Data and its analysis is a big part of the trade of performance improvement professionals. Yet data analytics is for naught if insight is not gleaned and its implications clearly communicated to others. Seeing data in visuals — graphs and tables — is, therefore, a critical aspect of the process improvement professional’s toolkit. Unfortunately programs like Keynote, PowerPoint, Excel etc. make it easy to create content but because they allow the user to decide how to display data, the quality and effectiveness of how data is displayed in tables and graphs is all over the map. Fortunately there are several authors who are working to try and raise the quality level of data presentation.
Core Reference Books
There are many books and courses on delivering presentations and on creating slides for stand-up presentations. This is not the focus of the discussion that follows. Although it is important to understand the best practices of presenting ideas and data as a speaker, I assert that one must first understand a fundamental set of skills, namely the foundational elements of conveying ideas and data effectively using graphs and tables. Therefore I omit references to books about stand-up presentation (slideshow) techniques such as “Presentation Zen” as these belong to a different discussion.
Here are some of my favorite books and authors.
Given the importance of tools such as Spotfire or Tableau, Stephen Few’s book is timely in that he not only provides timeless concepts on the most effective ways to display data, but he also shows how tools such as Spotfire and Tableau. He covers important concepts such as:
- Four core pre-attentive attributes of a visualization: form, colour, spatial position, and motion and their effect
- Effective formats for visualizing data such as tree maps
- Examples of poor visuals (and why they are weak)
- How and when to sort, filter, highlight, aggregate, zoom, annotate, bookmark
- Optimal quantitative scales
- Use of trellises and crosstabs
- Proper time series; using lines vs points vs bars; trend lines
- Concepts of data distributions (central tendency, dispersion etc.)
- Correlation patterns
If one had only one book to use as a guide to the proper construction and use of tables (a much under-used and mis-used device) and graphs then this is the one I suggest. A 4-day course for Executives and Process Excellence practitioners that I taught was titled “Structured Writing and Thinking” and this was a key reference. In addition to providing highly specific guidelines of when and how to use tables and various forms of graphs (as well as the big mistakes to avoid) it provides background on the cognitive elements of human eye sight and the processing of visual information in our brains.
The first course I took on information design was back in 1983 and it was taught by Edward Tufte. He still teaches and has since written several elegant books on the presentation of data in graphical form. The first of these books was this classic. It provides an inspiration and ideas on design but is less instructional than Stephen Few’s Show Me the Numbers.
This book takes the graphical principles of Few’s Now You See It and Show Me the Numbers and focuses them on the world of information dashboards. This book provides insights useful to not only how we would design software-based displays, but any type of data dashboard and in all mediums including old-fashioned paper and white boards.
A key element of Minto’s method, in terms of data visualization, is the importance of an underlying logic based on principles of MECE-ness (mutually exclusive, collectively exhaustive) and logical argument (whether inductive or deductive). While the Pyramid Principle was originally developed for word-based materials, the idea of a structured and logically coherent substructure in the visual presentation of data is equally critical. I tend to caution an over-interpretation of the inductive vs. deductive segment of the book as this sometimes causes some confusion and I also caution over-use of the Situation-Complication-Resolution pattern as not everything lends itself to this structure.
Nathan Yau’s book provides tips and tricks in the graphical display of information so often seen today in infographics and other applications leveraging the large datasets that are more commonplace today.
This book is more technical in that it shows how to create your own visualizations by writing code. Probably not for most people, I would not buy it unless you want to learn about visualization development.
This book is focused more on the theoretical underpinnings of visualization through a series of essays. Although a good book I would recommend it only if you are interested in a more academic angle.