Seeing is Believing… or is it?

Readers rely on journalists to adequately do their jobs. There has to be a degree of trust between the reader and the journalist: you trust that the information presented to you has been thoroughly verified and checked by the professional. This is a necessity. After all, most people simply do not have the time to fact-check every bit of information they encounter. They read the newspaper in the morning, before work, in order to stay properly informed. As such, it helps if a journalist is able to effectively distillate the essence of a story. As interesting as a subject may be, not everyone is willing to read 20 pages on a story.


Visualizations can be a helpful tool for telling your story. They have the ability to reveal patterns and trends that words cannot always convey. This is especially of invaluable importance to a data journalist. After all, numbers can be quite daunting at times. Especially in large quantities. Visualizations have the ability to translate numbers into comprehensible images that illuminate the story you want to tell. A good example of this is an election map made by Texty, that visualizes the results of the regional elections in the Ukraine.


                                         Blue is pro-Russia, orange is pro-Europe.

Ukrain_MapThis image illuminates the divide that exists in this country to the reader. As can be seen, the western part of the country voted for the pro-Europe party, the eastern part voted for the pro-Russia party. An image such as this one really tells a story within itself, because it is an insightful image that illustrates the divide; a divide that, in this case, fuels the tensions in this country.

Visualizations gone wrong

However, images are not always this insightful (though at first it might appear that they are). An example is the following poll I encountered:

At first glance, it gives the impression that the confidence in Obama’s Economic Plan is steadily increasing. However, the x-axis tells you a different story: confidence in the plan is actually decreasing. The months are displayed in reverse chronological order, making it easy to be misled by this visual when taking a quick look at it. Somehow, I think that was the intention. Now follows another example of a misleading visual:

The first pie chart might look more visually attractive, but it also makes Item A look as big (or even smaller) than Item C. However, in the regular pie chart it can be seen that Item A is actually more than two times bigger than Item C.

These are good examples of what can go wrong when it is attempted to visualize data. Alberto Cairo, teacher of the fifth module of the Online Data Journalism Course, stated that good visualizations are beautiful, functional and insightful. While you can attract viewers by creating an appealing visual, what matters the most is what the visual is able to convey. It seems that the 3D pie chart was a case of style over substance, in that regard.

Avoiding visual errors

Alberto Cairo named the following four features as essential for a great visualization:

  1. Functional: The shape of the graphic is altered to fit the questions you want to answer with the visual.
  2. Beautiful: Attractiveness will make more readers want to read it.
  3. Insightful: The visual puts your data in context.
  4. Enlightening: The information revealed by the visualization shapes the perception of the reader.

You should always keep in mind that the visuals you create for a particular story are graphical representations of evidence. While a beautiful visualization might attract viewers, it is the substance that matters the most in the long run. It falls in line with my earlier blog posts, where I (hopefully) illustrated the importance of being accurate. For me, accurate data is the fundament on which you build your story. Both good visuals and good writing can help you tell your story more clearly, but the facts count the most. You can be a verbose writer, throwing eloquent and difficult words in left and right, but a well-written, inaccurate piece is still inaccurate. The same is true for visualizations. You have to choose the best graphic form that is functional, represents the evidence, and is able to answer the questions your story might bring up. For example, the 3D pie chart: it might look fancier, but it also misrepresents the evidence you want to present to your readers. It is therefore an inaccurate way to display your evidence. A normal pie chart might be less attractive visually, but at least it represents your evidence more precisely. It is easier to extract the meaning from it. And that will only make it easier for you to tell your story.


8 gedachtes over “Seeing is Believing… or is it?

  1. I think, but forget me if I’m wrong, you title has to be: ‘Seeing is Believing… or isn’t it?’ A pretty ‘cool’ (I’m not sure if that is the right word for it) example of the regional elections in Ukraine! You took a clear example of how it could go wrong as well (Obama), however, what is the reference? And in that figure, you’re talking about the x-axis, but it are the columns that showing the percentages right? I’ve wrote exactly the same about accuracy.. Good point ;). But I like your blog overall! And more importantly, I agree on every aspect!


    • “you title has to be: ‘Seeing is Believing… or isn’t it?’”
      I think, since I’m casting doubt on the principle, that ‘or is it’ is the correct form.

      “And in that figure, you’re talking about the x-axis, but it are the columns that showing the percentages right?”
      Indeed, but the percentages aren’t the problem. If you check the x-axis, you can see that the months are in the reverse chronological order. Giving the impression that the confidence is going up; instead it is decreasing if you go from January to now.

      Thanks for the feedback 🙂

      Liked by 1 persoon

  2. This post contains a message that is consistent to the content of your previous blogposts, which I find very inspiring. I too agree that accuracy of reporting should be valued in terms of data journalism and visualization. It ranks as a priority, but I also consider the way in which the story is told as an essential value of both data journalism as well as data visualization. I wonder if people pay enough attention to misleading data visualizations if the visualization does not even capture their attention to begin with. In your examples, it is of course quite obvious that the visualization is misleading and anyone who does not notice that is probably just scrolling around and wasting time. I believe that there is no excuse for misleading data visualizations, whether readers will notice it or not. Journalists owe it to the people that wish to be accurately informed and will actually go for miles to acquire accurate information (as in actually scrutinizing news content for accuracy). I think that journalists should regard all readers as the well-informed bunch in order to prevent sloppy data visualizations and storytelling from happening. Even if it cannot be true that all readers will be willing to go through the same effort. But if journalists do not regard their readers as intellectually capable of noticing misleading data visualizations, then how can they value their own work? So my point is, I think the work of a journalist through visualizations and storytelling mirrors their perceptions about the audience to whom they are writing.


  3. You made a clear point, well done! I do not have much to say, only that you are right and that in case of journalism the visualizations have to represent the accurate data. I think it is the journalist’s duty to tell a truthful story, because it is about things that are happening in the world. But, if the story tends to be more in the light of advertising instead of the traditional journalistic work, I am not really agreeing on that one. Advertisement is about persuading and convincing, which can be done by misleading people and exaggerating. I do not think that it is a really bad thing to use misleading visuals in advertisement. But like I said before, in journalistic work it always have to be accurate because people trust the writer to be truthful.


    • Advertising is a different beast indeed, one that can both thrive on misleading and not telling (is that the same as not lying?). At the same time, misleading advertising can be pointed out, which might hurt a brand. On the other hand, depending on the product, people might not really care at all. 😛

      Thanks for the feedback! 🙂


  4. In my opinion your blogs is very good. The example of the Confident in Obama’s Economic Plan visualization is very useul and original. I also liked your writing. I think you made a point about the importance of accuracy in journalism and visualizations. However, I disagree with you in the point that says that facts are the most important thing in journalism. I agree with you that facts are basic for stories, without them there is not story, but I consider that a good representation of your findings, via text or visualizations, is mandatory for creating a decent story that can be shared (the essence of journalism), otherwise your facts can get lost in the middle of a horrible text or a bad infographic. Journalism is not only about facts, but also about effective communication. There has to be harmony between text, facts and visualizations.


    • I agree! But it’s a double edged sword – something untruthful can be communicated very convincingly through flashy visuals. I guess it can go both ways. Nevertheless, good point. Thanks for the feedback 🙂


Geef een reactie

Vul je gegevens in of klik op een icoon om in te loggen. logo

Je reageert onder je account. Log uit /  Bijwerken )

Google+ photo

Je reageert onder je Google+ account. Log uit /  Bijwerken )


Je reageert onder je Twitter account. Log uit /  Bijwerken )

Facebook foto

Je reageert onder je Facebook account. Log uit /  Bijwerken )


Verbinden met %s