99% Invisible - 433- Florence Nightingale: Data Viz Pioneer

Episode Date: March 3, 2021

Victorian nurse Florence Nightingale (played in this episode by her distant cousin Helena Bonham Carter) is a hero of modern medicine - but her greatest contribution to combating disease and death resulted from the vivid graphs she made to back her public health campaigns. Her charts convinced the great and the good that deaths due to filth and poor sanitation could be averted - saving countless lives. But did Nightingale open Pandora's Box, showing that graphs persuade, whether or not they depict reality? Cautionary Tales is a podcast by Tim Harford from Pushkin Industries. You can read more about the remarkable legacy of Florence Nightingale and the perils of misinformation in Tim Harford's new book The Data Detective (US/Canada) / How To Make The World Add Up (UK / International).

This is 99% invisible. I'm Roman Mars. It's pretty easy to lie with statistics, but it's really easy to lie with statistics when there is a pretty drawing attached. What follows is a story of data visualization and coercion in its centers on a person not particularly well known for either one of these things, the founder of modern nursing Florence Nightingale. And the perfect person to tell this story is Tim Harford, the undercover economist, the author of the Data Detective, and the host of the podcast, Cautionary Tales, which has just started its new season.

I could not be more excited, I love this show. I'm so happy it's back. And this episode that we have the honor of premiering is Taylor made for the 99PI crowd. So, without further ado, here's Tim Harvard. The grand barracks at Scootarian Istanbul were said to be the largest in the world when they were completed in the 1820s. By the 1850s, a grim war would turn them into the world's largest hospital. To an experienced eyes, the scutari buildings were magnificent. To ours, in their first state, there were truly whiteed sepulchres, pest houses.

Those are the words of a British nurse named Florence Nightingale. She sailed out to Istanbul during the Crimean War, a pointless conflict between Russia and an alliance including the British. Nightingale arrived late in 1854, the small team of nurses. Their task was to assist in the care of wounded British soldiers coming back from the battlefront to the hastily converted barracks hospital at Skutari. What did they discover waiting for them? Oh, you gentlemen of England can have little idea from reading the newspapers of the horror

and misery of operating upon these dying and exhausted men. That this is the kingdom of hell, no one can doubt. Within days of her arrival with fewer than 14 nurses, hundreds of casualties started arriving every day from the fighting in the Crimean Peninsula. These men were bleeding from abdominal wounds. Their faces black with gunpowder and mud, their bodies, crawling

with vermin. After each man died, he'd be stitched up in his own blanket and carried to a mass grave, making space for the next to take his bed. As for the hospital itself, Nightingale was appalled by the conditions and the Shambolic organization. The heating system didn't work, and there was no clean water. The heating system didn't work and there was no clean water. The army supply chain sent the wrong equipment to the wrong place at the wrong time as a matter of routine. And they also seemed to delight in refusing to deal with Nightingale, a woman in a man's world. No mobs, no plates, no wooden trays, no slippers, no shoe brushes, no blacking, no knives

and forks, no spoons, no scissors for the cutting of men's hair, which is literally alive, no basins, no towering, no chloride of lime. What unfolded that winter was a catastrophe. In January 1855, the British Army in Crimea lost one man in ten to the ravages of diseases such as dysentery and cholera. Many of them died at the hospitals in Skutari. Infectious disease tore the British army to shreds. Back in the UK, the reputation of the generals and politicians was also in tatters. One figure alone emerged with reputation intact. Florence Nightingale. The leader of the nurses in Istanbul was celebrated as the Lady with the Lamp, a near-religious

icon of gentleness and dedication, and the most famous woman in the British Empire, except Queen Victoria herself. There is not one of England's proudest and purest daughters who at this moment stands on so high a pinnacle as Florence's nightingale. The soldiers loved her too. If there is any angels on earth, she is one. What glory to see her delicate form gliding about amongst hundreds of great, rough soldiers and to see the looks of love and gratitude that they cast on her beloved face, it will

be a brave man at dare in sultan. I would not give a penny for his chance. In May 1855, with conditions at her hospitals improving, Florence Nightingale sailed to the front in Crimea, where she was moved both by the spectacle and by the devotion of the men. The men as a 39th regiment turned out and gave Florence Nightingale three times three as I rode away. There was nothing empty in that chair. And gave Florence Nightingale three times three as I rode away. There was nothing empty in that chair. Florence Nightingale was becoming a saint, but the battle with disease that had shaped her reputation

was about to take a sudden turn. On the 13th of May, a few days after arriving in Crimea and just a day after her 35th birthday, she collapsed. The rumours quickly spread around the British army. Florence Nightingale was dying. I'm Tim Halford and you're listening to cautionary tales. Even today, it is as a nurse that Florence Nightingale is revered in Britain. 25 years ago, my mother took her last breath in a Nightingale hospice. Nightingale's face adorned British banknotes and the front cover of magazines during the COVID-19 pandemic. All hail the hand-washing queen, and we even named our emergency COVID hospitals

the Nightingale hospitals. Nightingale is the ultimate nursing icon. It's as though she had died the day she collapsed in Crimea in May 1855 at the age of 35. Her mission as a nurse had been accomplished her march to heaven was assured, and there was nothing more to be said, which is strange. Because despite the trauma and the sickness staying with her, Nightingale lived until she was 90 years old. And she didn't bask in her celebrity, nor retire to her country home. She had a much bigger battle to fight. One woman and a hand-picked team of geeks versus the entire military and medical establishment of the country, with hundreds of thousands of lives at stake.

That huge fight is what this cautionary tale is about. That and the strangely modern weapon she used. Because Florence Nightingale was not only a nurse, she was also, and I mean this as a most sincere compliment, a total nerd. She became a statistician, the first female fellow of the Royal Statistical Society and an honourable member of the American Statistical Association. She was a master of data visualization. If you wanted to be dismissive and some people do, you'd say she was very good at drawing pretty diagrams.

But those pretty diagrams changed the world. This is the story of how to fight for a public health revolution armed with a souped-up pie chart. I don't need to tell you how ubiquitous data visualization is. Everywhere we look, whether we check social media, turn on rolling news, or flip through a newspaper, we see graphs and charts, flashy pictures of data, designed to persuade us of something. They're not just decorations. These graphs push and pull us into taking high stakes decisions.

COVID-19 reminded us of just how high the stakes can be. People have lived or died because of the decisions they've made after looking at a chart on Facebook. That's why I wanted to understand what Florence Nightingale did with graphs and how she did it. But the deeper I went into the Florence Nightingale archives, the stranger the story became. And it raises a question, if graphs are so powerful, shouldn't we worry about how that power is used? From the hospital in Skutari, Florence Nightingale had tirelessly lobbied for support and assistance, expertly dealing with the press and her political contacts to get what she needed. And in March 1855, came a turning point, a sanitary commission arrived from Britain with the task of cleaning up the hospitals in Scutari. There was a lot of cleaning to

do. Over the following weeks, they discovered that the drains leading away from the Baruch Hospital were blocked, effectively meaning that the hospital sat on a cesspool. The main water pipes, supplying part of the hospital, was blocked by a decomposing horse. Two dozen more animal carcasses were found on the hospital site. Prefabricated privies had been built in the central courtyard, but excrement was leaking out of the trench beneath them and into an adjacent water tank. By late March, the army was carrying out the commission's recommendations, clearing and flushing the sewers, cutting air vents in the ceilings, removing rotten wood floors and whitewashing everything.

The sanitary commission is really doing something and has set to work burying dead dogs and whitewashing walls, two prolific causes of fever. The death toll was far lower after the commission had done its work. And before, it was a perfect example of what could be achieved to save lives with simple cleanliness and keeping sewage away from the water supply, a nightingale did not forget the lesson. away from the water supply, a nightingale did not forget the lesson. When Florence nightingale returned from the war, Queen Victoria summoned her for a royal audience.

Ah, Miss Nightingale. Your Majesty? We have heard so much about you. Nightingale didn't think much of Victoria. She is the least self-reliant person I've ever known. But the Queen could be useful to her. Florence Nightingale had returned from the Kingdom of Hell with a mission. She wanted to make sure the awful toll of disease in Skutari never happened again in any British hospital anywhere in the world. So Nightingale persuaded Victoria to support a royal commission investigating the health of the army.

As a woman, Nightingale was unable to sit on the commission herself, but she assembled her geek allies and worked behind the scenes to figure out the problem. She turned down Queen Victoria's offer of a sweet at Kensington Palace, there would be far too many visitors. Instead, she took rooms at a low rent London hotel. So what had been the underlying cause of the death toll in the Scutari hospitals. To modernise, the answer is obvious. Disease spreads thanks to poor hygiene, poor ventilation and contaminated water, and the hospitals in Scutari suffered from all three. It was not so obvious to Nightingale.

Germ theory didn't exist in the 1850s. Although the science was mysterious, Nightingale was one of a group of Victorian thinkers, who were convinced that one way or another, good sanitation should help. With her ally, the great statistician William Far, she assembled and examined the data. Far and Nightingale became convinced that wherever they looked, premature death went hand in hand with open sewers, bad ventilation, and unclean conditions. It wasn't just about the Crimean War. It was an ongoing public health disaster in barracks, civilian hospitals, and beyond.

For pair began to campaign for better public health measures, and here the epic battle was joined. They faced powerful opposition. The government didn't want an embarrassing report about the Crimean War already regarded as a fiasco. The Queen, of course, was an instinctive conservative whose idea of reform was to replace one over promoted bureaucrat with another. And neither the army nor the medical profession cared to take instructions from a woman, not even the angel of scutari, Florence Nightingale. In any case, they believed she was surely wrong. A couple of years after the end of the Crimean War in 1858, the chief medical officer, John Simon,

acknowledged that contagious diseases such as cholera and dysentery were... A cause of premature death in every civilized country, but that they were... Practically speaking, unavoidable. These diseases just happened, said John Simon, and yes, they killed people. Deal with it, and don't take any lessons from Florence Nightingale. Nightingale was outraged at the complacency. The deaths from disease in British Army barracks were criminally high. It was just as bad, she said. As it would be to take 1,100 men out

upon Salisbury plain and suit them. The same for civilian hospitals, private homes, slums, all over the country, men, women, and children were dying, and self-satisfied men like John Simon insisted that these deaths were, practically speaking, unavoidable. The chief medical officer, the generals and the entire British establishment stood against her. Her geek sidekick, statistician William Far, warned her to be careful. Well, if you do it, you will make yourself enemies.

After what I've seen, I can fire my own guns. And the 19th century heroine had a 21st century weapon. It was a diagram. It's not an accident that these days we're surrounded by graphs and charts. They're the sweet spots between solid statistical evidence on one hand and shareable gifts and filtered photos on the other. Hard data plus striking images. Scientific evidence backs up what any news editor or social media consultant will tell you. Graphs attract attention and they persuade people.

Researchers at Tufts Visual Analytics Lab found that people formed an impression of a graphic within 500 milliseconds, just half a second. That's far too brief to understand what the graph is about, but it's not too brief to think what a mess or whoo shiny. We respond to images without conscious thought. Another team of researchers, data scientists at New York University, showed people evidence about practical policy questions. For instance, does a high corporate income tax drive jobs overseas, or does prison work as a deterrent? Sometimes the relevant data was in the form of a table, and sometimes in the form of a

chart. Unless people already had a strong position on the subject, the charts were much more persuasive than the tables. If you saw a chart, you were much more likely to change your view. That seems obvious today. It wasn't obvious in the 1850s. Statisticians were much more likely to present their data in the form of a table, even if the tables sprawled across page after page.

Beautiful design was thought to be superfluous. Florence Nightingale, not for the first time in her life, begged to differ. She would create a graph so compelling that the British establishment would have to bow in acquiescence. The graph in question is titled, Diagram of the Cause of Mortality in the Army in the East. It was published in 1859, the year after Dr. John Simon declared that death from infectious disease was practically speaking unavoidable.

Now, I'm going to try my best to describe this image. I've seen an original printing up close in the Library of the Royal Statistical Society in London. It's amazing. But you can find copies online. The first thing you would see, say if you were showing the graph for 500 milliseconds, is that it consists of two pale blue spirals,

one larger than the other. Look more closely and you see that each spiral is built of twelve equally angled wedges, like the hours of the clock. Some of the wedges are small, clinging near the centre. Others sprawl out hugely, which is what gives the diagram this sense of spiraling in or out. The Rose diagram is a beautiful image, but describes some horrifying numbers. Each of the wedges represents the deaths in a particular month, and the two circles describe the loss of life over two years from April 1854 to March 1856. 1854, to March 1856. The first circle spirals out like a snail. October is not too grave. November, when Nightingale arrived at the hospital in Scutari, is worse. December is worse still. January and February are awful. Swollen wedges of blue so large that they threaten to bleed

off the edge of the page itself. In the centre of the diagram are tiny black and red wedges of blue so large that they threatened to bleed off the edge of the page itself. In the centre of the diagram are tiny black and red wedges. They indicate a handful of deaths from miscellaneous causes and from wounds. The huge blue wedges show the overwhelming death toll from infectious diseases. No one ever made a decision because of a number. They needed a story. So said Daniel Kahneman and Amos Tversky, the two psychologists whose collaboration would win Kahneman and Nobel Prize after Tversky's death. And Florence Nightingale's diagram, more than anything, is a story. The first half of that story is a catastrophe, but the second circle continues the narrative in April 1855, just after the Sanctuary Commission arrived

in Scutari. The change is dramatic. The circle is much smaller, and while the first circle spiraled outward in an ever-worsening death count, the second circle shrinks inward as the casualty is dwindle. It's a tale with two halves. After the catastrophe comes the redemption. In between the two of them, flushing out the Suez, casting away the dead horse, disinfecting the hospital buildings. In 1858, John Simon, the chief medical officer, had declared that death from infectious

disease was practically unavoidable. In 1859 Florence Nightingale's graph said, that's a lie. Not only are deaths avoidable, but with simple practical measures, the army had avoided them. Those two pale blue circles delivered a powerful two-part payload. John Simon and his allies felt the force of both barrels. As Nightingale explained to an old friend and influential politician, whenever I am infuriated I revenge myself with a new diagram. The graphs and the Rose diagram in particular were part of a deliberate strategy. In another letter to the same friend, written on Christmas Day 1857, she sketched out a plan to use data visualization for social change. She declared her plan to have her diagrams glazed, framed, and hung on the wall at the

Army Medical Board and ward apartment. This is what they do not know, and did ought to. She added, none but scientific men even look into the appendices of a report. And this is for the vulgar public. Now, who is the vulgar public? Who is to have it? The Queen, Prince Albert, all the crown heads in Europe, with the ambassadors, all ministers of each, all the commanding officers in the army, all the regimental surgeons and medical officers, the chief sanitarians in both houses of parliament, all the newspapers, reviews, and magazines. Nightingale's visual story was impossible to ignore. Opinions started to shift, Parliament passed new laws,

doctors adopted new practices. Nightingale understood earlier than most that a chart has a special power. But I can't just end this cautionary tale there on a happy note, because perhaps charts have a little too much power. Our visual sense is potent, so potent that we even use the phrase I see as a direct substitute for I understand.

Seeing can be believing, but seeing can also mean fooling yourself. Edward Tufty, perhaps the most influential graphical guru alive, understands the power of visual explanation as well as anyone, his books include Invisioning Information and Beautiful Evidence. But we can envision misinformation too, and it can be just as beautiful, and we can now share it with a click. In mid-March 2020, as we were just beginning to grasp the enormity of the unfolding pandemic. Eric Fagelding, an epidemiologist with a large following and a quick fire style, fired off a graph from the Centers for Disease Control with a warning, Newsflash for young people.

You are not invincible. You're just as likely to be hospitalized as older generations. Even CDC says so. But the CDC's graph didn't show that at all. It showed that vastly more over 45s than under 45s were in hospital. But it was easy to misunderstand the graph if you didn't look closely at the tiny, tiny labels on the axes. And who looks at the tiny, tiny labels, eh? And so an epidemiologist who should have known better but was just a little too eager to tweet unwittingly spread

misinformation to his 300,000 followers. It gets worse. A few days later, the right-wing it and Coulter tweeted a pair of graphs and the comment, for people under 60 coronavirus is less dangerous than the seasonal flu. But the graphs showed the precise opposite of Coulter's claim. Covid-19 was about 8 times as deadly as flu for people in their 50s and 5 times as deadly for people in their 30s and 40s. It's like tweeting a picture of a cow next to a cat with a title, cows are smaller than cats. Observed, except, Coltas' up-is-down message was retweeted more than 11,000 times. Within a few hours of each other, two Twitter influencers, one overplaying the risk of

COVID and the other underplaying it, were tweeting graphs that they hadn't understood. It didn't seem to stop those tweets going viral. All too many people seem to think any claim with a graph attached must be true. People make life-changing decisions because of graphical misinformation like this, quitting a job in fear, but in fact the risk is low, or recklessly exposing others to deadly risk because they've been told the virus is fake news. It turns out that data visualization is a dual-use technology. It can be a tool or a weapon. Florence Nightingale was perhaps the first person in history to grasp that a well-designed

graph based on solid data can be remarkably persuasive. Experience has taught us the unfortunate lesson that a badly designed graph or a graph based on flimsy data, while they can be remarkably persuasive too. More of Tim Harford's cautionary tales are 99% invisible after this. There is a strange twist in this story, because while Florence Nightingale is revered by many graphic designers, many others despise her rose diagram. Edward Tufty, the influential author of Beautiful Evidence, criticised the graph on his website.

The inherent problem is the difficulty of making good comparisons across the wedges. In general for such small datasets, tables will outperform graphics. The diagram is unclear, and since there aren't that many numbers to portray, Nightingale should simply have used a table. Nightingale's statistical contemporaries would have agreed, but I've seen the data in Nightingale's graph presented as a table, and the first thing that sprang to mind was Nightingale's own comment to her friend in that Christmas Day letter of 1857. In this form printed tables and all in double columns, I do not think anyone will read it. Remember to whom she sent this earth-shaking diagram?

The Queen, Prince Albert, all the crown heads in Europe, all the newspapers, reviews and magazines. Tables can be clearer, but tables don't grab your attention in 500 milliseconds, and kings and queens and ministers and newspaper editors are busy, as Nightingale rather acedly noted when she sent her report to Queen Victoria. She may look at it because it has pictures. Fine, the message demands a graph, but surely there's a better, clearer graph conveying the same message.

Certainly, that's what Edward Tufty's followers believe in the comments on his website. Is it wrong that I'm erased by a graphic? Good design is not drawing pretty pictures and shoe-horning the facts in later. These knighting-gale roses are just a type of pie chart and contain all the disadvantages of pie charts. Wow, the disadvantages of pie charts. Wow, that's quite a burn. The charts are difficult to read. I would have thought that a stacked bar chart on a time scale would have been a better choice.

Good idea. Let's try a bar chart. And here's where the plot thickens. Because when I first saw the data presented as a bar chart, my jaw dropped. It is absolutely clear and easy to read, and that's the problem. When you see the data presented in a clear, modern format, you start to realise something. Maybe Florence Nightingale wasn't quite as saintly as everyone thought. Florence Nightingale's diagram divides the data into two halves. That's not an accident. The colour diagram number one shows the sanitary state of the army before the arrival of the sanitary commission.

The colour diagram number two shows what it became after that event. Catastrophe before, recovery after. And as I've mentioned, Nightingale was aiming not at a post-mortem of the Crimean War, but at the far bigger goal of public health reform. Similar diagrams might be constructed for towns in their unimproved and improved state. Nature is the same everywhere, and never permits her laws to be disregarded with impunity. The argument is powerful, and the conclusion is correct. Life expectancy strikingly

improved in the second half of the 19th century, and the sanitary and revolution, cleaner water, cleaner homes, cleaner air, deserves much of the credit. But that's why it's so shocking to see the data from the Rose diagram re-plotted as a bar chart. When you do that, the stark, before and after story is lost. By the time the Sanitary Commission arrived in March, flushing horses out of the water supply and carrying away tons of human excrement, deaths had already been falling sharply for a couple of months. Mark Bostridge, the author of an award-winning biography of Nightingale, argues that deaths were falling because new arrivals were in better health. In part, thanks to the better weather.

They were less numerous, so the hospital was less overcrowded, and with fewer soldiers in the hospital, of course, there'd be fewer deaths. There's no doubt that better sanitation works, but an unvarnished presentation of Nightingale's data would have suggested that the truth was complicated, and complicated, wasn't going to serve her purposes. So she created her rose diagram. The same data, artfully presented, tells a very different story. There's a famous remark in a letter that passed between Nightingale and her ally, the great statistician, William Far. You complained that your report would be dry. The dry are the better. Statistics should

be the driest of all reading. Several biographers have reported that remark as being written by Far to Nightingale. That makes sense. The Fusty, middle-aged statistician was advising the fiery, younger advocate to rein in her righteous campaigning influences and thankfully she ignored him. Except the biographers are wrong. Confused, perhaps, by the fact that the surviving draft of this letter was dictated to an assistant and unsigned, but while researching my new book, the data detective, I tracked the letter down, and it wasn't from far to nightingale, it was the other way around. You complain that your report would be dry. The dry at the better. Statistics should

be the driest of all reading. She was telling him to play it straight and avoid editorialising. Solid evidence first, she said, and worry about the sales pitch later. Good advice for any scientist. In the same letter, she added, We want facts. Fact-a-fact-a-fact-a is the motto which ought to stand at the head of all statistical work. It's puzzling. How could she produce the famous rose diagram, an artfully constructed piece of statistical storytelling, while being the same person who told William Fart to keep it dusty dry? My guess is that she was far too clever to build an argument on shaky foundations. The more spectacular the statistical acrobatics, the more solid the numbers needed to be. But it's just a guess. I don't

know. I don't even know if this cautionary tale has a happy ending. If the end justifies the means, I suppose it does, because she won. Nightingale and her allies saved countless lives, transforming the health of Victorian Britain and arguably of the world. Most of Nightingale's campaigning took place while she was confined to her bedroom by the long illness she had acquired in Crimea, and she emerged. Triumphant. Jerm theory had vindicated her focus on hygiene and public health. Her sanitary and reforms had been broadly implemented.

The everyday health of ordinary citizens had been transformed. Even Dr. John Simon, it seems, had quietly recognized his mistake. He published a collection of his essays, and without acknowledging the change, he altered the line that said that deaths from disease were, practically speaking, unavoidable. Instead saying they were, in some degree, unavoidable. From saying, there's nothing we can do to save lives. John Simon had softly sidestept into saying, well, we can't save everyone. Florence Nightingale and her rose diagram had defeated him. But let's be careful. Because it seems that if you give us 500 milliseconds alone with a pretty graph, we're all suckers. The first data visualization to change the world did so by exploiting our visual gullibility.

Florence Nightingale's beautiful graphic proved a powerful weapon. And now it's a weapon that anyone with any motive can pick up and use. Key sources for this episode include Mark Bostritch's biography and Lin McDonnell's collected works of Florence Nightingale, Hugh Small's presentation to the Royal Statistical Society, and my own book, The Data Detective, Ten Easy Rules to Make Sense of Statistics. For a full list of references, see TimHalford.com. Corsion Retails is written by me, Tim Halfafard with Andrew Wright, it's produced by Ryan

Dilly and Marilyn Rust. The sound design and original music is the work of Pascal Weiss. Julia Barton edited the scripts. Starring in this series of Corsion retails are Helen Abotten Carter and Jeffrey Wright, alongside Nazar Eldorazi, Ed Gohan, Melanie Guthridge, Rachel Hanshaw, Kobyna Holdbrook Smith, Greg Lockett, Masey M. Rowe, and Rufus Wright. This show wouldn't have been possible without the work of Mia LeBell, Jacob Weisberg, Heather Fein, John Schnarrs, Carly McLeory, Eric Sandler, Emily Rostick,

Maggie Taylor, Anneela LeCarn and Maya Canig. Cautionary Tales is a production of Pushkin Industries. If you like the show, please remember to rate, share and review. The new season of cautionary tales is out now. Go subscribe and listen. I love the show. I'm so happy it's back. And then once you subscribe, go buy all of Tim Hartford's books because they're fantastic. The latest one is called The Data Detective. rules to make sense of statistics.

99% of visible is a product of 91.7K LW in San Francisco and produced on Radio Row and beautiful downtown Oakland, California. We remember a radio topia from PRX a fiercely independent collective of the most innovative shows in all of podcasting. Find mall at radiotopia.fm. You can find the show and join discussions about the show on Facebook. You can tweet me at Roman Mars and the show at 9.9.pi org or on Instagram and read it to that our home for beautiful nerds is 99pi.org. Radio tapio. From P-R-X.