The most common mistakes made by novice data instructors and how to correct them

When you create a data story, it's easy to get lost in the details, and you can't create a data story that inspires others to do it. Here are the most common mistakes made by novice data instructors and how to correct them. These techniques come from the "thinking like a data narrator" seminar.

1. The data story is not suitable for the audience

Not all listeners are the same, not all listeners have the same goal. Even if you look inside your own team, consider how a technical support specialist and an operations executive can have different perspectives. Although both have a common goal of serving customers, each has a different perspective on the methods and reasons that this goal can be achieved.

Many lecturers who present data stories only create a set of slides that are too broadly targeted. In these stories, the audience may not understand what will happen next, or may not understand the problem. So be sure to think carefully, what questions your audience may have and answer these questions through the story.

2. The data supports a different version of the story.

Many times when researching data, researchers allow deviations to spread. Deviations may stem from the careful selection of specific data in order to get the results we want. In the article "The chaotic relationship between drunkenness and cancer," the authors say that previous studies showing that alcohol is healthy have been affected by biased choices. The original data in this article mixes healthy drinkers with those who are sick and drinkers. The authors describe them as non-drinkers compared to moderate drinkers. In fact, these “non-drinkers” are former alcoholics or The disease can no longer be composed of people who continue to drink alcohol, and these people are usually more ill than healthy moderate drinkers. This result proves that the study has reached the wrong conclusion.

When making a data story, make sure you are using the right statistical techniques and let others rigorously review your conclusions.

Figure 1 Woman with a sticky note on the wall

3. The lack of focus on the data story

For a data story, the information must be simple in order to effectively facilitate the audience. Novice data instructors mostly use multiple pieces of information to form a complex story. The audience will become confused and even frustrated.

In 1854, Dr. John Snow used a simple message to tell a data story: "We can stop the cholera epidemic by turning off the polluted Broad Street pump." This is not only a simple message, but also a viable information.

Make your information simple and focused. Think about it, "What do I want the audience to do after reading my data story?" Then process your information around that element.

4. The data story is full of data

Our brains are more sensitive to stories than data. In the book "Made to Stick," author Chip and Dan Heath mentioned a study that asked students to present a one-minute persuasive speech to classmates. On average, each presentation contains 2.5 data. Only one speech contains a story.

Ten minutes later, the researchers asked the students to take out a piece of paper and write down the information they remembered. Only 5% of the students remembered the data; 63% of the students remembered the story. Think about it - this story is what the audience can keep. This is because the story can inspire our emotions and enable us to successfully recall the story.

When it comes to data stories, try to limit your statistics to the most important aspects. Any statistics you use should be excellent and helpful for narrative. Please carefully select those parts that have a significant impact.

5. Proper data visualization methods are ignored

Many data storytellers are not proficient in the way data is presented. As a result, information becomes confusing and confusing in inappropriate ways of expression. You can compare this to someone who speaks two languages ​​at the same time, but you only understand the language you know.

Be sure to understand how to code variables before making a data story. Study data visualization techniques or work with experts who can help you successfully complete your presentation.

Think about the story told by the late Professor Hans Roslin in his TED talk, or the story he wrote for BBC4 called "The Joy of Stats." He skillfully used a chart to simplify 120 numbers into simple information about health and wealth.

6. Lack of fascinating narrative

When the data shows unexpected results, the audience will be very excited. But if your data story doesn't show any new conclusions, then the audience will lose interest and have no incentive to try.

Steve Levitt and Steve Dubner, authors of Freakanomics, found unexpected answers in the data. In one study, they found that low-level drug dealers might not make more money than working in fast food restaurants. This discovery is surprising because most people believe that drug trafficking can bring luxury life. However, in every podcast, they tell a compelling story.

If your data doesn't even make you say "wow", maybe your narrative is wrong.

7. The lack of "people" in the data story

People are interested in the stories of others. If you have learned about the great stories of history, then you will know that all of these stories are about how someone successfully handles adversity. "Someone" may be an individual like Luke Skywalker or King Lear, or it may be a group, such as how the army defeats the enemy. The story helps us understand how life changes and why it changes. These stories can teach the audience how to deal with objections, how to face difficult decisions, and how to take the best action.

Our brains focus on the stories of others. In Roslin's "Numbers are Boring, People are Interesting" speech, he used his own examples to encourage listeners to change their views on the family. He tells the story of his own career goal and his childhood at home. His topic reveals a secret to telling a good story - this is a story about people, not a story about numbers.

Your key point

You may repeat your data story multiple times, just like writing a novel. It doesn't matter, this is part of the process, every time there will be promotion