138 internautes sur 143 ont trouvé ce commentaire utile
- Publié sur Amazon.com
"Now You See It" is the latest book by author Stephen Few. The description says this book is a companion to "Show Me The Numbers," which is a favorite book of mine. "Show Me The Numbers" is about building charts and tables that will help you show others what you see in your data. "Now You See It" is about helping you to find new ways to display your data for your own analysis. Before you can show others your data you need to understand it yourself, and that's where this book fits in.
Creating charts is front and center as the focus of this book. "Show Me The Numbers" focused on charts and tables that could be built with simple tools such as Excel. Now You See It shows you the types of analysis you can do when you employee more advanced software such as Tableau and R. Some of the illustrations are really cool to look at and inspirational, even if I will never have the tools or time to prepare my own version of them.
"Now You See It" is broken up into 3 sections:
In Part 1 - Building Core Skills for Visual Analysis. Stephen Few covers the history of information visualization, the basics of analysis, and how we perceive data. There is some overlap with "Show Me The Numbers," but it's only one chapter, and not a deal breaker for me. I found the history of information visualization chapter interesting, and I imagine that in 50-100 years there are going to be new kinds of visualization methods available that we haven't even thought of yet.
In chapter 4, Analytical Interaction and Navigation, the author covers the role of good software in the data analysis process. He lists a few requirements that good software should have, and in many cases popular software such as Excel fall short. This is when you realize that learning another program like R could be useful. I almost feel like this chapter was written for software developers who are trying to create their own data analysis software, so if you're in that camp this is your book.
Part 2 - Honing Skills for Diverse Types of Visual Analysis, goes in depth with various types of charts that you can use to analyze your data. There is a chapter for each of the major types of visual analysis: Time Series, Ranking and Part-to-Whole, Deviation, Distribution, Correlation, and Multivariate.
Within each chapter Stephen Few shows you which types of patterns you should look for in your data and shows you what those patterns mean. He then shows you different ways of displaying the data, which can range from simple Excel charts to complex visualizations which could belong in a magazine. Finally he finishes each chapter with a list of best practices for analyzing the data, such as scaling chart intervals properly or using logarithmic scales to compare the percent change of data with different starting points (look at almost any stock market graph to see a logarithmic chart in action).
These chapters form nearly 50% of the book, and could be very useful reading to a student getting started with statistics, or anyone else who is not completely comfortable with numbers.
Part 3 - Further Thoughts and Hopes. The first chapter of the book opens with the history of information visualization, and the final chapters conclude with the author's thoughts on the future. As computing power gets stronger and the internet becomes more ubiquitous new innovations are in the works, and some of them are covered here.
I finished reading this book about a week ago, and at first I didn't think much of it. I already have a strong analytical background and didn't feel like I got much out of this book in terms of learning anything new. But after a few days I noticed that I starting thinking about problems differently - I started thinking about how I could present them in a visual manner, and I started sharing my simple charts with others.
I am finding that being able to throw together a chart quickly and effectively is extremely helpful for me and a great way to share results with coworkers. Despite having seen almost everything in this book before, reading it has got me thinking about using charts more to analyze data. It is also the kick I needed to start learning to do charts in SAS so I can expand my visualizations beyond what Excel can do.
The benefits of this book may not be immediately apparent like "Show Me The Numbers,", but if you give it some time to sink in I think you will start thinking of new ways to visualize your data. The charts shown by Few in this book are, for the most part, accessible to those of us in business, versus Edward Tufte who emphasizes charts created with design tools such as Adobe Illustrator. There are some examples shown in Tableau and Spotfire, which are both quite expensive. But there are also illustrations created in R, which is free. Of course if you are going to use those programs you have to learn to use them, but that will only increase your job appeal that much more.
If you work as a business analyst and are looking for practical ways to expand your knowledge and abilities, I highly recommend this book.
18 internautes sur 20 ont trouvé ce commentaire utile
- Publié sur Amazon.com
Stephen Few has put together a book that is useful as a reference but readable as well. I heard him speak at a conference and really think he has hit the mark. The book is accessible without being shallow. He gives worthwhile examples of how to combine the features of the software with good graphing techniques. As someone who has worked with many of the techniques, he provides a reason for what I did intuitively. This is invaluable for both displaying findings and training subject experts to analyze their data.
26 internautes sur 31 ont trouvé ce commentaire utile
Jock D Mackinlay
- Publié sur Amazon.com
As someone who's done over two decades of research and development on visualization technology, I highly recommend "Now You See It" for everybody - novice to expert. Stephen Few explains visual analysis clearly and conversationally. His examples are accessible, appropriate, and beautiful.
The book is well-structured. Part I focuses on core concepts, principles, and practices. It prepares the general reader for Part II, which focuses on more technical material involving specific types of analysis (time-series, deviation, correlation, etc). Part II contains practical advice that will help everyone become better at visual analysis.
I particularly like the recommendations Stephen Few has included for visual analysis techniques that should be supported by commercial systems that are helping us work with data. After all, computers are now automatically collecting data. This book teaches us how to use this data to inform our individual work and to enhance our communication with each other. I believe these are key skills that will help us improve our modern, complex world.
15 internautes sur 19 ont trouvé ce commentaire utile
- Publié sur Amazon.com
Clarity and calm are great virtues in making difficult problems seem easy. Stephen Few offers an abundance of these virtues in his book, Now You See It: Simple Visualization Techniques for Quantitative Analysis. He methodically guides readers from example to example in an orderly journey made even more tranquil by his gentle personal style of writing. The "I", "you", and "we" phrasing make for easy reading, even as the information visualization concepts get progressively more complex and potent.
The example data sets are easy to understand and the lessons of good design seem to pop up from the surface of the pages. Color is used cautiously and appropriately, with no wasteful distractions. The clean designs show respect for Tufte's data-to-ink ratio.
As early as 1965, statistician John Tukey recognized that one of the great payoffs of interactive computing was the potential for exploratory data analysis. Stephen Few reiterates Tukey's vision and then fulfills it by showing that good graphical representations "pave the way to analytical insight." Few has a potent advantage in that modern software tools enable him to show off the good and bad approaches for each concept. Successful commercial tools like Spotfire and Tableau are put to work repeatedly, while university research projects show up where appropriate. Over all, Few lays out the territory and gives us a grand tour.
Few closes with this declaration: "I love information, in part for the understanding that it offers...Mostly, though, I love it for what I can do with it to leave the world a little better off than I found it." Few proudly presents this noble aspiration to his readers in a compelling way; now it's up to us to realize this goal through the emerging discipline of information visualization.
12 internautes sur 15 ont trouvé ce commentaire utile
M. D. HEALY
- Publié sur Amazon.com
Like Tufte, Few is a well-known evangelist for clear data visualizations that yield actionable insights. Unlike Tufte, Few is more of a craftsman and less of an artist: his style is less elegant than Tufte's, but for that reason I actually find his advice has greater practical value. Tufte's graphics inspire and awe me in the same way that a Bach fugue does, but I can scarcely imagine creating anything so elegant myself. Few gives examples that I can actually emulate in my work.
PS: I have met Tufte in person, but have never met Few though I'd like to!