(Roughly) Daily

Posts Tagged ‘statistics

“There are three types of lies — lies, damn lies, and statistics”*…

 

In today’s world, we are constantly bombarded with averages and medians: the average temperature in New York in April is 52 degrees; Stephen Curry averages 30 points per game; the median household income in the United States is $51,939

But the concept of taking many different measurements and representing them with one best number is a relatively recent invention. In fact, there are no historical examples of the average or median being used in this manner prior to the 17th Century.

So how did the concept of averages and medians develop? And how did the average triumph as the measurement of our times?  The supremacy of the average over the median has had profound consequences about how we understand data. In many cases, it has led us astray…

More at “How the Average Triumphed Over the Median.”

* Benjamin Disraeli

###

As we average it out, we might recall that it was on this date in 1913 that employees of the City of New York held a “Parade of Statistical Graphics,” replete with large graphs on horse-drawn floats, and a photograph with people arranged in a bell-shaped curve.  The crowd’s favorite was the float devoted to the decline in death rate due to improvements in sanitation and nursing.

 source

 

Written by LW

May 17, 2016 at 1:01 am

“Data without some sort of analysis is just noise”*…

 

The Scientific American Reference Book of 1913, compiled by Albert A. Hopkins and A. Russell Bond, is available to read on the Internet Archive… a delightful snapshot of the concerns of readers of the magazine a hundred years ago, as well as an interesting tour through infographic design strategies au courant at the time.

“The Editorial staff of the Scientific American receive annually about 15,000 inquiries covering a wide range of topics,” explained Hopkins in a preface. While the magazine had earlier published a Scientific American Cyclopedia of Receipts, Notes, and Queries (which did well, selling 25,000 copies), the book didn’t quite fill the needs of every reader; letter-writers wanted general information about the world, not just scientific formulas.

They wanted information about the Antarctic region, the Panama route, shipping, navies, armies, railroads, population, education, patents, submarine cables, wireless telegraphy, manufactures, agriculture, mining, mechanical movements, astronomy, and the weather…

The infographics are also, of course, an opportunity to compare those times to ours (e.g.: the most recent data on expenditure on public education in the U.S. suggests that “teachers salaries”– “instruction”– has fallen to 54% of the total.)

More backstory and examples at “A Treasure Trove of Awkward Early-20th-Century Infographics.” Page through the entire volume here.

* Irene Ros, Bocoup; interview, October, 02, 2014

###

As we compress, 1000 words at a time, we might send statistically-significant birthday greetings to Lambert Adolphe Jacques Quetelet; he was born on this date in 1796.  An astronomer, mathematician, statistician, and sociologist, he was instrumental in introducing statistical methods to the social sciences. From 1825, he wrote papers on social statistics, and in 1835 gained international recognition for publication of Sur l’homme et le developpement de ses facultés, essai d’une physique sociale, in which he used the normal curve (which had previously been applied to error correction) to illustrate a distribution of measured human traits about a central value–  pioneering the concept of “the average man or woman.” One of his primary foci was public health: his establishment of a simple measure for classifying people’s weight relative to an ideal for their height, the body mass index (AKA, the Quetelet index), has endured with minor variations to the present day.

 source

 

Written by LW

February 22, 2016 at 1:01 am

“I keep saying the sexy job in the next ten years will be statisticians”*…

 

Nathan Yau, the force behind the fabulous Flowing Data, was lamenting the loss of a classic resource, when he decided to take on the challenge himself:

Ever since I found out about the Statistical Atlas of the United States, historically produced by the Census Bureau, it annoyed me that there wasn’t one in the works for the 2010 Census due to cuts in funding. The last one was for 2000. Actually, the 2000 edition was called the Census Atlas, but whatever. With more data than ever, it seems like there should be one.

Maybe that’s why there’s isn’t one. Too much data, too much of an undertaking, and too many bureaucratic decisions to make.

The first Atlas, by Francis A. Walker, was published in 1874 using the data from the prior 1870 Census. Counting cover, credits, and all that, it was 56 pages.

I got to thinking, hey, I could do that. And if I did, I wouldn’t have to be annoyed anymore. So I recreated the original Statistical Atlas of the United States with current data. I used similar styling, and had one main rule for myself. All the data had to be publicly available and come from government sites

See the stunning– and stunningly useful– results (with larger versions of each chart) at “Reviving the Statistical Atlas of the United States with New Data.”

* “I keep saying the sexy job in the next ten years will be statisticians. People think I’m joking, but who would’ve guessed that computer engineers would’ve been the sexy job of the 1990s?”  – Hal Varian, The McKinsey Quarterly, January 2009

###

As we discriminate between the median and the mean, we might spare a thought for Sir D’Arcy Wentworth Thompson; he died on this date in 1948.  A classics scholar who was also an accomplished biologist and mathematician, Thompson is best remembered for On Growth and Form (1917, new ed. 1942), a profound consideration of the shapes of living things, starting from the simple premise that “everything is the way it is because it got that way.”  Thus one must study not only finished forms, but also the forces that molded them: “the form of an object is a ‘diagram of forces’, in this sense, at least, that from it we can judge of or deduce the forces that are acting or have acted upon it.”

The book paved the way for the scientific explanation of morphogenesis, the process by which patterns are formed in plants and animals.  Thompson’s description of the mathematical beauty of nature inspired thinkers as diverse as Alan Turing and Claude Levi-Strauss, and artists including Henry Moore, Salvador Dali, and Jackson Pollock.  Peter Medawar, the 1960 Nobel Laureate in Medicine, called On Growth and Form “the finest work of literature in all the annals of science that have been recorded in the English tongue.”

 source

 

Written by LW

June 21, 2015 at 1:01 am

“Eventually, all things merge into one, and a river runs through it”*…

 

Cartographer Daniel Huffman has created a series of maps in which American river systems are visualized as subway maps (specifically, in the manner of Harry Beck’s 1930s London Tube maps), with nodes representing connections between streams and tributaries.

Huffman strikes a particular chord in the map-lover’s heart. On an Internet brimming with sleek, sharp geo-visualizations, Huffman’s maps offer a sweetly idiosyncratic view of the world. The signatures of historic figures turn into street vectors. Oregon’s wine country gets the ‘90s computer-game treatment. A simple bike path becomes a work of calligraphy.

“What it really probably comes down to is a desire to do things differently than others,” he says in an email. “I crave variety, and so it often leads me to thinking of weird ideas and saying, ‘I wonder if I can do X[.]’”

Lately, a trend has emerged out of Huffman’s impulses towards novelty—an idea he calls “Modern Naturalism,” in which his maps present natural features in the type of “highly-abstracted, geometrically precise visual language that we often apply to the constructed world on maps,” according to his website

More on Huffman and his marvelous maps at “When Rivers Look Like Subway Systems.”

* Norman Maclean, A River Runs Through It and Other Stories

###

As we hop onto our rafts, we might send healing birthday greetings to Florence Nightingale, born on this date in 1820. Famed for her work as a nurse in the Crimean War, she went on to found training facilities and nursing homes– pioneering both medical training for women and what is now known as Social Entrepreneuring.  Less well-known are Nightingale’s contributions to epidemiology, statistics, and the visual communication of data in the field of public health.  Always good at math, she pioneered the use of the polar area chart (the equivalent to a modern circular histogram or rose diagram) and popularized the pie chart (which had been developed in 1801 by William Playfair).  Nightingale was elected the first female member of the Royal Statistical Society, and later became an honorary member of the American Statistical Association.

“Diagram of the causes of mortality in the army in the East” by Florence Nightingale, an example of the the polar area diagram (AKA, the Nightingale rose diagram)

 source

 source

 

Written by LW

May 12, 2015 at 1:01 am

“Correlation does not imply causation”*…

 

From stat-enthusiast (and full-time law student) Tyler Vigen, entertaining examples of patterns that map in compelling– but totally-inconsequential– ways…

More (and larger) examples at the sensational Spurious Correlations.

* a maxim widely repeated in science and statistics; also rendered: (P&Q)≠(P→Q)٧(Q→P).  It addresses the post hoc, ergo propter hoc (“affirming the consequent”) logical fallacy

###

As we think before we leap, we might send energetic (really energetic) birthday greetings to Enrico Fermi; he was born on this date in 1901.  A physicist who is best remembered for (literally) presiding over the birth of the Atomic Age, he was also remarkable as the last “double-threat” in his field:  a genius at creating both important theories and elegant experiments.  As recently observed, the division of labor between theorists and experimentalists has since been pretty complete.

The novelist and historian of science C. P. Snow wrote that “if Fermi had been born a few years earlier, one could well imagine him discovering Rutherford’s atomic nucleus, and then developing Bohr’s theory of the hydrogen atom. If this sounds like hyperbole, anything about Fermi is likely to sound like hyperbole.”

 source

 

Written by LW

September 29, 2014 at 1:01 am

“Geography is just physics slowed down, with a couple of trees stuck in it”*…

 

From Google Maps, an altogether-engrossing geographical trivia game: Smarty Pins.

* Terry Pratchett

###

As we find our bearings, we might spare a thought for Florence Nightingale; she died on this date in 1910. Famed for her work as a nurse in the Crimean War, she went on to found training facilities and nursing homes– pioneering both medical training for women and what is now known as Social Entrepreneuring.  Less well-known are Nightingale’s contributions to epidemiology, statistics, and the visual communication of data in the field of public health. Always good at math, she pioneered the use of the polar area chart (the equivalent to a modern circular histogram or rose diagram) and popularized the pie chart (which had been developed in 1801 by William Playfair).

Nightingale’s “Diagram of the causes of mortality in the army in the East”

source

 source

 

Written by LW

August 13, 2014 at 1:01 am

“The fact is that no species has ever had such wholesale control over everything on earth, living or dead, as we now have”*…

 

From the always-amazing Randall Munroe, who reminds us that bacteria still outweigh us thousands to one– and that’s not counting the pounds of them in each of our bodies…

* “The fact is that no species has ever had such wholesale control over everything on earth, living or dead, as we now have. That lays upon us, whether we like it or not, an awesome responsibility. In our hands now lies not only our own future, but that of all other living creatures with whom we share the earth.” 
― David AttenboroughLife on Earth

###

As we watch our weight, we might send carefully-calculated birthday greetings to John Graunt; he was born on this date in 1620.  A London haberdasher by trade, Graunt was fascinated the human tide that swelled around him– a fascination that led him to create the first statistically-based estimation of the population of London in his book Natural and Political Observations Made upon the Bills of Mortality, undertaken as Charles II and other officials were trying to create a system to warn of the onset and spread of bubonic plague in the city.  Profiled as one of Aubrey’s Brief Lives, Graunt has been called the first statistician, the first demographer, and was in any case the first statistician to become a fellow of the Royal Society of London.

 source

 

Written by LW

April 24, 2014 at 1:01 am

Follow

Get every new post delivered to your Inbox.

Join 1,318 other followers

%d bloggers like this: