|
|
|
Books
|
How to Lie with Charts - Second Edition - Reader Support
What's New in the Second Edition?When the first Sybex edition of How to Lie with Charts went out of print in 2000, it was republished as a print-on-demand book by iUniverse. That kept the book on the market and selling steadily, even though it hadn’t been updated. A drawback was that, in those early days of POD, master files for digital printing were created by optically scanning the pages of the existing book. As a result the print quality was rather fuzzy, and the reproduction of the illustrations, particularly the gray halftones, was poor. For this second edition, we’ve gone back to the original digital files (designed in Aldus PageMaker, another program of yesteryear), converted it all to Adobe Prepress PDF, and in the process made many corrections, updates, and clarifications. Readers will find the most substantial changes in Chapter 11, where the original emphasis on color design for different output media (such as slide film and overhead transparencies) has now shifted to consideration of room lighting conditions for video display or projection (dark versus bright rooms). The discussion of color palettes, which follows closely the way color was implemented in Freelance Graphics, might seem dated since PowerPoint doesn’t give the user as much access to these choices. However, on closer examination, the principle of palette-driven color schemes is essentially the same as in the Cascading Style Sheets (CSS) that are so much a part of Web design, especially when used in conjunction with “format-agnostic” XML. The introduction to the first edition, “Truth Is the Best Revenge,” is nowhere to be found in this book. It had a tongue-in-cheek literary style, a hymn to the demise of graph paper, and a sense of humor that may have died with the comic novel. The inordinately curious can still read it here. Introduction to the Second Edition: How to Lie with PowerPoint (excerpt)A publisher once told me that I would sell 10,000 more copies of this book if I were to put the trademarked name of the popular Microsoft slide presentation software in the title—perhaps even more if we were to point a finger at Excel. And a well-intentioned reviewer of the first edition wrote: “A better title would be How to Avoid Being Fooled by Charts and How to Prepare Clear Informative Ones.” Hmmm. Honest, if not catchy. Then a newspaper reviewer took me to task for teaching people how to lie. But he should have accused me of theft. I borrowed the title from Darrell Huff’s How to Lie with Statistics, which has endured mightily since it appeared in 1954. (And then Mark Monmonier did How to Lie with Maps.) This LaPuerta publication is the second edition of How to Lie with Charts, which was first published by Sybex in 1995. Despite the fact that the first edition had a few embarrassing errors (yes, lied in its own way), it has become something of a classic among math teachers and business communicators. I assumed this reputation had preceded me when, years later, I met one of the new Sybex editors at an authors’ conference. As he shook my hand, he commented that he admired the book. Swelled with pride, I asked him if perhaps one of the senior editors at the publishing house had recommended it to him. “No,” he replied. “We studied it in school.” Can I be that old? Would you think I were lying if I claimed, like Mozart, to have turned pro at age four? What's the Problem?Are you worried about using PowerPoint? You should be. However, you can’t beat the rap by blaming the computer. Oh, don’t get me wrong—the ability to generate charts and graphs at a click does compound the crime. Why? Because if you had to laboriously lay out your drawings using the ancient tools of graph paper, ruler, and pencil—you’d have time to think about what you were doing. But when the process is automated, many of the crucial design decisions happen behind the screen. Isn’t it marvelous digital magic? Sure. But when did you ever hear of a magician who was honest? What, again, is the problem, exactly? If you are using PowerPoint or Excel or any other computer application to make charts for an audience—it is almost certain that you are lying to them in some way. So I don’t need to teach you how to lie. You’re doing it already. But the problem is more serious than that. You don’t know you’re doing it. And neither does your audience. So you’re getting away with it— until a manager or an investor or a sales prospect makes a bad decision based on the information that you were so helpful to provide. But What about PowerPoint?The main focus of this book is on the principles of visual communication. It’s about careful thinking and clear expression. Computer technology is a side issue. If you’re still worried about using PowerPoint, I should back up my claims. By way of introduction to the underlying reasons, here are just three ways that you and your computer helpmate can make downright unreliable charts. 1. Start by loving ExcelAnd who doesn’t? Life without spreadsheet software would be like an eternal morning without coffee. PowerPoint’s Datasheet window is an embedded Excel object. You can also import data from external sheets and databases. Your chart is no more accurate than the numbers from which it’s generated. We’re now living at a time when many young financial managers never did any manual calculation after high school (and maybe not even then). Unfortunately, students and managers alike are more concerned with getting results quickly than questioning the underlying logic of their sheets. And even if the logic is sound, inaccuracies from truncation or rounding in formulas or as a result of number formatting can be multiplied with every update (especially when a formula uses the results of another formula). It’s all the more likely in a sheet that gets updated frequently—daily reports, say. Through iteration, margins of error can become significant—enough to matter. The remedy? Learn to analyze the logic in complex sheets. And apply tests of transactions before you use the sheet on live data to generate charts. Oh, yeah—everybody knows that. But hardly anyone does it— life’s too short. 2. Make everything 3DWhen you get around to selecting a chart type in PowerPoint, if you want to mislead or distort, choose dimensional pies or bars. It’s virtually impossible to make a 3D chart that doesn’t give a wrong impression. In a 3D pie chart, the slice with the thickest edge will always seem larger than its true percentage. (Chapter 2 will tell you why.) In a 3D bar chart, it’s a challenge to find a rotation at which the short bars are still visible but the tall bars aren’t distorted by the perspective. Judging the heights of bars in a perspective view is a real challenge. So your audience can’t reliably estimate values (do visual take-offs) from your dimensional bars. (See Chapter 4.) Unfortunately, 3D bars are now the staple of the wildly popular “dashboard” displays of executive information systems. Managers are actually making operational and strategic decisions based on this type of reporting. It’s scary. 3. Heed Rothman’s Rule: “Don’t show the don’t-knows”I named this rule for British economist James Rothman, who preaches vehemently against following it. What he means is, in a data series, if you have some leftovers you’re tempted to label “Miscellaneous,” “All Others,” or “Unknown”—it’s so much tidier just to leave them off. Showing them is not only messy but it also invites questions about why you can’t be more specific. In a bar chart, it’s a negligent practice. But in any pie chart, it’s an outright lie. A pie chart, by definition, must show 100 percent of the data set. (Read more in Chapters 3 and 4.) Furthermore, if you are unsure of the value of the Unknowns, the relationship among all known items is almost useless. Since each pie slice is some percentage of the total, none of their sizes will be reliable if you don’t know what the total is! Yes, you know how the knowns relate to one another, but you have no clue how significant any of them is in relation to the whole. Be warned: There is no built-in control in PowerPoint or in Excel or in any other computer charting application that will force you—much less, remind you—to account for 100 percent of the real-world data in a pie, or in any other, chart. If you’re ready to swear you’d never resort to this low trickery, think about what programmers call the “unk-unks,” the unknown unknowns. Whatever the situation is that you’re trying to depict, I’ll bet you a trip to the snack bar there’s something about it you don’t know. If you think there isn’t, you’re lying to yourself. Introduction to the First Edition: Truth Is the Best Revenge (excerpt)If you feel a twinge of guilt as you pick up this book, don't worry- you are among friends. I admit that the title is provocative, promising a tantalizing debasement of moral values, at least in the realm of business intercourse. But don't be ashamed that you are tempted to look behind the peepshow curtain. We have all been there, or wanted to. Make no mistake: The promise of the title is not false. In these pages can be found the potent means to work serious mischief. Call me an optimist, but I have a better opinion of your motives. I can think of several legitimate-even honorable-reasons for your wanting to know how to lie with charts, and I like to think those are the real reasons I wrote this book for you. For the moment, then, let's assume that you're not a shameless, unprincipled liar who will stop at nothing in your frenzied scurry to the top of the heap. What is there for you here? You may have been drawn to this book because you feel, as most of us have at one time or another, that you have been lied to. Whether you are a manager being presented with a suspiciously rosy sales forecast or an investor being enticed with a pretty addition to your portfolio, you could be easy prey for seductive chartmakers. Learning their nasty tricks is one way to even the odds, if not the score. Another indication that you need to get acquainted with the tricks of the liars' trade stems from a deep-seated fear of lying. When you are the one poised to present, you don't want to get caught with your pants down (unless that's your deliberate plan). In this litigious era, an xv overdose of caution might be downright healthy. Therefore, go ye and study the liars, that ye may abjure their ways! Let's posit a Golden Rule of the Information Age: Don't show unto others what you don't want shown unto you. Beyond your own sense of conscience, this crack in your confidence, there's the yawning chasm of public ignorance. For the most part, we are a society of trusting illiterates, where charts are concerned. And the situation is getting worse. Why? Like so many complexities of postmodern life, computers are at the root of the problem. Charts have become the lingua franca of the information age, and you would think that would be a boon to communication. After all, charts are prettier than dreary tables of numbers. And, these days, those pretty pictures are so easy to make! Gone is the tedium of the careful draftsman, especially now that we have pint-sized computers popping up all over the place in our homes and offices. Thanks to the developers of graphics software, it has become a trivial task to create charts of all kinds-in full color, no less. These days, if your work routine includes charting, it just doesn't make sense to do without the labor-saving miracle of digital personal computing. And it's probably just as well that charts can be made so effortlessly, because otherwise the only people who would be interested in looking at the numbers would be the privileged few who make the decisions, and that strikes me as decidedly undemocratic! It's been almost two centuries since Charles Babbage and Lady Ada Lovelace tried to invent the computer using a steam engine, but, all things considered, it has been worth the wait. We are fortunate that electronics technology came along just in time to finish the job, that the nuclear arms race stimulated the development of high-performance computer displays, and that the far-sighted wizards of Silicon Valley responded so capably to the marketplace demand for computer games from a generation called X that is fast maturing into a new breed of video-junkie financial analysts who want their information on the screen, in color, and fast. Thanks to these happy accidents of history, we can now create charts literally at the press of a button. But I wonder. Although I'm not ready to totter off with the likes of my fellow curmudgeons, I am old enough to dimly remember a time when we were not so fortunate. I have a vague recollection of peering out through the bars of my crib to watch my honorable forebears labor mightily with pencil, ruler, and graph paper to eke out a meager living on the dusty plain of a world that had yet to invent cheap electronics. (I always knew I had very special parents.) As I look back on it, I wonder if there wasn't a greater wisdom in those primitive efforts. Having such crude tools might have forced those early chartmakers into slower thought processes. It is conceivable that they actually pondered carefully the composition-maybe even the content!-of those pathetically simple charts and graphs. Can it be that in their technological poverty they achieved a higher level of consciousness? Did they actually come to grasp the meaning of their graphic creations? As has happened again and again throughout human history, in gaining new knowledge we have had to shed the innocence of the past, which once enfolded us like a protective garment. We now stand naked before our computer monitors (at least until two-way video comes into general use). Empowered as the new gods of cyberspace, we have been granted the ability to devour megabytes of data at a single gulp, digest it in mere microseconds, and spew it forth without further thought as visually stunning color imagery. My fear, though, is that the wisdom of old has disappeared in that brief, electronic zap! that transforms raw data into those pretty pictures. The good news (and there will be plenty of it if the Gaia Effect proves to work as advertised) is that, having generated this visual information so thoughtlessly, we can change it in a blink of the other eye. I harbor the hope, then, that there is just time to shape the wet clay of our understanding about charts before it solidifies into something that future generations will not be able to grok. Now that I have you bound and gagged by virtue of your having paid your hard-earned, devalued dollars to remove this book from the opulent comfort of your favorite superstore, you may be wondering why you should heed my advice on the perils of lying with charts. Well, despite the lessons of my early childhood education, I did not set out from a young age to become a guru to chartmakers. My career, like the other happy accidents of history, seems only in retrospect to have been guided by an unseen hand. And, believe it or not, when I peered out of the bars of that crib and saw my father soldering together his first computer from an assortment of colorful resistors and other tiny components, I did not aspire to join the ranks of the digital generation. As soon as I was old enough to get my hands on his graph paper and wrap my chubby little fingers around a pencil, my thoughts turned not to drawing diagrams but to writing silly stories. As I matured, I fell victim to the economic reality that our society does not cheerfully subsidize aspiring writers. Like a disenfranchised aristocrat in a bygone age, I was forced to learn a trade. I soon discovered that, not surprisingly, technocrats would pay handsomely for magazine articles and books on technical subjects. The narrower the audience and the more arcane the subject, the higher the fee. Somewhere along the way in my budding career as a techno-apologist, I underwent a Kafkaesque transformation. I awoke one morning and found to my dismay that I had changed into a computer nerd. The genesis of my evolution was my clients' petulant demands for pretty pictures to accompany their droning speeches. As my deadlines grew ever shorter through the innate impatience of my executive taskmasters, I quite naturally yearned for a technological labor-saving solution. I happened on computer-generated slide graphics at a time when it took a roomful of shiny hardware to crunch the numbers. The price of these air-conditioned electronic behemoths would sink a small conglomerate, but fortunately the technocrats had their corporate checkbooks at the ready, because they lusted after those pictures more intently even than they wanted my carefully chosen, jargon-filled prose. So I found myself the master of a big juice guzzler called a Genigraphics computer, which took many minutes and sucked half the watts out of downtown Detroit to record a single chart on a color slide, but it was a big improvement. And it marked the beginning of the end of profitability for manufacturers of graph paper. I noticed even then that a disturbing change was taking place. In the old world of wizened artisans who labored with graph paper and the tools in their tabourets, there was time to reflect. In fact, there was a venerated ritual called a cradle review (flashback to my crib!) in which artwork pasted on cardboard panels was examined and approved by the client before the monumental expense was incurred of photographing it on color slide film. But no sooner had this Genigraphics thing warmed up than the cherished cradle ceremony- and all other comforting mechanical rituals-disappeared in that same electronic zap! The technocrats cheered, delighting in the blinding speed of our new toy, unaware that during their excitement the god of digital imagery had stolen their understanding of their own charts. Now that electronic microminiaturization has shrunk the Genigraphics machine to the size of a cuddly small rodent, the sagacity of the pet owners shows no signs of improving. These days, you don't need custody of a million-dollar silicon brain to do the job. Anybody with a credit card who knows the way to the local computer store can command the tools of the chartmaker's craft. It's time that I spoke out, and strongly: These people are lying to you. They are lying to themselves. And if you are doing it, you're both getting it and giving it back. And the shame of it is that there's so little malice involved. For the most part, people don't know that they are telling shameless lies with their gorgeous, new charts. They have put their faith in the zap! Someday they will discover-perhaps not too late for the good of the planet-that their new computer god lacks a crucial skill called human judgment. This book is offered in the sincere hope that you will save its dear price many times over by avoiding the kinds of graphical gaffes that could end up costing you or your capitalist masters big bucks. But beware. This is potent stuff. An unscrupulous person could work this magic to unfair advantage. If you unwisely choose the sinister bend in your life road, you will find no comfort here. You will be on your own through the Dark Wood of Error. But if you desire instead to protect yourself from ruthless liars who somehow manage to lure you into their audience, rest assured that if you study these pages, their tricks will have no effect on you. I am surprised that the literati I so admire-Anne Tyler, John le Carré, Margaret Atwood, and the incomparable Peter De Vries (whose poor imitation you might have recognized by now)-have not already mined the rich lode of the deceptive chartmaker's craft as a plot device. Having reeducated myself in these dark arts in the process of writing this book, I now realize that I have a unique artistic opportunity. You see, there was this nefarious chartmaker who alone had the knowledge that the CEO was color-blind-strike that-visually challenged, colorwise…. Gerald Everett Jones
|
|
Send mail to info@lapuerta.tv with
questions or comments about this web site. All postings to weblogs (forums) of this site and email to the author(s) become the property of La Puerta Productions. By posting or submitting you indicate that you agree with this policy, and you represent and warrant that the content of your submission does not infringe on the rights of any third party. For Dummies®, the Dummies Man logo, Certification for the Rest of Us! are trademarks or registered trademarks of Wiley Publishing, Inc. PMP® and PMI® are registered trademarks of Project Management Institute, Inc. All other trademarks used on this site are the property of their respective owners. This site is not affiliated with or endorsed by PMI. Some copyrighted material on this site is licensed to the publishers named herein and is used by permission. |
Vacation
at "The Roost" in Sedona, Arizona