Signal: Understanding What Matters in a World of Noise

by Stephen Few

Teaches the analytical skills necessary to glean value from the warehouses of accumulating data

In this age of so-called Big Data, organizations are scrambling to implement new software and hardware to increase the amount of data they collect and store. However, in doing so they are unwittingly making it harder to find the needles of useful information in the rapidly growing mounds of hay. If you don’t know how to differentiate signals from noise, adding more noise only makes things worse. When we rely on data for making decisions, how do we tell what qualifies as a signal and what is merely noise? In and of itself, data is neither. Assuming that data is accurate, it is merely a collection of facts. When a fact is true and useful, only then is it a signal. When it’s not, it’s noise. It’s that simple. In Signal, Stephen Few provides the straightforward, practical instruction in everyday signal detection that has been lacking until now. Using data visualization methods, he teaches how to apply statistics to gain a comprehensive understanding of one’s data and adapts the techniques of Statistical Process Control in new ways to detect not just changes in the metrics but also changes in the patterns that characterize data.

Genres: Nonfiction, Design, Business, Bisexual

225 pages, Hardcover

First published June 1, 2015

Reviews

[Michael Scott · Rating 2 out of 5]

--- Stephen Few does not like big data. He does not like big data because he believes it corrupts people working with data. The chain of his arguments goes along the lines of refuting the strictest form of pro big data thought, where quantity is the only thing that matters, where bigger is not only better but also the quality of information is irrelevant. News: it's 2018, big data is everywhere, and still nobody is saying what Few is refuting. Much ado about nothing; skip this and you'll find a good treatment of basic graphing techniques.

(the remainder, from a previous review)

Many pluses and minuses match the other books of Stephen Few:
+/- The target audience is the layperson starting with visualizing information, not the sophisticated crowd targeted by Edward Tufte. So, a good primer, but nothing to get excited about. Also, not enough for information visualization for scientific evidence, but a good start in that direction.
+++ Good overview of what we know about information visualization. More systematic treatment than Show Me the Numbers.
+++ Excellent references: Edward R. Tufte (design of graphs and visual information elements), William S. Cleveland (design and interpretation of visual information artifacts), Colin Ware (human perception and memory model associated with visualization), John W. Tukey (statistics). Also some good references, less known: Gene Zelazny (practical guidelines on charts and slideware), Jonathan G. Koomey (high-level process from data to knowledge, Robert L. Harris (reference), Manfredo Massironi (psychology), Nancy Duarte (slideware presentations).
--/+ Reads like a good, but obvious rehash of what the others have said. Useful to traverse once.
-- Poor book design, with large imagery competing for attention with the many sections elements, and the buleted lists often obfuscating what is actually being said.

[Manuel Frias · Rating 3 out of 5]

I like Few's style - he knows how to explain complex topics in a clear and understandable way. I enjoyed reading this book as I did with his previous ones. However, I have the feeling that he has already written about most of what he covers in Signal.

The last section about XmR charts was totally new for me. I appreciate that Few even shares instructions on how to produce them in Excel!

[Roy · Rating 3 out of 5]

This book is focused primarily on data exploration and visualization, which are typically central to the first stage in any data science project, and does a very thorough job of it. The most valuable part, however, pertains to the fundamental mindset and framework with which one should be equipped before approaching data analysis, and thus the book is probably best suited for novice data analysts. In addition, the discussions about XmR charts offer some good ideas about how to set up a system to monitor for key signals.

[Glenn Burnside · Rating 4 out of 5]

Not as deep as his earlier books. I’d recommend “show me the numbers” and his “operations dashboard design” book instead of this one, but I did pick up some new ideas from this book on how to do comparisons between two measures.

[Zachary King · Rating 5 out of 5]

Simple, insightful, and practical with plenty of followup readings cited. Good stuff.

[Willis · Rating 4 out of 5]

This was the first of Few's books that I've read and I want to read his other books. This is a good book on data visualization, talking about the graphical tools that effectively communicate information. It is very accessible, not statistics knowledge required. If you read any of Edward Tufte's classic books, this one is very similar in feel and style. But Few blends in a little bit of Donald Wheeler's work into it as well. I like the counterbalancing argument to all of the hype around Big Data. The first chapter should be required reading for any one caught up in the hype. He makes the point that the broad availability of data makes it harder than ever to find the signals and that the fundamental principles in data exploration are still as relevant as ever.

[Eileen · Rating 5 out of 5]

As an engineering working in signal processing, I feel fed up with all the software engineers who always brag about big data but have limited understanding about what information is. They should first take a class on information and signal before processing the large volume of data in hand.
This book perfectly expresses my anger and give a quite fundamental but essential lesson to all those idiots.