Tim Harford Credit: Árni Torfason for PopTech
Tuesday 22 September
When was the last time you read a grand statement, accompanied by a large number, and wondered whether it could really be true?
In this talk, Tim Harford drew on his experience as both an economist and presenter of the BBC’s radio show ‘More or Less’ to take us deep into the world of disinformation and obfuscation, bad research and misplaced motivation to find those priceless jewels of data and analysis that make communicating with numbers worthwhile.
Tim’s latest book ‘How to Make the World Add Up: Ten Rules for Thinking Differently About Numbers’ is available on amazon or from your local independent bookstore.
Statistics are vital in helping us tell stories – we see them in the papers, on social media, and we hear them used in everyday conversation – and yet we doubt them even more than ever.
Tim Harford is an economist, journalist and broadcaster. He is author of ‘The Next Fifty Things That Made the Modern Economy’, ‘Messy’, and the million-selling ‘The Undercover Economist’. Tim is a senior columnist at the Financial Times, and the presenter of Radio 4’s ‘More or Less’, the iTunes-topping series ‘Fifty Things That Made the Modern Economy’, and the new podcast ‘Cautionary Tales’. Tim has spoken at TED, PopTech and the Sydney Opera House. He is an associate member of Nuffield College, Oxford and an honorary fellow of the Royal Statistical Society. Tim was made an OBE for services to improving economic understanding in the New Year honours of 2019.
The following are notes from the on-line lecture. Even though I could stop the video and go back over things there are likely to be mistakes because I haven’t heard things correctly or not understood them. I hope Mr Harford and my readers will forgive any mistakes and let me know what I got wrong.
Mr Harford started his talk by relating a story about Abraham Bredius.
Bredius, by Hendrik Haverman, 1899.
Dr. Abraham Bredius (18 April 1855 in Amsterdam – 13 March 1946 in Monaco) was a Dutch art collector, art historian, and museum curator. He was also an expert on spotting forgeries, especially Rembrandts
Rembrandt Harmenszoon van Rijn (15 July 1606 – 4 October 1669) was a Dutch draughtsman, painter, and printmaker.
In 1937 Dr Bredius was approached by Gerard Boon, a Liberal State Party parliamentarian, to see if he would validate a painting entitled “The supper at Emmaus” as a painting by Vermeer.
The Liberal State Party, “the Freedom League” (Dutch: Liberale Staatspartij “de Vrijheidsbond”, LSP), was a conservative liberal political party in the Netherlands from 1921 to 1948.
Detail of the painting The Procuress (c. 1656), believed to be a self-portrait by Vermeer
Johannes Vermeer (October 1632 – December 1675), original name Jan Vermeer van Delft, was a Dutch Baroque Period painter who specialized in domestic interior scenes of middle-class life. During his lifetime, he was a moderately successful provincial genre painter, recognized in Delft and The Hague. Nonetheless, he produced relatively few paintings and evidently was not wealthy, leaving his wife and children in debt at his death.
The fake “Supper at Emmaus”
Even though Dr Bredius had never seen a painting like it he described it as a virginal, unknown, pure untouched Vermeer.
However, it was a fraud and Bredius and the whole of the Dutch art world fell for it.
Vermeer did not paint much in his lifetime so this new find should have made people ask questions.
In 1945 the threads of the fraud unravelled.
Han van Meegeren was a very successful art dealer and during the war helped the Nazi’s to acquire art.
Van Meegeren in 1945 painting Jesus Among the Doctors
Nazi Reichsmarschall Hermann Göring traded 137 looted paintings for Christ with the Adulteress, and showcased it at his residence in Carinhall (about 65 kilometers north of Berlin). On 25 August 1943, Göring hid his collection of looted artwork, including Christ with the Adulteress, in an Austrian salt mine, along with 6,750 other pieces of artwork looted by the Nazis. On 17 May 1945, Allied forces entered the salt mine and discovered the painting.
The painting was traced back to van Meegeren. On 29 May 1945, he was arrested and charged with fraud and aiding and abetting the enemy. He was remanded to the Weteringschans prison as an alleged Nazi collaborator and plunderer of Dutch cultural property, threatened by the authorities with the death penalty. He labored over his predicament, but eventually confessed to forging paintings attributed to Vermeer and Pieter de Hooch. He exclaimed, “The painting in Göring’s hands is not, as you assume, a Vermeer of Delft, but a Van Meegeren! I painted the picture!” It took some time to verify this and van Meegeren was detained for several months in the Headquarters of the Military Command at Herengracht 458 in Amsterdam.
Van Meegeren painted his last forgery between July and December 1945 in the presence of reporters and court-appointed witnesses: Jesus among the Doctors, also called Young Christ in the Temple in the style of Vermeer. After completing the painting, he was transferred to the fortress prison Blauwkapel. Van Meegeren was released from prison in January or February 1946.
The trial of Han van Meegeren began on 29 October 1947 in Room 4 of the Regional Court in Amsterdam. The collaboration charges had been dropped, since the expert panel had found that the “Vermeer” sold to Hermann Göring had been a forgery and was, therefore, not the cultural property of the Netherlands. Public prosecutor H. A. Wassenbergh brought charges of forgery and fraud and demanded a sentence of two years in prison.
The court commissioned an international group of experts to address the authenticity of van Meegeren’s paintings. The commission included curators, professors, and doctors from the Netherlands, Belgium, and England, and was led by the director of the chemical laboratory at the Royal Museums of Fine Arts of Belgium, Paul B. Coremans. The commission examined the eight Vermeer and Frans Hals paintings which van Meegeren had identified as forgeries. With the help of the commission, Dr Coremans was able to determine the chemical composition of van Meegeren’s paints.
He found that van Meegeren had prepared the paints by using the phenolformaldehyde resins Bakelite and Albertol as paint hardeners. A bottle with exactly that ingredient had been found in van Meegeren’s studio. This chemical component was introduced and manufactured in the 20th century, proving that the “Vermeers” and “Frans Halses” examined by the commission were in fact made by van Meegeren.
On 12 November 1947, the Fourth Chamber of the Amsterdam Regional Court found Han van Meegeren guilty of forgery and fraud, and sentenced him to a minimal one year in prison. During his trial he admitted to painting supper at Emmaus.
Further investigation such as the use of gas chromatography and analysis of the lead paint used showed that it was, indeed, a forgery.
Radiometric dating, radioactive dating or radioisotope dating is a technique which is used to date materials such as rocks or carbon, in which trace radioactive impurities were selectively incorporated when they were formed. The method compares the abundance of a naturally occurring radioactive isotope within the material to the abundance of its decay products, which form at a known constant rate of decay.
So why do people fall for frauds like this?
Because, sometimes we want to be fooled.
What went wrong for Bredius? He said “I had difficulty controlling my emotions”
Lecturer at the Department of Economics, University of Sydney
In 2011 he carried out an experiment which tested whether and in what circumstances people are more likely to believe an event simply because it makes them better off. Subjects observed a financial asset’s historical price chart, and received both an accuracy bonus for predicting the price at some future point, and an unconditional award that was either increasing or decreasing in this price. Despite incentives for hedging, subjects gaining from high prices made significantly higher predictions than those gaining from low prices. The magnitude of the bias was smaller in charts with less subjective uncertainty, but was independent of the amount paid for accurate predictions.
Forecasts shouldn’t be affected by hope, but it is.
Linda Babcock and George Loewenstein research used a real-life case
To investigate the role of self-serving assessments in bargaining, they designed an experimental paradigm, which was then used in a number of experimental studies. They developed a tort case that was based on a trial that occurred in Texas, in which an injured motorcyclist sued the driver of the automobile that collided with him, requesting $100,000. Subjects were randomly assigned to the role of plaintiff or defendant, and attempted to negotiate a settlement. Subjects first received a page explaining the experiment, the sequence of events, rules for negotiating, and the costs they would face if they failed to reach an agreement. Both subjects then received the same 27 pages of materials from the original legal case in Texas. The materials included witness testimony, police reports, maps, and the testimony of the parties. Subjects were informed that identical case materials were given to a judge in Texas, who reached a judgment between $0 and $100,000 concerning compensation to the plaintiff.
Before negotiating, subjects were asked to write down their guess of what the judge awarded. They were told they would receive a bonus of $1 at the end of the session if their prediction was within $5,000 (plus or minus) of the actual judge’s award. They were also asked what they considered a fair amount for the plaintiff to receive in an out-of-court settlement “from the vantage point of a neutral third party.” Subjects were told that none of this information was shown to the other party. The two subjects were then allowed to negotiate for 30 minutes. Delays in settlement were made costly to the subjects by imposing “court costs” that accumulated in each period in which the subjects failed to settle. If they failed to reach a voluntary settlement within 30 minutes, then the judge’s decision determined the defendant’s payment to the plaintiff and legal costs were levied on the parties.
The experiment was designed to test for the effect of the self-serving bias in a contextually rich and controlled experimental setting. Since both parties were given the same case information and neither party had private information about the judge, differences in estimates between defendant and plaintiff couldn’t be attributed to differences in information.
The first experiment with this framework found strong evidence that the negotiators formed self-serving assessments of the judge’s award, and that the discrepancy between the plaintiffs’ and defendants’ assessments was correlated with the parties’ ability to reach voluntary settlements (Loewenstein, Issacharoff, Camerer, and Babcock, 1993). The subjects were 80 undergraduates from the University of Chicago and 80 law students at the University of Texas at Austin. Subjects were assigned randomly to roles as either the defendant or plaintiff immediately upon entering the experiment.
The self-serving bias was clear in that plaintiffs’ predictions of the judge averaged $14,527 higher than defendants’, and plaintiffs’ fair settlement values averaged $17,709 higher than defendants’, with both differences statistically different from zero.
Brexit was an example where emotions overtook facts. Anti-Brexit people regarded themselves as Europeans where Brexit supporters thought about “The Good Old Days”.
The media didn’t help either.
Problems arise when we don’t consider facts.
In his book, Mr Harford lays down ten rules.
Number one: Reflect on whether a claim makes sense. Is it fake news? How does it make you feel?
Pay attention to your emotions. Reflect more calmly. Notice and trust your feelings before accepting what you are being told.
Gain more experience.
One overwhelming example of not reflecting us the Republican and Democrats opinion of climate change.
Republicans don’t think climate change is a problem.
Democrats do think climate change is a problem. However, there is a bigger gap in opinions.
However, more facts don’t necessarily bring us together.
Seeking facts that you want and ignoring the ones you don’t.
People tend to use the media that will support their views.
There was a paper written once that said that drinking lots of coffee could cause cancer in women. It was accepted by everyone except women who drank lots of coffee.
It has been found that if the stock market goes up then investors are more likely to check the performance of their shares than if the stock market goes down.
The ostrich effect. People don’t want painful news.
Expertise helps but we need to be careful. The more we have the easier it is to draw the conclusions we want.
Abraham Bredius should have known that the supper at Emmaus was a forgery.
Girl Reading a Letter by an Open Window is a genuine Vermeer and is obviously much better than the supper at Emmaus, which doesn’t actually look like a Vermeer.
How was Bredius fooled?
van Meegeren used an original 17th century painting, which he painted over. He made sure that the colours on the bread and jug looked authentic for the time. Bredius picked up on these subtle things, including the fact that the oil paint was hard (oil paints can take up to 15 years to truly harden). But it was hard only because the paint had been mixed with bakelite.
Bredius knew that Caravaggio has painted a version of supper at Emmaus and thought that this had inspired Vermeer. There was big gap in Vermeer’s timeline with no paintings. van Meegeren knew this and Bredius accepted that Vermeer could have done paintings like supper at Emmaus during this time. This was proof to Bredius he was right.
Michelangelo Merisi (29 September 1571 – 18 July 1610) was an Italian painter active in Rome for most of his artistic life.
We are all guilty of reading statistics that we don’t really understand. We believe them if it suits us and we disbelieve if it doesn’t.
We need to recognise our emotions before we take the next step.
Emotions can make us disbelieve things that are true, climate change for instance.
Rejection of facts goes back a long way.
In 1954 a certain book was very popular
How to lie with statistics
How to Lie with Statistics is a book written by Darrell Huff in 1954 presenting an introduction to statistics for the general reader. Not a statistician, Huff was a journalist who wrote many “how to” articles as a freelancer.
There has been lots of books debunking statistics – it’s fun
Perhaps we should be doing the opposite
In 1950 a very important paper was published
Sir William Richard Shaboe Doll CH OBE FRS (28 October 1912 – 24 July 2005) was a British physician who became an epidemiologist in the mid-20th century and made important contributions to that discipline. He was a pioneer in research linking smoking to health problems.
In 1950, Richard Doll published research in the British Medical Journal showing a close link between smoking and lung cancer. Beginning in December 1952, the magazine Reader’s Digest published “Cancer by the Carton”, a series of articles that linked smoking with lung cancer.
We need to check facts as this can save lives,
There are two different views of statistics.
They are a useful tool, but only if there is enough information.
At the start of the Covid-19 pandemic there was very little information for statistics so it was difficult to make decisions.
Coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). It was first identified in December 2019 in Wuhan, Hubei, China, and has resulted in an ongoing pandemic.
Statistics are important in decision making. They have been used in dealing with unemployment and inflation
Dealing with statistics takes care, courage and money but can empower people.
Try to compare the information with something you understand.
What is really going on?
What is the context?
What do the numbers really mean?
Is there something missing from the data?
Notice your own emotional reaction.
We can fall for anything if we want to.
People repeat false stories because they want them to be true.
The Dutch wanted to believe that van Meegeren was a plucky little forger, getting one over the nasty Nazi’s.
However, there was a different take on his story.
During the Nazi occupation of Holland, he was a Nazi sympathiser. A noted anti-Semitic, he threw parties for noted Nazi’s and had plenty of food and alcohol whilst the rest of Holland starved.
A book he had given Hitler was found with an inscription to Hitler from him however people ignored these facts because it spoilt the narrative. People didn’t want another story about a Nazi collaborator.
Questions and answers
1) How do you stop misinformation spreading?
Check the facts.
Hans Rosling (27 July 1948 – 7 February 2017) was a Swedish physician, academic, and public speaker. His posthumously published book Factfulness, coauthored with Anna Rosling Rönnlund and Ola Rosling, became an international bestseller.
Factfulness: Ten Reasons We’re Wrong About the World – and Why Things Are Better Than You Think is a 2018 book by Swedish statistician Hans Rosling with his son Ola Rosling and daughter-in-law Anna Rosling Rönnlund. In the book, Rosling suggests the vast majority of human beings are wrong about the state of the world. He demonstrates that his test subjects believe the world is poorer, less healthy, and more dangerous than it actually is, attributing this not to random chance but to misinformation. Rosling recommends thinking about the world as divided into four levels based on income brackets (rather than the prototypical developed/developing framework) and suggests ten instincts that prevent us from seeing real progress in the world.
Many people think the world is getting worse when, in fact, it is not. The survey at the beginning states that with over 10,000 poll recipients 80% knew less about the world than chimps would have [had] they just guessed. That, the authors claim, shows that the media systematically skew data and trends and select stories to make people think that the world is getting worse.
It should be noted, however, that the book does have its critics.
However even if the facts are known it is very easy to believe the opposite.
A famous example is the Iraqi weapons of mass destruction. Even though there was absolutely no evidence that they existed it just made people believe they existed more.
In my own personal opinion George Bush junior wanted to punish Saddam Hussein for the Gulf war and finish the job his father, George Bush senior, started.
We know vaccines are good and do not cause autism. People who don’t accept this push back even harder. This is the back-fire effect.
There is a bad way to fact check (although the media has improved)
Nigel Farage made a statement “Malmo in Sweden was the rape capital of Europe because so many immigrants live there”
The BBC News web page had a headline
“Is Malmo the rape capital of Europe?”
You had to scroll down a lot of rubbish to find the answer, which was, of course, No.
The untrue claim was repeated over and over again and unfortunately people just remember the false claim.
Things are better now and the fact is made more prominent.
You have to get people interested. There’s nothing that an audience can’t understand if they are interested.
Fact-checking can involve some humour but you need to play it straight. It can be boring.
2) How do I convince someone to change their mind?
Explain. Get them interested. Don’t try to persuade.
People think they know more than they do.
Illusion of explanatory depth
For instance, you could ask someone if they understand how a ball point pen works. They will reply they are, say, 80%, sure because they know how to find those answers. Knowledge Adjacency seems to be is a misleading form as a form of intelligence and understanding. We confuse what our community collectively knows with our own individual understanding.
However, if you ask them, there and then, to explain how a ball point pen works they probably won’t be able to. People stumble very quickly.
It can also happen between people. I once had someone think I would know why her VCR didn’t work because I had a physics degree.
The same thing happens with politics. People have strong views which moderate only if they have to try and explain it.
I have to say though that I didn’t see this when it came to Brexit. When reporters asked some militant Brexiteers to give a reason for Brexit most of them couldn’t. It didn’t stop them wanting Brexit.
Most people know less than they think. Although my experience as a physics teacher is that girls tend to know more than they think.
Can you explain this thing you are arguing about?
Nobel winning physicist Richard Feynman suggested “If you want to master something, teach it.”
I totally agree with this statement. I had taught electromagnetic induction to A level students for four years before I thoroughly understood it.
Richard Phillips Feynman ForMemRS (May 11, 1918 – February 15, 1988) was an American theoretical physicist, known for his work in the path integral formulation of quantum mechanics, the theory of quantum electrodynamics, the physics of the superfluidity of supercooled liquid helium, as well as his work in particle physics for which he proposed the parton model. For contributions to the development of quantum electrodynamics, Feynman received the Nobel Prize in Physics in 1965 jointly with Julian Schwinger and Shin’ichirō Tomonaga.
Sir Patrick John Thompson Vallance FRS FMedSci FRCP (born 17 March 1960) is a British physician, scientist, and clinical pharmacologist who has worked in both academia and industry and, since March 2018, has been the Chief Scientific Adviser to the Government of the United Kingdom.
Sir Patrick was involved in producing what has been jokingly called the “devil’s graph”. It shows the worse case scenario if nothing is done to prevent the spread of the virus. Cases would double every seven days.
Yes, it looks bad but should we just be considering hospitalisation. You can have the virus but with very little symptoms. It should be noted that even at the height of the crisis not all of available hospital beds were needed.
Look at all the information. What do the numbers actually say?
4) Debunking is good for getting rid of bull-shitting. It is fun to call out nonsense but there must be room for the truth.
5) As the American’s say “Do the math”.
Matt Hancock, Secretary of State for Health and Social Care, at time of writing, came up with a slogan “lose 5 pounds and save the NHS more than £100m”.
Matthew John David Hancock (born 2 October 1978) is a British politician serving as Secretary of State for Health and Social Care since 2018 (at time of writing).
The sums are not hard and Mr Harford came out with a value of 30p saved per person. It will probably cost more to put in place things to help people lose weight.