5 Steps to Make Your Graphs (& Students) into Better Storytellers

Book Review October 12, 2021

By: Kristin Hunter-Thomson

At the heart of communicating information is telling a story. It is the same when that information are data and the medium of the story are data visualizations.

Unlike in English Language Arts and Library classes where we spend a lot of time unpacking stories, exploring their various structures, and practicing writing prose to communicate a story, these story aspects are rarely part of looking at data in a math, science, or social studies context. Similarly, we spend a lot of time unpacking images for their composition, medium, and message as well as practicing different approaches to creating visual stories in Art classes, but not these aspects are rarely part of looking at data visualizations in math, science, or social studies classes.

Part of this is due to the content focus of the courses as well as the reality that there are only so many hours in a given school day let alone year. And this completely makes sense.

But there are some consequences to not integrating storytelling and design aspects of working with data into math, science, and social studies classes that limit students’ growth in overall data and graph literacy skills. For example,

• Understanding that any and all data visualizations are made by a human to communicate a particular story or message is a foundational skill to developing the subsequent skill to stop and consider who made the visual and if they have any additional motivations for a particular storyline.

• Or, understanding that how we construct a data visualization in terms of its style, color, annotation, etc. influences what someone takes away from the story is critical to developing the skill of constructing visualizations that clearly communicate a message AND the skill to read the various pieces in a visualization to interpret the story.

I am NOT suggesting that we dive into story arcs and aesthetics of data visualizations every time we use data in math, science, or social studies classes. But I am strongly suggesting that we at least do it sometimes as a way to help our students build their data and graphical literacy over time.

So, where to start? Fortunately, there is a whole field of Data Visualization to help us out!

I would highly recommend Cole Nussbaumer Knaflic’s Storytelling with Data: A Data Visualization Guide for Business Professionals as a great place to start. And yes, I know it says “for business professionals”…but the lessons are broadly applicable to any work with data, and the business side can be more relevant than you may imagine for communicating assessment data to your administration or School Board.

Throughout the book Knaflic breaks down and explains in tangible details 5 steps to better communicate your story with data. I have highlighted each below with some suggested connections to teaching with data:

1. The importance of context: Depending on the to whom, what, and how the ways in which we communicate a story are influenced. Spending time to think about who, what, and how before making your data visualization can be the difference between success and a waste of time. How can we integrate this into the classroom? The next time your students are making a graph for a project or lab report, ask them first to write down a response to this prompt: “If you had only 3 minutes to tell your audience what they need to know, what would you say?” (p.30) Make sure they have an identified audience, preferably not you as their teacher but rather an authentic (even if hypothetical) audience. Also encourage them to use their response when considering how to create their graph.

2. Choosing an effective visual: What data visualization type we use is dependent upon what data we have and what question we are asking (or said another way, what story we are trying to tell). In fact, the type of visualization we use can help or hinder our abilities to communicate the story regardless of how well thought out executed it is. How can we integrate this into the classroom? Students need to learn when to use different visualization types just as much as they need to learn how to make those visualizations. Integrate graph choice into their work when they create their own visualizations as well as when they are reading those made by someone else.

3. Clutter is your enemy!: Any ink (e.g., line, color, text) on a data visualization takes time for the reader to see, process, and make sense of. Leveraging that brain power towards understanding your story is key. Therefore, removing any extra ink that does not help with understanding your story is also key. This doesn’t mean you need to be a strict minimalist, but it does mean that you should review and consider what ink is there purposefully. How can we integrate this into the classroom? Help your students see that the default output from a software platform or app is RARELY the best way to construct a visualization to communicate their story. The computer does not know their data or message they want to communicate. Instead, empower your students to actively choose what to keep and what to remove when making their data visualizations.

4. Focus your audience’s attention: Part of leveraging brain power of your reader to understand your story is making sure unnecessary things are not included, but another key part is strategically including things that will help the reader understand the story. Preattentive attributes are things that our eyes and brains cue into without us consciously being aware of it. Data visualizers use that to their advantage. How can we integrate this into the classroom? Looking at visualizations that others have made can be a quick way for students to identify what aspects of the visualizations are effective and noneffective. Seeing what works in someone else’s graph makes it easier to mimic that in your own visualizations.

5. Think like a designer: “Form vs function” is an expression that can be applied to data visualizations as well. If the form is how you create the visualization, the function is what you want the audience to do with the story. How can we integrate this into the classroom? This one can be trickier for students to think about, as we usually ask students what the story (claim) is from the data but rarely ask them what someone should do with that information. But that is NOT how data visualizations are made in the real world, there is always a motive for readers to do something with the story. Having students look at data visualizations from a range of sources and thinking about what the motive was of the data visualization designer can be a way to help students start to develop this skill.

As a note, while these steps can be used at any point in working with data, they are truly most effective and time efficient when you are doing Explanatory Data Analysis (you have some specific you want to explain or show from the data).

So, take a look and see what you can learn from Knaflic’s suggestions across these 5 steps. Enjoy!

LOOKING FOR MORE RESOURCES?

• Listen and subscribe to Knaflic’s Storytelling with Data podcast, ~1-2 new episodes a month in which she tackles various data visualization related topics and interview others in the field of data visualization

• Watch the mini-lesson on Exploratory vs Explanatory Data Analysis

• Explore the Graph Type Matrix Resources to find suggested ways to help students develop their graph choice skill

• Join the Data Literacy Series workshop for strategies to integrate into any K-12+ classroom

  • #2 Create & Iterate Data Visualizations - graph choice

  • #3 Identify Patterns & Relationships in Your Data - pre-attentive attributes

  • #7 Communicate with Data in Multiple Ways - removing clutter

• Check out some suggested ways for Making Effective Data Visualizations

• Join the Not Just Bar Charts: Making Better Graphs workshops to dive deeper into graph choice, best practices, and storytelling