Good Books: How to Lie with Statistics, Darrell Huff

I think I was in fifth grade when I discovered a copy of this book on my school library shelves. I was enthralled and How To Lie With Statistics became one of the most influential books I have ever read. It taught me to be a skeptic.

Darrell Huff had been an editor for several mainstream magazines before he returned to free lance writing. In 1954 he published How to Lie with Statistics which was the single biggest success he ever had in his writing life. I saw the book a year or two after it was first published and it was already into its 8th (as I recall) printing. I believe it had fifty or more reprintings — more than a half million copies sold — before it was re-packaged by Norton in 1993 into the volume that is still selling today. It has been translated into many languages — the Chinese version was published 2003 — and called the most widely-read book on statistics ever published, which might be true.

The first four chapters introduce concepts familiar to most statistics students — biased or inadequate sampling, distorted use of the average as opposed to the mean, and so on. Later, Huff writes about “correlation not meaning causation” and the post hoc fallacy. All of this is common street knowledge now, less so in the 1950s. So statistics-smart people sometimes pooh-pooh this book — unless they teach an introductory course and then they are hoping that their students will get as much out of a semester as Huff presents in less than 140 pages. The concepts are illustrated with examples taken from advertising or promotional material and by sharp caricatures by Irving Geis.

So we learn that an organization wishing to appear progressive may hype its average salary, but in a company where the pay rates range from $2000 to $45000 a year, the average doesn’t mean much. Note how Huff/Geis illustrates the concepts of average/median/mode:

huff1

That item brings another criticism that I have seen often about this book, that it is too old and people can no longer relate to the examples from the 1940s and ’50s used by Huff. After all, no one expects only $2000 a year wages any more. These critics, obviously, are very limited thinkers.

The chapters that excited me as a kid were the ones about using graphics to lie. For instance:

huff2

Same data, different charts.

Besides charts, Huff also pointed out how pictures were used to distort fact. One of the figures below is twice the other, but the illustration for the larger has been doubled in height, which means that it is four times larger than the smaller.

huff3

There was more of this, a lot of information in a short book, but the last chapter has a valuable set of tests to apply to claims and propositions:

1. Ask “Who Says So?”
2. How Does He Know?
3. What’s Missing?
4. Did Somebody Change the Subject?
(…[W]atch for a switch somewhere between the raw figures and the conclusion.”)
5. Does It Make Sense?

I probably apply these rules to stuff I read ten or twenty times a day. Huff taught me to be skeptical. But there’s a dark side to that.

In 1965, Huff appeared before a US Congressional Committe investigating the 1964 Surgeon-General’s Report linking smoking to cancer. Huff was being paid by the tobacco lobby and he pointed out to the Committee, in entertaining fashion, that correlation does not equal causation and that post hoc does not mean propter hoc. Afterwards Huff picked up some money to write a book, How to Lie about Smoking Statistics, from the tobacco companies but it was never published. Some suggest that this was because Huff sabotaged his own manuscript or that he recognized that the evidence was against him, but I doubt it. In the last analysis, Huff was a hack, a freelance writer trying to make a buck — a Mad Man, if that is meaningful for you.

So here is the thing: skepticism is fine, but you have to remember that Holocaust deniers, climate change skeptics, and tobacco apologists all use Huff’s methods to bolster lies. The point is not to be skeptical so much, as it is to assess the evidence as well as you can. You have to be skeptical about everything, even your own analysis.

 

Advertisements

Ulfberht’s Swords

Thousands of swords from Europe’s early Middle Ages have been recovered. Many of these are too corroded to show any detail but 19th Century archaeologists noticed that some had markings on the blades. Further investigation showed that many of these had the name Ulfberht on one side and geometric markings on the other.

Drawing by Norwegian archaeologists, published 1889.

The archaeologists believed that they had discovered the work of a master swordsmith and, since these swords were often found in Scandinavia, they were thought to be of Viking manufacture. A bit later, scholarship held that Ulfberht was a Frankish name. Later still, more rigorous dating showed that the swords were made over a period of two hundred and fifty years or more — from 850 – 1100 AD. Now the theorists held that Ulfberht was the medieval equivalent of a tradename, possibly the place where the swords were turned out. Ulfberht was the sign of quality — like Porsche, one archaeologist suggested.

An example from the Netherlands dated 950 – 1000.

Iron isn’t that easy to turn into a good finished tool or weapon. It must be heated to 1500° Celsius — difficult in Europe a thousand years ago — and that’s only the beginning. The molten iron must be cooled and worked and reheated, each time resulting in a slightly different composition of iron. The goal was steel: steel that was hard, but not brittle, steel that could strike a hard object and neither bend nor shatter, steel that could hold a sharp edge.

Bog iron isn’t that difficult to find in northern Europe and it can be melted to a stage where it can be worked, although it may only be the slag ingredients, not the iron, that is truly molten. Archaeologists do this kind of reconstruction all the time, locating bog iron and melting it and forming it a bit. Probably there were a lot of swords made this way. These were not great weapons and there are accounts of men straightening bent swords with their foot in the midst of a battle.

Somehow, at the end of the 8th Century, the Franks began turning out good swords, ones much easier to produce than the old pattern-welded blades. The Carolingian monarchs tried to forbid their export so that the weapons could not be used against Frankish troops but was unable to stop the traffic in arms.

Allen Williams has examined some forty-four Ulfberht swords and discovered that the earlier-made weapons were forged from fine crucible steel possibly from Persia or Afghanistan. In this process, iron was smelted in a sealed crucible and slowly allowed to cool. The resulting steel is of good quality with enough carbon content so that its melting point had lowered and it could be finished by local craftsman into excellent swords. Many Ulfberht swords have been found east of Frankia all the way into present-day Russia, possibly along Viking trade routes. But not all the Ulfberht swords are of good quality — many, especially the later-made weapons, are brittle and might shatter when struck by the real thing. Even so, some have very decorative hilts, finely worked with silver or other inlays, so they probably were made for wealthy customers. And they are still marked “Ulfberht”.

Imitation Ulfberht with silver wire worked into the hilt and pommel.

Mind you, the markings differ slightly. The original swords are marked “+VLFBERH+T” where the “+” is a cross. Later versions have the cross after the T, or two crosses, one quite fancy, on either side of the name. This is the equivalent of those “Rollex” watches that guys try to sell you from the trunk of their car.

An Ingelrii sword from the London Museum.

So far, only a dozen or so of forty-four examined Ulfberht swords are entirely of crucible steel, though some of the knock-offs are of pretty good quality and some have crucible steel edges. There were other swordmakers who signed their work — Ingelrii, Cerolt, and Ulen, for example — but only Ulfberht, whether he was a smith, a guy who owned a shop, or a patron of the swordmaker’s art, was famous enough to attract this kind of imitation, one of history’s great trademark thefts.

Notes:

Anne Stalsberg, “The Vlfberht Sword Blades Re-evaluated”. A hundred and sixty-six Ulfberht blades are listed with geographic distribution and considerable speculation on just who Ulfberht might have been.

Alan Williams,”A Metallurgical Study of Some Viking Swords” (PDF). Williams’ paper shows the difference between original and knock-off Ulfberht swords. There are a lot of photos at the article’s end.

An Ulfberht auctioned by Christie’s. It realized more than $18000, even though it’s broken. Real or imitation? At this distance, it probably doesn’t affect the value.

A swordmaker looks at Ulfberht’s work. Here and here.

First Photo on the World Wide Web

[via vice.com]

In 1992, Tim Berners-Lee was looking for a picture to demonstrate the image-handling ability of his baby, the brand-new World Wide Web. Well, actually the Web had been around for a little while but only as a network for scientists involved with CERN. In 1991, though, it was opened to the public. This is the image that Berners-Lee chose as the first to go public:

So what is that?1 The Cernettes were a group formed from workers and scientists’ wives and girl friends at CERN. They sang take-offs on girl-group songs with lyrics aimed at particle physicists: “Liquid Nitrogen”, “Collider”, and so on. They were quite a big deal, at least in Geneva, and Berners-Lee was a fan. The group was managed by an IT developer named Silvano de Gennaro. He needed a photo for a CD cover so asked the group to pose backstage at a gig:

Berners-Lee asked de Gennaro for a digitized photo that could be uploaded to the WWW as a test. De Gennaro happened to have on hand a GIF file of the photo that he intended as a CD cover. Berners-Lee insisted that he add words — “It has to be fun!” — so de Gennaro got to work with PhotoShop 1 (that’s “one”, folks) and arched the lettering over top. The resulting image was part of an article about CERN music.

No one much noticed. Probably more people saw a Cernettes poster than the WWW image. But this bit of retro humor was the first. The next big steps in Web history — on-line commerce, for instance — followed with the development of internet porn as people discovered that they could sell digital images.

The Cernettes are calling it quits after twenty years and giving a final performance this month. The original GIF file vanished when the Mac that held it in memory died in 1998.

There is much more on this topic including video of Cernettes’ performances and Tim Berners-Lee’s cross-dressing here.

 1 I think it looks like an album cover for a side-project connected with a Francophone Slim Cessna. Something like the Lee Lewis Harlots, for instance.

Blaschka Glass Models

Leopold Blaschka was a Bohemian glass worker with a love of natural history. He moved to Dresden after the birth of his son, Rudolph, so that the boy would have better educational opportunities. Blaschka’s paying work was mostly glass eyes and test tubes until Prince Camil de Rohan became intrigued with some botanical models that Leopold had made for himself. The prince had a huge collection of plants and flowers and commissioned orchid models from Blaschka in 1860. That same year the Dresden State Museum asked if he could create glass models of invertebrate creatures. Blaschka could. Museums around Europe began commissioning models of plants and sea creatures and Blaschka soon dropped all his other work.

Rudolph joined his father in the project and the pair turned out thousands of glass models over the years. Harvard became interested in the models and, from 1887, thanks to the genrosity of the Ware family, the Blaschkas worked exclusively for that university, creating more than three thousand botanical models.

The Blaschka family.

Leopold’s first efforts were based largely on Ernst Haeckel‘s famous drawings but after dealing with Rohan, he worked more from preserved or living specimens. Rudolph took a long ocean voyage in 1892 to study plants in North America and the Caribbean.

Blaschka drawing

The models are made of layers of glass sometimes reinforced with metal wires. The colors are powdered glass painted onto the main body, then scratched or formed into leaf veins or other details. The glass was heated over and over again, something which has affected its durability.

Harvard display case for Blaschka Cacao models. (c) courtesy http://www.hmnh.harvard.edu, photo by Adam Blanchette.

The models are very fragile — the first specimens shipped to the United States were broken in Customs — and repairs sometimes need to be done.

“I don’t know how we could clean that,” sighed a Harvard curator. (c) President & Fellows, Harvard College, courtesy Blaschka archives, photo by Hillel Burger, http://www.hmnh.harvard.edu.

Before and after shots of a Paper Nautilus at the National Museum of Wales

But some of the glass seems to be deteriorating as well. Powdery white glass corrosion is visible on some specimens and, in others, parts of the models are separating. Most of the troubles seem to be from models made in the late 1880s and early 90s. Rudolph complained about the quality of glass supplied to him then and began making his own glass after this time.

[Video of Micro CT scan of above squid model done to determine how the glass is layered.]

Probably these museum models could have been done in wax and served the stated purpose as study objects, but the beauty of glass entrances everyone who sees these wonderful creations and their value today is as works of art.

Leopold died in 1895 and Rudolph carried on alone until 1938 when, at the age of 80, he said he was just too tired to continue. He died the next year. The Blaschkas never hired an apprentice and Rudolph had no children. Their methods died with them.

Notes:

“Flowers Out of Glass” by Nancy Marie Brown. A very good article that gets into the Blaschkas’ methods and attempts to restore and preserve these flowers.

Many museums have put photos of their collections on line. The Harvard collection has been exhibited in several places and a couple of short videos  have been made about it.
Also:
Dresden photos by Heidi and Hans-Jurgen Koch.
National Museum of Wales.
Cornell.

Chicken Specs

Anyone who’s ever observed chickens has seen them peck each other. It’s not the roosters that are hen pecked, it’s other hens and they may be pecked to death. This has long been a problem for chicken farmers and various solutions have been proposed.

In 1903, Andrew Jackson, jr. was granted a patent for chicken spectacles. These are, in effect, safety glasses meant to protect a chicken’s eyes. But that’s not the only vulnerable spot on a chicken. Hens will peck at another bird’s tailfeathers until they expose skin and then attack the exposed area until it bleeds. Once blood is drawn, other birds will peck at the wound until the hurt bird suffers serious injury.

So, at about the same time Louis Harwood and Joseph Haas developed spectacles for chickens — not to protect the victim’s eyes, but to prevent the aggressor from attacking. Once these spectacles are attached to a chicken’s beak, it will not peck at blood. You see, these are rose-colored glasses; the lenses filter out the color red. The glasses from Haas are hinged and fall away from the chicken’s eyes when it lowers its head so that it can still forage for red stuff. [See them in action here.]

The National Band and Tag Company sold many thousands of the Haas glasses until the 1970s when they quit manufacturing them. Now they are collectible. There is a problem with the specs though, one that caused the United Kingdom to ban them: they are attached to the bird’s head by a cotter pin run through its nostrils. I suppose that sounds cruel. Anyway, that’s one reason chicken ranchers have turned to other anti-pecking measures, like clipping the bird’s beak.

Backyard chicken-raising has become a bit of an urban fad in the last few years. Sooner or later, these city farmers are going to be faced with the problem of restraining their flock’s natural murderous instincts. Perhaps some new form of chicken specs will be introduced for their benefit.

Blue Men, Blue Women.

There are a surprising number of reports of people with blue skin. A doctor might suspect cyanosis and think such a person was about to keel over from a coronary event or perhaps that this was a person who had ingested a great deal of colloidal silver which can turn the skin blue. An ethnologist might think this was a Tuareg, a north African people who wear blue-dyed clothing from birth and whose skins become blue — at least so long as they wear traditional clothing. But there are other cases of people who are permanently blue and they aren’t having a heart attack.

When Madison Cawein, a hematologist at the University of Kentucky, heard stories of blue people living up in the hills, his ears perked up. He suspected some kind of blood condition and, in 1960, began searching the Kentucky hills for blue people. And found them. “They were bluer’n hell,” said Cawein and he began tests to discover why.

Paul Karason of Kentucky who quit taking his medication for this photo.

Cawein came across an article in Blood, the journal of hematology, that described a similar condition among some Alaskan Indians and Inuit.  The authors, E.M.Scott and Dale Hoskins, discovered that the blue-skinned people were deficient in the enzyme diaphorase causing a condition known as methemoglobinemia. Scott hypothesized that this was due to genetic factors that were exacerbated by a lack of vitamin C. Injections of methyl blue, a substance that is non-toxic and excreted in the urine, would reverse the process. Sometimes daily doses of 400 mg. of vitamin C would do the same thing, depending on the nature of the particular condition.

Cawein tested the blue people he found and, sure enough, they had low levels of diaphorase. He began injecting people with methyl blue and their skins turned pink almost immediately. Daily oral dosage would keep it that way. Some people were able to reverse the blue process with vitamin C which, effectively, repairs the hemoglobin damaged by lack of diaphorase.

The Kentucky blue people could all trace their ancestry back to a French settler named Martin Fugate. Martin had a recessive “blue gene” and, remarkably, married a young woman named Elizabeth Smith who also carried the gene. Their descendants all carried the same propensity for blue skin. It isn’t that two recessives lead to a dominant; it isn’t that simple. Having a single recessive gene can lead to a certain degree of blueness or not. There is a gradation in genetic effects that cannot always be predicted.

Blue people have been reported in other places than Kentucky and Alaska. In the 1930s a Doctor Deeny treated two Dublin brothers with vitamin C to reverse their blue tint. It may be worth noting that Elizabeth Smith was of Irish ancestry.

So the blue people can be cured. This is good news for them since, if there is too much blueness, too high a gradation of the gene, then serious medical problems may result. The blue people have brown blood which does not transport oxygen very well. And most of them felt a certain stigma. Mountain people are used to outsider mockery — think Jay Leno as Doofus with those stupid false teeth giggling about hillbilly incest and you understand why these folks are sensitive to genetic issues. John Stacy recalls that his father-in-law was blue. “All them old fellers way back then was blue.”

The only thing Stacy can’t or won’t remember is that his wife Luna was blue.When asked about it, he shakes his head and stares steadfastly ahead. It would be hard to doubt this gracious man except that you can’t find another person who knew Luna who doesn’t remember her as being blue.

 Notes:
The Blue People of Troublesome Creek” by Cathy Trost, reprinted from Science 82, available in a number of places on the Net.
This version of the Trost article includes some other abstracted articles that I drew on for this post.