Share your ideas for the JMP Scripting Unsession at Discovery Summit by September 17th. We hope to see you there!
Choose Language Hide Translation Bar
Staff (Retired)
Stephen Few's New Book Is a Must-Read

“Now You See It” is data visualization expert Stephen Few’s new book, explaining how to use simple visual techniques for quantitative analysis. In this textbook-sized offering, Stephen explores one of the more overlooked aspects of analysis: the graphic representation of information.

Stephen lays the foundation for good visual analysis in Part I by defining how we perceive information. He states, "…there are ways to visually display data that are effective because they correspond naturally to the working of vision and cognition, and there are ways that break the rules and consequently don’t work. If we wish to display information in a way that will enable us and others to make sense of it, we must understand and follow the rules.”

He goes on to explain that we "…perceive several basic attributes of visual images pre-attentively, that is, prior to and without the need for conscious perception."

Below is a list of those pre-attentive attributes that are quantitatively perceived in and of themselves, without having values arbitrarily assigned to them:

Length – longer =greater

2-D Position – offset higher/lower, or left/right=greater

Width – Wider=greater

Size – Bigger=greater

Intensity – Darker=greater

Blur – Clearer=greater

These pre-attentive attributes help us all consume and digest information and make sense of their meaning. However, Stephen points out that "pre-attentive symbols become less distinct as the variety of distracters increases. It is easy to spot a single hawk in a sky full of pigeons, but if the sky contains a greater variety of birds, the hawk will be more difficult to see."

Good visualizations not only take advantage of the pre-attentive attributes mentioned above, but they also use them appropriately while considering the limitations of our visual memory.

Stephen explains that we have both "working memory and long-term memory. Working memory stores information only briefly. Working memory is where information resides when we are thinking about it. If we think about it long enough, it will end up in long-term memory."

And visual memory (which is part of working memory) is very limited. Visual memory processes information in “chunks.” How much is a chunk of information? Well, it depends on how it is conveyed. Information chunks have to be relatively small when they consist of text or numbers. However, they can be larger when served up graphically.

Part II, which is the meat of the book, takes the reader through several different analyses and shows examples of good visual techniques to best convey the information used in each one.

Prediction Profiler in JMP, which is used in the new book Now You See It by Stephen Few

Part III covers promising new trends. Here, Stephen discusses “Illuminating Predictive Models,” and this is where JMP is prominently featured.

Noting that his book largely focuses on analysis of existing information, or descriptive statistics, here he highlights the benefits of predictive statistics. "If we understand the past well enough to describe it clearly and accurately, we can often build a model that we can use to predict what will likely happen as a result of particular conditions, events, or decisions in the future," Stephen writes.

He explains that "the goal of predictive analysis is not to produce certainty about the future, but to reduce uncertainty to a degree that enables us to make better decisions.“

He then describes what he refers to as “transparent predictive models.” And it is here where Stephen uses JMP to explain how transparent predictive models help us make better decisions with less risk. He says, “This level of involvement in the analytical process [using transparent predictive models] takes advantage of our brains in a way that throws open the windows to insights that we might never otherwise experience.”

Many of the visualizations discussed in Stephen’s book are available in JMP. But even more importantly, JMP’s visualizations also incorporate world-class analytics. This is why, when Stephen turns his attention to more advanced analytics, like predictive modeling, JMP is prominently featured.

“Now You See It” is a must-read for anyone who needs to explore and understand data that guides his or her decision-making process. Following Stephen’s advice will help readers explore their data better, discover trends and patterns more quickly, and make decisions with confidence.

Article Labels

    There are no labels assigned to this post.

Article Tags

Martin Owen wrote:

I've not yet read "Now you see it", but I have just finished reading Stephen Few's previous book "Information Dashboard Design".

This is an outstanding read - both pragmatic and thought-provoking. I immediately deployed Stephen's advice and used these techniques to redesign an existing dashboard to provide both added value and clarity. And once you are aware of the 13 common mistakes in dashboard design you see appalling examples of these everywhere, standing out like sore thumbs.

Being good with statistical tools is not enough.

Who needs to know the output?

What do they need to know?

How much do they need to know?

Why do they need to know?

If you want to make an immediate impact on sifting out what's most important and then getting across complex data effectively I strongly recommend this book.

Martin Owen


Stephen McDaniel wrote:

Chuck, great post about a great book.

I think JMP fits a middle ground between SAS and R that has been overlooked.

While SAS 9.2 has made significant improvements around visualizing analytics, JMP has a much higher level of interactivity due to the nature of the product. While R has often been prominently touted by academic and theoretical statisticians for great visualizations combined with analytics, I personally think JMP is a much more productive environment than R for the applied statistician. I especially think this is true in a corporate settings where multiple projects are often attached in short periods of time. Finally, JMP has made bold leaps forward in interactive visualization, which I applaud.

In my opinion, JMP is the best interactive tool that combines advanced analytics with visualization. I especially appreciate the data mining algorithms and interactive statistical modeling capabilities. SAS scales better, but JMP is quite simply fast, flexible and powerful for advanced statistical visualization.


Stephen McDaniel

Co-founder and Principal, Freakalyticsâ ¢ LLC

Rapid Analytics to Explore, Understand, Communicate & Actâ ¢