1. Please read chapters 5 (The Gee Whiz Graph) and 6 (The One Dimensional Picture) from the book "How to Lie with Statistics" by Darrel Huff.
2. Come up with a brief reflection on the reading.
3. Find an example of a graph in the media that falls victim to one of the many errors/issues raised by the chapters (do not use the examples from the reading). Paste a screenshot, picture, or link of a graph that has errors/issues that are discussed.

THERE is terror in numbers. Humpty Dumpty's condence in telling Alice that he was master of the words he used would not be extended by many people to numbers. Per- haps we sner from a tram induced by grade-school arithmetic. Whatever the cause, it creates a real problem for the writer who yearns to be read, the advertising man who expects his copy to sell goods, the publisher who wants his books or magazines to be popular. When numbers in tabular form are taboo and words will not do the work well, as is often the case, there is cue answer left: Draw a picture. . About the simplest kind of statistical picture, or graph, is the line variety. It is very useful for showing trends, 60 m Genvmnz men 5, something practically everybody is interested in showing or lcnowing about or spotting or deploring or forecasting. We'll let our graph show how national income increased ten per cent in a year. Begin with paper when! into squares. Name the months along the bottom. Indicate billions of dollars up the side. Plot your points and draw your line, and your graph will look like this: Billion Dollars Jf'AMJJASDND Now that's clear enough. It shows what happened during the year and it shows it month by month. He who runs may see and understand, because the whole graph is in proportion and there is a zero line at the bottom for THE GEE-WHIZ GRAPH 63 comparison. Your ten per cent looks like ten per cent-an look. Simply change the proportion between the ordinate upward trend that is substantial but perhaps not over- and the abscissa. There's no rule against it, and it does whelming. give your graph a prettier shape. All you have to do is let That is very well if all you want to do is convey informa- each mark up the side stand for only one-tenth as many tion. But suppose you wish to win an argument, shock a dollars as before. reader, move him into action, sell him something. For 22.0 that, this chart lacks schmaltz. Chop off the bottom. 21.8 24 22 21.6 Billion Dollars 20 21.4 18 F M M S ON D Billion Dollars 21-2 Now that's more like it. (You've saved paper too, some- 21.0 thing to point out if any carping fellow objects to your 20.8 misleading graphics.) The figures are the same and so is the curve. It is the same graph. Nothing has been falsi- 20.6 fied-except the impression that it gives. But what the 204 hasty reader sees now is a national-income line that has climbed halfway up the paper in twelve months, all be- 20.2 cause most of the chart isn't there any more. Like the miss- 20.0 ing parts of speech in sentences that you met in grammar F M M S D classes, it is "understood." Of course, the eye doesn't "un- That is impressive, isn't it? Anyone looking at it can just derstand" what isn't there, and a small rise has become, feel prosperity throbbing in the arteries of the country. visually, a big one. It is a subtler equivalent of editing "National income rose Now that you have practiced to deceive, why stop with ten per cent" into ". . . climbed a whopping ten per cent." truncating? You have a further trick available that's worth It is vastly more effective, however, because it contains a dozen of that. It will make your modest rise of ten per no adjectives or adverbs to spoil the illusion of objectivity. cent look livelier than one hundred per cent is entitled to There's nothing anyone can pin on you,THE GEE-WHIZ GRAPH 05 And you're in good, or at least respectable, company. honest red line that rose just four per cent, under this Newsweek magazine used this method to show that caption: GOVERNMENT PAY ROLLS STABLE "Stocks Hit a 21-Year High" in 1951, truncating the graph at the eighty mark. A Columbia Gas System advertise- ment in Time in 1952 reproduced a chart "from our new GOVT. PAY ROLLS UP! GOVT. PAY ROLLS STABLE! Annual Report." If you read the little numbers and an- alyzed them you found that during a ten-year period $20. 000,000 living costs went up about sixty per cent and the cost of 20 gas dropped four per cent. This is a favorable picture, Millions of dollars but it apparently was not favorable enough for Columbia Gas. They chopped off their chart at ninety per cent ( with no gap or other indication to warn you ) so that this $19,500,000 was what your eye told you: Living costs have more than tripled, and gas has gone down one-third! Steel companies have used similarly misleading graphic Dec Sept methods in attempts to line up public opinion against 1937 1937 wage increases. Yet the method is far from new, and its impropriety was shown up long ago-not just in technical Collier's has used this same treatment with a bar chart publications for statisticians either. An editorial writer in newspaper advertisements. Note especially that the in Dun's Review in 1938 reproduced a chart from an middle of the chart has been cut out: advertisement advocating advertising in Washington. 3,205,000 est. D. C., the argument being nicely expressed in the head- 3,200,000 line over the chart: GOVERNMENT PAY ROLLS UP! The line in the graph went along with the exclamation 3,150,000 3,140,709 point even though the figures behind it did not. What they showed was an increase from about $19,500,000 to $20,- 3,100,000 3,104,729 200,000. But the red line shot from near the bottom of the graph clear to the top, making an increase of under four First nine months Last quarter First quarter 1952 1952 1953 per cent look like more than 400. The magazine gave its From an April 24, 1953, news- own graphic version of the same figures alongside-an paper advertisement for COLLIER'SM CHAPTER 6 The One - Dimensional Picture A DECADE or so ago you heard a good deal about the little people, meaning practically all of us. When this began to sound too condescending, we became the common man. Pretty soon that was forgotten too, which was probably just as well. But the little man is still with us. He is the character on the chart. A chart on which a little man represents a million men, a moneybag 0r stack of coins a thousand or a. billion dollars, an outline of a steer your beef supply for next year, is a pictorial graph. It is a useful device. It has what I am afraid is known as eye-appeal. And it is capable of be- coming a uent, devious, and successful liar. The daddy of the pictorial chart, or pictograph, is the 66 can unnu vmm'nu r In; A un- \"I ordinary bar chart, a simple and popular method of repre- senting quantities when two or more are to be compared. A bar chart is capable of deceit too. Look with suspicion on any version in which the bars change their widths as well as their lengths while representing a single factor or in which they picture three-dimensional objects the vol- umes of which are not easy to compare. A truncated bar chart has, and deserves, exactly the same reputation as the truncated line graph we have been talking about. The habitat of the bar chart is the geography book, the cor- poration statement, and the news magazine. This is true also of its eye-appealing offspring. Perhaps I wish to shew a omnparison of two guresthe average weekly wage of carpenters in the United States and Botundia, let's say. The sums might be $80 and $30. I wish to catch your eye with this, so I am not satised merely to print the numbers. _I make a bar chart. (By the way, if that $60 gure doesn't square with the huge sum you laid out when your porch needed a new railing last summer, remember that your carpenter may not have done as well every week as he did while working for you. And anyway I didn't say what kind of average I have in mind or how I arrived at it, so it isn't going to get you anywhere to quibble. You see how easy it is to hide behind the most disreputable statistic if you don't include any other information with it? You probably guessed I just made this one up for purposes of illustration, but I'll bet you wouldn't have if I'd used $59.83 instead.) 68 now so in: wrrn STATISTICS DOLLARS QER WE'HK There it is, with dollars-perweck indicated up the left side. It is a clear and honest picture. Twice as much moneyr is twice as big on the chart and looks it. The chart lacks that eye-appeal though, doesn't it? I can easily supply that by using something that looks more like money than a bar does: moneyhags. One moneybag Americans wage. 01' three for the Rotundian, six for the American. Either way, the chart remains honest and clear, and it will not deceive your hasty glance. That is the was an honest Photograph is made. That would satisfy me if all I wanted \"as to communi- cate information. But I want more. I want to say that the American workingman is vastly better off than the Rotun- 'l'El: ONE-DMNSIDNAL Film 59 dian, and the more I can dramatize the dierence between thirty and sixty the better it will be for my argument. To tell the truth { which, of course, is what I am planning not to do), I want you to infer something, to come away with an exaggerated impression, but I don't want to be caught at my tricks. There is a way, and it is one that is being used every day to fool you. 1 simply draw a moneybag to represent the Rotundian's thirty dollars, and then I drew another one twice as tall to represent the American's sixty. That's in proportion, isn't it? Now that gives the impression I'm after. The American's wage now dwarfs the foreigner's. The catch, of course, is this. Because the second bag is twice as high as the rst, it is also twice as wide. It occupies not twice but four times as much area. on the page. The nnmlJers still say two to one, but the visual impression, which is the dominating one most of the time, says the ratio is four to one. Or worse. Since these are 70 HOWTOIJEWITHSTATIHKS pictures of objects having in reality three dimensions, the second must also be twice as thick as the rst. As your geometry book put it, the volumes of similar solids vary as the Cube of any like dimension. TWO times two times two is eight. If one moneybag holds $30, the other, having eight times the volume, must hold not $60 but $240. And that indeed is the impression my ingenious little chart gives. While saying "twice,\" I have left the lasting impression of an overwhelming eight-tonne ratio. You'll have trouble pinning any criminal intent on me, too. 1 am only doing what a great many other people do. Newsweek magazine has done it~with moueybags at that. The American Iron and Steel Institute has done it, with a pair of blast furnaces. The idea was to show how the industrv's steelmaking capacity had boomed between the 19303 and the 1940s and so indicate that the industry was doing such a job on its own hook that any governmental interference was uncalled for. There is more merit in the Principle than in the way it was presented. The blast hn'nace representing the ten-million-ton capacity added in the '305 was drawn just over two-thirds as tall as the one for the fourteen and a quarter million tons added in the '405. The eye saw two furnaces, one of them close to three times as big as the other. To say \"almost one and onehalf\" and to be heard as \"three\"that's what the one- dimensional picture can accomplish. This piece of art work by the steel people had some other points of interest. Somehow the second furnace had fattened out horizontally beyond the proportion of its TEE one-numesromn mom {:1 STEEL CHANT? IDDEB me mLuau ms Adapted by scenery of rmware. neighbor, and a black bar. suggesting molten iron, had become two and one-half times as long as in the earlier decade. Here was a 50 per cent increase given, then drawn as 150 per cent to give a visual impression of unless my slide rule and I are getting out of their depth ~Over 1500 per cent. Arithmetic becomes fantasy. (It is almost too unkind to mention that the same glossy tour-color page offers a fair-to-prime specimen of the truncated line graph. A curve exaggerates the per-capita yowth of steelmaking capacity by getting along with the lower half of its graph missing. This saves paper and doubles the rate of climb.) Some of this may be no more than sloppy draftsmanship. But it is rather like being short-changed: When all the mistakes are in the cashier's favor, you can't help wonder- mg. 72 HOW TO LIE WITH STATISTICS THE ONE-DIMENSIONAL PICTURE 73 Newsweek once showed how "U. S. Old Folks Grow Older" by means of a chart on which appeared two male THE DIMINISHING RHINOCEROS figures, one representing the 68.2-year life expectancy of today, the other the 34-year life expectancy of 1879-1889 It was the same old story: One figure was twice as tall as the other and so would have had eight times the bulk or weight. This picture sensationalized facts in order to make a better story. I would call it a form of yellow journalism. The same issue of the magazine contained a truncated, or gee-whiz, line graph. THE CRESCIVE COW 1515 AFTER BL 1860 1936 There is still another kind of danger in varying the size of objects in a chart. It seems that in 1860 there were something over eight million milk cows in the United States and by 1936 there were more than twenty-five million Showing this increase by drawing two cows, 1936 one three times the height of the other, will exaggerate the impression in the manner we have been discussing. Apply the same deceptive technique to what has hap- But the effect on the hasty scanner of the page may be even pened to the rhinoceros population and this is what you stranger: He may easily come away with the idea that get. Ogden Nash once rhymed rhinosterous with prepos- cows are bigger now than they used to be. terous. That's the word for the method too