General Douglas MacArthur was supreme commander for the Allied Powers in Japan, in the years following World War II. MacArthur used his mandate to destroy Japan’s militaristic tradition and feudal social and economic system and set Japan on the road to becoming a democratic economic superpower. Japan had been devastated in World War II, being the only country to suffer nuclear attack. [1] MacArthur brought in a then obscure statistician named W. Edwards Deming, who would play a key role in the Japanese economic miracle.
Statistics, the Language of Proof in Science
Statistics allows for objectively (that hate speech word) determining whether the difference between a control and experimental group should be attributed to the treatment (for example the drug administered) or failing to rule out it occurred by chance. It was not until 1908 that the tools for that determination were invented by Wiliam Gosset, he published under pseudonym of “student”. [2]
I remember, in the mid-1960’s, when an elderly friend of my mother showed me
her college graduation ring from 1908. That is how new statistics is. It only took 4,000 years after the creation of sophisticated mathematics by the ancient Babylonians for statistics to begin. Not having a concept of randomness, because of religious and cognitive grounds, prevented it.
Statistics is the study of randomness. Even the ancient Greeks, who were on the verge of calculus before ancient Greek civilization was destroyed, did not understand that concept. The ancient Romans added not one iota to Greek mathematics. After that civilization fell, it became a matter of what was to burned and forgotten during the dark ages.
It was only in the 1654 that Blaise Pascal and Pierre de Fermat a could look at probability for the first time, and without fear of “cancel culture”. A bit earlier could have resulted in a trial for heresy.
By the end of the 20th century evolution was discovered and made the concept of randomness permissible and indisputable. The second industrial revolution after 1870 made statistics essential. Industrialization required standardization of manufacture and the ability to make valid conclusions about populations from small samples. Gosset got the ball rolling when he worked for the Guinness brewery company and was faced with the problem of how to test the quality of the ingredients used (i.e the “population”) when he could only test a small sample.
Statistics is Racist
The “woke” isn’t satisfied with simply destroying every monument to Galton, Fisher, Pearson, the fathers of statistics. They want statistics itself, the only thing that separates science from primitive tribal superstition, banned!
In 1662 statistics did not exist, science didn’t exist. Europe was taking its first baby steps after 1,800 of the dark ages. John Graunt, in a pioneering work, analyzed London birth and death records and published the first life table He found life expectancy was just under 20 years! Eventually science, which relies on statistics, would change that.
Cancel Culture is now cancelling statistics. [3] Statistics ranks with calculus as the greatest of mathematical inventions. It and only it allows us to objectively determine whether data obtained support a hypothesis or not. But that is its sin.
As put by a researcher and scientist,“Classical statistical methods have a politically conservative bias…The tools that we use on a day to day basis to interrogate data and understand the world, were developed by white supremacists…Why don’t scientists worry about statistics’ sordid roots? The main reason is that scientists have fully bought into the myth of objectivity.” [4]
The same sentiments were in an article in Nautilus,“The method of significance testing and the reputations of the people who invented it are crumbling simultaneously. Crumbling alongside them is the image of statistics as a perfectly objective discipline…To get rid of the stain of eugenics, statistics needs to free itself from the ideal of being perfectly objective.” [5]
That article also quotes a data scientist in healthcare, who said, “Objectivity is extremely overrated,” and “What the future of science needs is a democratization of the analysis process and generation of analysis…Just because you haven’t measured something doesn’t mean that it’s not there. Often, you can see it with your eyes, and that’s good enough.”
“Science” will not be allowed to contradict what the woke “see”. Any wonder why there is public distrust of scientists.
Electric Lunacy
Japan will ban the sale of gasoline-powered cars by 2035. This will devastate Japan’s car industry. Toyota President Akio Toyoda has said “the current business model of the car industry is going to collapse.” Mr. Toyoda maintains this would cost hundreds of billions of dollars, make cars unaffordable for average people, and lead to blackouts as the electric grid would be overwhelmed. He also pointed out that most of Japan’s electricity is generated by burning fossil fuels. But hey, destroying Japan’s car industry is a small price to pay to pay in the name of the global warmist hoax.
When I was a child my father never owned a car. But I remember a conversation he had with a friend who once gave us a lift in what was probably the first Japanese car I saw. This would have been the 1960’s. Japanese cars were new in the United States, and “made in Japan” was looked at as meaning low quality. [6] This man said he took a chance buying this car because it was inexpensive, even though it was small, plain, and unproven. The car was a source of amusement to his neighbors. But he had the last laugh. While his neighbors’ big expensive American cars routinely broke down, he was delightfully surprised to find that nothing ever went wrong with his small spartan car.
How did Japanese cars become the quality standard of the world, while simultaneously US cars went from envy of the world to undesirable? As we will see below, part of the answer involves the work of a “made in America” statistician.
The Quality Control Paradox
How do you insure quality in manufactured goods? The first thing that will occur to one is the need to reject faulty pieces. But it is too expensive to produce faulty pieces, only to throw them out. You need to make sure faulty pieces are not produced, by eliminating variability within what is one model, as if a piece is faulty it variates from the intended specifications. So far so good. But manufactures ran into a seemingly bizarre problem at this point. Western Electric, for example, manufacturer of phone parts, found that the more they tried to eliminate variability, the lower product quality became!
Walter Shewhart (1891-1967) was hired by Western Electric to improve quality. He came to understand the problem and solution. Shewhart had a science background, and acquired vast statistical knowledge. He was heavily influenced by pragmatism and emphasized the centrality of the scientific method as our tool for acquiring knowledge about objective reality (that now dirty word, objective, keeps coming up).
Shewhart wrote about quality control “specification, production, and inspection correspond respectively to making a hypothesis, carrying out an experiment, and testing the hypothesis. The three steps constitute a dynamic scientific process of acquiring knowledge.”
Shewhart divided variability into chance variation, also known as common cause, which is random, and assignable cause variation, which is due to a specific cause, and is systematic. Assignable cause variability can and should be eliminated ,but random variation will always be there. Trying to eliminate it is counterproductive. Rather the goal is to keep it under control. But “When management does not understand variation their decisions can, and usually do, make things worse.” A simple graph of his own construction, the Shewhart Control Chart, revealed if chance variability was under sufficient control, or as Shewhart put it, “under economic control”.
His knowledge of Statistics, combined with an appreciation of pragmatic considerations, lead Shewhart to invent statistical quality control, which he would become known as the father of. Statistical quality control is a science, and as Shewhart put it “In this sense, specification, production, and inspection correspond respectively to making a hypothesis, carrying out an experiment, and testing the hypothesis. The three steps constitute a dynamic scientific process of acquiring knowledge.” [7]
W. Edwards Deming (1900-1993) would earn a degree in engineering and a PhD in physics. He needed to work his way through school. This turned out to be a big break, for he got a job at the Western Electric Company where he met Walter Shewhart. Shewhart because Demings’s friend and mentor. Deming would take Shewhart’s ideas, add to them, and carry Shewhart’s banner for the rest of his long life.
W. Edwards Deming
Although his academic training was in mathematics and physics, Deming became an expert on statistics, helped by getting a year leave from the agriculture department, to study under Sir Ronald Fisher at the University College of London. The same Ronald Fisher who was the greatest contributor to statistics ever, and who now has been cancelled by UCL.
In 1947 Deming were just another statistician, best known for working on the US census, when General MacArthur’s office asked him to assist on Japan’s 1951 census. An accurate estimation of the damage done to Japan and a revitalization of Japanese industry was in America’s interest. No one then could predict that Japan would soon surpass the USA in major industries. And Deming’s arrival in Japan in 1950 was a big part of this process. That year, Deming was invited to come to Japan and lecture on theories on applying statistics to quality control and improvement. Japan’s industry leaders eagerly attended his lectures.
One talk Deming gave in 1950 was attended by top industry leaders representing 75% of Japan’s industrial production, As Deming put it in that lecture”Statistical product quality administration is a splendid new tool.” Japan’s industrial leaders gobble up what Deming offered.They had no choice, Japanese industry in shambles. In the USA however his methods were ignored. Why bother? The USA was then the world only economic power.
Companies such as Toyota Motor Corp. and Sony Corp utilized Deming’s ideas, and became dominant players. “Made in Japan” went from meaning junk to meaning highest quality. For a time Deming was the best known and most admired American. Deming developed an affinity for Japanese culture, and Japan even created a Deming prize for companies making great advances in quality.
In the United States, he only became well-known in his old age .While Japan was implementing Shewhart-Deming methods for quality control, the attitude in the US was that quality did not matter to consumers. Just make it bigger, shinier, with bigger fins and more chrome. I remember those days. It was an attitude that lead to the end of the US electronics industry and almost the end of the US auto industry.
W. Edwards Deming died of cancer at age 93 on December 21 1993. He needed all those years. The United States and Western Europe only discovered the Shewhart-Deming’s methods in the 1980’s, when Shewhart was dead for almost two decades and Deming was in his 80’s. Deming finished his last seminar only 10 years before his death. Deming was generously giving Shewhart credit right to the end.
Deming in the USA
Deming’s fame in the USA began in 1980 when NBC News produced a documentary called “If Japan can… Why can’t we?”. The documentary involved sending crews to Japan and its purpose was to find out why Japanese products were so much better than their American competition. At this point point Deming could charge astronomical fees for his advice in the USA, after decades of neglect in his home country.
The New York Times once described Deming in his old age as “a tall, formal man who habitually wore frayed three-piece suits and spoke to senior executives as if they were school boys.” At age of 86, in 1987, speaking before America’s industry leaders, Deming said “Can you blame your competitor for your woes? No. Can you blame the Japanese? No. You did it yourself.”
Similar to Japan being forced to listen to Deming because of the devastation of WWII, The USA was forced to listen to him because of the devastation faulty products wrecked on American industry. However a recent article in Industry Week writes. “After making considerable progress in the 1980s and 1990s, corporate America has turned its back on Dr. Deming’s teachings.” [8]
Conclusion
Of course Deming preached the importance of statistics in quality control, but he emphasized that getting the actual factory workers involved was essential to controlling manufacturing problems. Deming’s methods included forcing companies to view workers as partners and understanding the importance of meeting customer needs. Deming believed the customer is the only boss. But that's only true in capitalistic systems. A system that was firmly rejected in the 2020 election, whether by the will of the voters or by by the will of the Marxist Democrat Party and their Chinese role models.
Capitalism is the free market creating progress through a process similar to how evolution produces organized complexity. Capitalism, like evolution, produces value from the bottom up. Change occurs due to changing technology and customer demands. In capitalism, change can not be dictated from above. Electric cars are far inferior to those with internal combustion engines. In addition to waiting times for recharging and limited range, they rely on lithium batteries which use rare and costly metals and will pollute and pose health hazards when dumped after use. They are created not to meet consumer demand, which is prohibited to talk about in Marxism, but because of government rules. And not many will buy them unless internal combustion engines are banned.
Gasoline powered cars, plastic bags, red meat, all verboten.
Statistical methods made Japanese cars better. Japanese cars became a huge success because one by one American buyers voted to buy them. It was a free election, remember those.
Notes and References
1. The current administration and Democrat party are eager to let Iran male Israel number two. But that is another story.
2. Student, The Probable Error of the Mean, Biometrika 6 (1): 1–25. March 1908. (This is the most important article in the history of statistics.)
3. By statistics I mean current hypothesis testing statistics used to objectively test predictions made by hypotheses based on the scientific method. Leftists will have something they call “statistics”.
4. Cleather, Daniel. (March 11, 2020) Is Statistics Racist? https://medium.com/swlh/is-statistics-racist-59cd4ddb5fa9
5. Clayton, Aubrey. ( Oct. 28, 2020) How Eugenics Shaped Statistics. Nautilus.
6. When my father was repairing watches in the early and mid-1960s, the movement were largely Swiss, even the cheapest pin levers which my father regarded as wind-up toys. There were no Chinese movement in the US market. But there were Russian movements which my father regarded as junk (Superfluous useless wheels, fake jewels glued down to the plate to create 50 Jewel monstrosities).Around 1968, my father saw his first watch with a Seiko Japanese movement. He was expecting it to be likewise junk, and he was shocked how good it was. I remember because he was so shocked that he was motivated to tell me.
7. Shewhart, W. A. 1939. Statistical Method from the Viewpoint of Quality Control. Department of Agriculture. Dover, 1986, page 45
8. Holland, Charles. (July 3, 2019). Forgotten Lessons of W. Edwards Deming. IndustryWeek.
