprehend the data, you must consult all five graphics.
The visualization is about How Many Avocado Toasts Does It Take To Afford A Deposit On A House. Ten cities are used in the study – Mexico City, Johannesburg, Berlin, Tokyo, New York, Sydney, Vancouver, San Francisco, Hong Kong, and London.
Conversely, two cities are used for each graphic. Despite this, it’s difficult to grasp what each of the five graphics is talking about.
The designer represented 100 avocado toasts with one cup and then one dollar notes for an avocado toast price per city. The varying number of avocado toasts consumed and the price of the toasts make the representation disorganized.
The data could be fully prepared and communicated in a straightforward bar chart.
Check Out: Best Tableau Retail Dashboard Examples
4. India Today, Chances of NDA Coming into Power
Ahead of the 2019 elections in India, India Today published an article to discuss the chances of Prime Minister Narendra – NDA – Modi winning a second term. While NDA did win his second term, understanding his chances via this visual data is puzzling, unless all India Today readers are professional data analysts.
It’s ideal to use a speedometer chart in data visualization like this one (or a pie chart in general) when you have collectively exhaustive and mutually exclusive quantities. In this chart by India Today, none of the two is the case.
The chart is divided into 3 – NDA staying below the 220 mark, NDA crossing the 250 mark, and NDA getting a majority. The probability is 9%, 72%, and 50% respectively.
Commonly, a speedometer chart has just one pointer. However, the graphic designer here decided to feature two pointers – one points in-between the 9% and 72% probability, while the other points at the 50% probability.
An ideal explanation here would be that the chances of NDA winning a second term fall on the 72% probability as it’s in between the pointers. However, there’s no way to be sure.
5. Walt Disney’s Companies Worldwide Assets
Walt Disney has worldwide assets totaling close to $200 billion. If you want to know every company Disney owns, it’s easier to read about it than consulting this infographic by Titlemax. The infographic is a definition of bad data visualization.
This should be highly informative and valuable data visualization if it were not so large and complicated. There is so much information to include, and the designer didn’t do so well with his choice of font size, line weight, circle sizes, etc.
Mickey Mouse’s shape (one of Walt Disney’s most popular characters) is featured, which brings about 3 different circle sizes. Viewers can easily discern that the companies in the larger circle are most important which isn’t so true.
Also, it makes it seem the companies in smaller circles outside the Mickey Mouse frame are less important. Furthermore, the yellow on white color combination is never ideal to use.
Inside the big circles, too many other shapes were used. There are smaller circles, rounded rectangles, sharp cornered squares, and whatnots. Finally, the resolution of such a large data visualization should be at the highest possible rate else reading the small texts will be impossible.
6. Dialect Map of India
English and Hindi are the two state languages in India. However, there are about 121 recognized languages out of 1,369 recognized dialects. This dialect map of India aims to represent this data visually via a Choropleth.
A Choropleth is a good way to visually interpret data, but it becomes unreliable when there is so much data to interpret. That’s what this map designated failed to perceive. The map features 60 Indian dialects and others. Hence, there are so many colors, which makes the data representation appear as a color riot.
You can’t tell which dialect is the most dominant by looking at this map. Matching the colors to the top languages will also be challenging as colors of close shades are featured next to each other.
The black background makes things a bit more complicated as it mixes with the black colors featured on the map.
Choropleths are best for representing how variables change across different areas. When it comes to population distribution, which this map is all about, choropleths are not ideal because it results in uneven distribution.
7. The Economist, Why ticket prices on long-haul flights have plummeted
The Economist isn’t a publication you would expect a bad data visualization example from but here’s one. The core here is the featured protractor.
You would expect to grasp a relationship between long-haul flights and their ticket prices. However, what you get are overlapping lines such that differentiation is a problem.
The protractor features two lines. The blue-colored lines are for transatlantic flights while the ash-colored lines are for other flights. There are three axes with two being Distance in Km and the other, Change in price of economy-class tickets.
Since the length of the lines depends on the flight’s distance, some lines are terse. Hence, it’s difficult to trace them to grasp the percentage change in ticket prices. Furthermore, the actual prices of the flight tickets are not pointed out.
Except for their not-so-straightforward titles, the three graphs featured below the page are easier to understand. The first represents shares of Norwegian seats on six transatlantic routes, and the second represents jet fuel $ per liter. In comparison, the third represents average ticket prices on six transatlantic routes.
8. ASEC Data, Household Income Percentiles
This was supposed to be a simple double bar graph, but the designer got the opposite in trying to make it simple.
The graph was meant to represent household income percentiles for 2017 and 2018. However, it was labeled 2017 and 2016. This alone renders this chart unreliable since you cannot grasp what the data is representing.
If you want to utilize this graph, you’ll have to consult the original ASEC data to verify the variables. Nevertheless, with 2017 coming before 2016, 2016 could be an error.
As a double bar graph, the household income percentiles are represented side by side. While the yellow and light blue color combination isn’t awful, a stronger color variation would be better. For example, yellow and standard blue color.
The next thing wrong with this data visualization is the alignment of the text within the bars. You would need to twist your head or rotate the image on your device to read it which can be uncomfortable.
Furthermore, with the dollar cutoff figures at the y-axis, there’s less need to write the actual figures within the bars.
9. HuffingtonPost, How India Eats
This is a data visualization by HuffingtonPost India on how India eats. It’s not a complex graphic to understand, but the designer’s choice of interpreting the data isn’t the best.
This infographic shows the percentage of vegetarian and non-vegetarian eaters in different parts of India. The percentages are represented with concentric circles such that there’s always an overlay.
The sum of the two percentages for vegetarians and non-vegetarians equals 100 – which is correct. However, how is the size of the circle for each percentage measured? Was the radius or the area of the circle used?
Looking at the graphic, all the outer circles seem identical. Meanwhile, they mostly feature different percentages. The inner circles better represent their percentage size.
Furthermore, in the case of equal distribution (50% vegetarians and 50% non-vegetarians), it’ll be impossible to represent using such concentric circles.
Take, for example, the case of Madhya Pradesh (MP) as represented in the diagram. There are 50.6% vegetarians and 49.4% non-vegetarians – a 0.6 difference. While the distribution is almost equal, it looks like the state has more non-vegetarians as red is the color in front.
10. Vox, All life on Earth, in one staggering chart
The first thing you’ll notice from this Vox data visualization is its length. That’s the first reason why the visualization is bad. The graph won’t easily fit on any web page, so it’ll take a viewer multiple scrolls to view all the details.
The graph shows all life on earth which includes all plants, animals, and humans. With plants at 450 Gt C, Animals at 2 Gt C, and Humans at 0.06 Gt C.
Looking at the graph, there are more plants than both animals and humans. Also, there are more animals than humans.
It’s the simple fact the chart aimed to interpret, but the choice to use 3 dimensions was a flaw that made it complicated. Most viewers who look at this data visualization will perceive more to it than the fact mentioned.
It would be difficult to represent the actual figures of plants, animals, and humans in the world – or even their size ratio – in a graph. This makes the lengthiness more irrelevant.
11. Bloomberg, Polluted Cities
If you’re a regular Bloomberg reader, you’ll know this is the standard way the publication prepares their bar graphs. They aren’t totally bad, but they aren’t the best either.
In this particular graph, Bloomberg is not just representing polluted cities in India and China. The percentage increase in these cities from 1998 to 2016 is also being represented.
There are 20 cities in total with 10 Chinese cities and 11 Indian cities. Both are differentiated with ash and white colors, respectively.
The first thing wrong about his chart is how the labels are placed – far away from the graph. Although this is due to the inclusion of the -% and all but one city has a positive increase in pollution, it links the bars somewhat abstruse.
The designer employs a progressive pattern in arranging the cities. However, there are no start and endpoints. Also, you can’t point out just how much of a percentage increase in population was recorded for each year.
Finally, the use of an all-black background is always not ideal when preparing a data visualization.
12. Visual Capitalist, 80 Trillion World Economy One Chart
In an article on the 80 Trillion World Economy One Chart, Visual Capitalist features this data visualization prepared by Howmuch.net. The graphic represents the revenue and percentage of the economy of all countries.
Looking at the visualization, you can only easily pick out countries with booming economies which include the United States, China, Japan, Germany, France, etc. If your country doesn’t have a major economy, it’s difficult to find it.
Another thing wrong with the chart is positioning. It’s represented in a circle which can be associated with the Earth’s spherical shape. However, the countries are placed just anywhere which doesn’t mean anything.
Nevertheless, countries of the same continent are positioned close to each other but the sizes are not relative.
The positioning makes it hard to sum up all the revenue of each economy and percentage to check if it equals 80 Trillion and 100 percent respectively.
Furthermore, the shapes representing the countries are unusual. Although the size of a country’s economy determines its size, nothing relates the shape’s size to the size of the full circle.
13. P&G Annual Report
Here we have a data visualization of the P & G annual report. The visualization represents sales by business segment, geographical region, and market maturity in the year 2018. It also displays financial highlights for five previous years (2014 to 2018).
Thanks to the alignment and orderliness, this visualization could easily pass for a properly prepared one. However, there are a few errors.
First of all, the light blue background color doesn’t sit well with white texts. Furthermore, similar colors are used in the charts such that it could be difficult to pinpoint the chart. Thankfully, the yellow color makes it more noticeable.
Donut charts are not the best when it comes to representing more than four categories. In such charts, the categories are separated by areas and lengths, making them difficult to distinguish. This is why it’s easier to grasp the last chart and the first two charts.
Furthermore, the visualization features a legend at the top which is somewhat distant from the charts. Hence, there’s a need for constant head movement, which can inconvenience a viewer.
14. World Debt in One Visualization
How much do world governments owe? As of 2018, the figure was estimated at $63 trillion as represented in this data visualization. Visual Capitalist prepared the chart.
The United Nations Of Dept graphic by Visual Capitalist is grotesque and it has a lot to do with the shapes used. Varying shapes – with some very bizarre ones – were featured.
To properly represent data visually, it’s ideal to stick with regular shapes to make it intuitive. Furthermore, they are easier if you want to calculate areas for sizes.
Like the 80 Trillion World Economy One Chart mentioned earlier (still from Visual Capitalist), it is difficult to locate countries with small debts. Only countries like the United States, Japan, China, Italy, etc., with large debts, are easily located.
The color scheme of this data visualization doesn’t sit right either. A lot of similar colors were used to represent most countries in which very abstract colors were used in one or two places.
Finally, this visualization holds no value logically. This is because countries and companies owe each other. Should each country’s liabilities be crossed with their assets, there’ll be zero debt.
15. LiveMint, Homosexuality in India
For this bad data visualization example, LiveMint intends to interpret a time series. To show how much support for homosexuality in India has grown over the years.
Here, the publication used an area chart displaying broadly supportive and broadly opposed respondents. The problem with this chart is that it doesn’t particularly point out how much support growth was recorded per year.
It seems like the years were set randomly. The displayed years include 1990, 1995, 2002, 2006, and 2014 – no orderliness.
The color used to represent the broadly supportive respondents can be easily missed. Considering this is the most important category as it shows the support growth, a more visible color should have been used.
The graph aims to show the growth of homosexuality support, making the broadly opposed respondents’ data quite irrelevant. Furthermore, the broadly opposed graph also has an up-trend (which signifies growth), and this makes the visualization counterintuitive.
When representing information and data graphically, it’s important to apply best practices, so they don’t appear complex. However, the creators of the 15 bad data visualization examples mentioned in this post failed at that.
To create good data visualizations, you should keep it simple. Likewise, it shouldn’t be boring.