Data Visualization

The guiding principles of making informed visualizations included not misleading readers and prioritizing conveying a clear message. These speak to being mindful of how your figures may be perceived and presenting your data ethically and responsibly.

“Scientific visualization is classically defined as the process of graphically displaying scientific data. However, this process is far from direct or automatic. There are so many different ways to represent the same data: scatter plots, linear plots, bar plots, and pie charts, to name just a few. Furthermore, the same data, using the same type of plot, may be perceived very differently depending on who is looking at the figure. A more accurate definition for scientific visualization would be a graphical interface between people and data.”
— Nicolas P. Rougier, Michael Droettboom, Philip E. Bourne, Ten Simple Rules for Better Figures

Exploratory plots (for your own purposes)

Exploratory plots are just for you, they focus solely on data exploration. They

  1. Don’t Have to Look Pretty: These plots are only needed to reveal insights.
  2. Just Needs to Get to the Point: Keep the plots concise and straightforward. Avoid unnecessary embellishments or complex formatting.
  3. Explore and Discover New Data Facets: Use exploratory plots to uncover hidden patterns, trends, or outliers in the data. Employ different plot types to reveal various facets and aspects of the dataset.
  4. Help Formulate New Questions: Use exploratory plots as a tool to prompt new questions and hypotheses. As you discover patterns, think about what additional questions these findings raise for further investigation.

Explanatory plots

“…have obligations in that we have a great deal of power over how people ultimately make use of data, both in the patterns they see and the conclusions they draw.”
— Michael Correll, Ethical Dimensions of Visualization Research

Explanatory plots are mainly for others. These are the most common kind of graph used in scientific publications. They should

  1. Have a Clear Purpose: Define a clear and specific purpose for your plot. Identify what scientific question or hypothesis the plot is addressing. Avoid unnecessary elements that do not contribute to this purpose.
  2. Be Designed for the Audience: Tailor the design of your plot to the characteristics and expertise of your audience. Consider their familiarity with technical terms, preferred visualizations, and overall scientific background.
  3. Be Easy to Read: Prioritize readability by using legible fonts, appropriate font sizes, and clear labels. Ensure that the axes are well-labeled, and use a simple and straightforward layout. Avoid clutter and unnecessary complexity that could hinder comprehension.
  4. Not Distort the Data: Maintain the integrity of the data by avoiding distortion in your plot. Ensure that scales and proportions accurately represent the underlying data, preventing misleading visualizations.
  5. Help Guide the Reader to a Particular Conclusion: Structure your plot in a way that guides the reader toward the intended conclusion. Use visual elements such as annotations, arrows, or emphasis to highlight key findings and lead the reader through the data interpretation process.
  6. Answer a Specific Question:Construct your plot with a specific research question in mind. The plot should directly address and answer this question, providing a clear and unambiguous response.
  7. Support an Outlined Decision: If the plot is intended to support decision-making, clearly outline the decision or action it is meant to inform. The plot should provide relevant information that aids in making well-informed decisions based on the presented data.

The table below summarises Ten Simple Rules for Better Figures, a basic set of rules to improve your visualizations.

Rule Name Rule Description
Know Your Audience Understand the characteristics and expertise of your audience to tailor the figure accordingly.
Identify Your Message Clearly define the main message or takeaway that you want the audience to understand from the figure.
Adapt the Figure to the Support Medium Tailor the figure’s complexity and design to suit the medium it will be presented in (e.g., print, online).
Captions Are Not Optional Craft informative captions that provide essential details and context for interpreting the figure.
Do Not Trust the Defaults Adjust default settings to optimize the visual elements of the figure, such as colors, scales, and labels.
Use Color Effectively Apply color purposefully, taking into account accessibility considerations and cultural interpretations.
Do Not Mislead the Reader Avoid creating misleading visualizations and be aware of formulas to measure the potential misleading nature of a graph. There are formulas to measure how misleading a graph is!
Avoid Chartjunk Eliminate unnecessary decorations and embellishments in the figure that do not contribute to the message.
Message Trumps Beauty Prioritize conveying a clear message over making the figure aesthetically pleasing.
Get the Right Tool Choose the appropriate visualization tool (e.g., R) or chart type that best communicates the data and intended message.

TASK Read this short paper and find examples in your choice of literature of one or more of the rules in action.

TASK Put your data wrangling and visualization skills to the test and find the hidden picture in this dataset.