Slashdot Mirror
Florence Nightingale, Statistical Graphics Pioneer
Science News has a fascinating look at an under-appreciated corner of the career of Florence Nightingale — as an innovator in the use of statistical graphics to argue for social change. Nightingale returned from the Crimean War a heroine in the eyes of the British citizenry, for the soldiers' lives she had saved. But she came to appreciate that the way to save far more lives was to reform attitudes in the military about sanitation. Under the tutelage of William Farr, who had just invented the field of medical statistics, she compiled overwhelming evidence (in the form of an 830-page report) of the need for change. "As impressive as her statistics were, Nightingale worried that Queen Victoria's eyes would glaze over as she scanned the tables. So Nightingale devised clever ways of presenting the information in charts. Statistics had been presented using graphics only a few times previously, and perhaps never to persuade people of the need for social change."
For Christ's sake, spell her name right.
My web domain.
So Nightingale devised clever ways of presenting the information in charts.
So, in other words, she invented PowerPoint.
The higher the technology, the sharper that two-edged sword.
I searched around for a more readable graph and found one here, at the bottom of the page.
So if it wasn't for Ms. Nightingale, I would never have understood the deleterious effect my cat was having on my homework performance, as it might never have been properly explained.
If she'd been a man presenting this, she'd have made the equivalent of surgeon general in her career. -_- No joke--Despite the blessing of Queen Victoria herself, she was denied a chairman position that oversaw general health affairs in the military. I doubt there's an academic statistics book currently in circulation that gives her any credit for this. Even this--a zine read by only a tiny, tiny fraction of the people who go to school every year and rely on her innovation. Hell, the entire field of field medicine was in disrepute at that time in history -- who needs medicine? Most nurses spent at least part of their time in the kitchen, which was viewed as more important. She made it important. It's been two centuries since then and she's still only a footnote. Today, graphical statistics are used in every trained discipline from engineering to medicine to management, but nobody knows this woman's name. They should -- they owe her a lot.
#fuckbeta #iamslashdot #dicemustdie
...by David Salsburg mentioned Florence Nightingale (jeez, who proofreads Slashdot contributions?):
Although I think her most lasting legacy was to lend her name to the daughter of her friends, Florence Nightingale David, who went on to make valuable contributions to combinatorics and statistics.
Comment removed based on user account deletion
Nightingale didn't get the credit she deserved, but I don't think that credit is really what matters here. Yes, she did it, but shouldn't we look at the results rather than the person who caused them? I do not know much about Nightingale, but most people who choose to work in the medical field back then did it out of selflessness. You didn't become a doctor because you wanted glory, you became a doctor because you wanted to help people.
Except calling Obama "black" is inherently racist.
His mother is white.
His father is black.
Shouldn't that make him "half-black" or "half-white"? How about being "Kenyan-European"? Or does the European parts of his mother's heritage need to be broken out, making him "Kenyan-English-Irish-German"?
How about him just being "American", and only mention his heritage when it is relevant, like when asked where his parents were born.
If I have nothing to hide, don't search me
I'm just going to comment on the graph itself, without connection to the person:
Presenting this kind of data - abolute numbers and their breakdown into individual contributors,
for consecutive, identical intervals of time - in a polar graph as some kind of piechart is a very bad idea.
Piecharts are good to represent relative parts of a whole, by segmenting a circle. That's it. As soon as a radial
component is included (as is the case here - it even is the main component), they become at least misleading, and I'd call it useless.
What does the radial dimension mean? Is it linear (I presume it is - but how can I be sure?)? Are we supposed to interpret
the sizes of the areas? Note that the exact same radial length leads to a much larger area on the outside of the graph than
at its center. And we tend to see areas, not lengths. Imagine the order of subnumbers switched: The area of the orange parts
("death from battlewounds") would grow considerably, while the blue area ("death from disease") would shrink quite a bit, thereby
of course reducing the point this graph tries to make. This point is actually a perfectly valid one - but trying to "sex it up"
with a misleading graph is a bad idea. Also, the segments themselves are unnecessary and don't contribute to the information:
We know months are roughly the same length each. No information gained here.
This graph is actually a very good example how not to do it. The unambiguous, easy to interpret graph to use here is a simple histogram.
It appears they are finaly getting around to last years issues.
http://www.economist.com/world/europe/displaystory.cfm?story_id=10278643
>
From even the summery:
"""
Statistics had been presented using graphics only a few times previously
"""
So, she didn't invent them then, now did she. One of the first, fine. One of the ones to popularise its use, fine. But, invent, hardly.
Proves that good intentions may cause unexpected harm. Here development ultimately lead to Death by Powerpoint casualties.
Full circle. She couldn't win.
Bert
History is full of women who's contributions have been forgotten. Another one is Lucy Parsons. Her and her husband were anarchist labour leaders in Chicago where they helped organize the events known as the Haymarket Riots which gave the rest of the world May Day.
The Chicago police called her "more dangerous than a thousand rioters" and she was a major influence on labor politics until she died in a house fire in 1942 that also consumed most of her many writings.
In 1905 she wrote this piece for The Liberator, published October 22:
FAMOUS WOMEN OF HISTORY: Florence Nightingale
Amid the general consternation, the minister of war wrote a letter to Miss Nightingale, stating that he considered her the only person in Great Britain capable of bringing order out of confusion, and imploring her to organize and direct the reform of the military hospitals; and this letter was crossed by one from Miss Nightingale, volunteering to place her strength and ability at the service of her nation. Good trained nurses were almost unknown quantities in those days; yet, nothing daunted, Florence Nightingale sailed from England with thirty of the best nurses that she could muster within the week from her letter. In required a good deal of tact to overcome the prejudices and jealousies among the physicians and surgeons at the "womanly prominence" and the conciliate the general disapproval of medical and military officials. For these were the days when it was considered that "the proper place for the woman is at home."
Overcoming professional jealousy, she set herself to the task of cleansing the Augean hospitals containing over 4,000 patients. These barrack hospitals at Scutari, which had been loaned to the British government by the Sultan of Turkey, were 100 feet above the Bosporus. The day before the arrival of the staff of nurses the wounded from Balaclava had been landed; packed in the overcrowded transports, their wounds had not been dressed for five days, and cholera and fever were reaping their fearful harvest. The poor men outside with cold and starvation were faring far better than the sufferers in the tainted wards of the disordered hospitals.
-------------
I got this out of "Lucy Parsons: Freedom, Equality and Solidarity".
Off the top of my head, some other woman who have been mostly forgotten include Elizabeth Gurley Flynn (a co-founder of the ACLU), Ada Lovelace (perhaps earliest programmer), Hedy Lamarr (co-invented spread spectrum wireless communications years before it was technologically practical to implement, but better known for being a babe). How many people here know the name Rosalind Franklin? All of these women and many more excelled in male dominated fields.
Not to take away from Nightingale's achievements, but the most groundbreaking and impactful innovation in graphical representation of disease vectors came from Dr. John Snow, who created a map of SoHo's (London) devastating 1854 cholera outbreak that convincingly made the case that cholera was water born and not the result of miasma. The medical establishment at the time largely dismissed Snow's findings, but the power of the graphical representation convinced the people it needed to in the end and Snow's theories were ultimately vindictated. Unfortunately, Snow didn't live long enough to see his ultimate triumph. Some speculate that his habit of experiementing on himself with ether and chlorophorm may have contributed to his early demise. (Snow was also a pioneering anathesiologist, and even assisted in the birth of Queen Victoria's eight and rather difficult childbirth.) All this is recounted in Steven Johnson's excellent book The Ghost Map (2006). He talks about Nightingale as well, though not about her charts and graphs. Nightingale was, at least through the 1850s a proponent of the eventually discreted miasma theory.
"No live organism can continue for long to exist sanely under conditions of absolute reality;..."
"...an innovator in the use of statistical graphics..."
Really? I'd have said that she was an innovator in the use of statistical graphics to MISLEAD and 'spin' her data to enhance what she wanted to show (so in that sense, I guess she was in fact ahead of her time, the foremother of all crappy powerpoint presentations).
Why do I malign such a wonderful woman? Because her presentation is misleading and not so terribly well-presented in terms of either accuracy or simplicity.
1) while the method of graphing the data is perhaps novel in it's way of advancing over time, it's NOT USEFUL. It's finite - once you've determined the proportion each pie piece is of the circumference, that's it. If your pie pieces are going to each be 30 degrees, you get 12 data points, and that's IT...have a 13th point? Sorry, need to start another roundel (or whatever it's called) subsecting the data in ways that are at least hard to interpret and possibly misleading.
2) circular (area) presentations of linear data should always make the viewer suspicious, and this is no exception. Circular data emphasizes change in disproportional ways, as recognized and explained perfectly by Tufte. For example if you're showing your information as 'circles of relative size' but your data is implemented as the diameter of those circles, a simple doubling of the diameter actually increases the AREA of the circles (what your eye instinctively recognizes) by FOUR. So if you want to mislead people that a small increase really 'feels' quite a bit larger, circular graphs are the ticket. This is precisely what FN did here. Her goal was to show the HUGE number of 'preventable' deaths, and she did this in two ways: first, she chose the circular-presentation which exaggerates increases by ballooning the area disproportionally to the actual numeric increase. Secondly, she even further stacked the graphs, pushing preventables out to the circumference of the circle, further exaggerating the numbers because they were then stacked ATOP the death data, sneakily increasing the radius (and thus the displayed 'area') even if preventables did NOT increase.
She obviously had the best of intentions, but let's recognize this 'graph' for what it is: a very clever presentation of highly massaged data to induce an administrator to come to the conclusion desired. It's propaganda, nothing more. Well intentioned, but still propaganda.
So clearly, she's not simply the mother of the Red Cross, but the ancestor of all modern hatable powerpoint quackery to the present day.
-Styopa
There's an interesting discussion thread on the Ask ET section of the Edward Tufte web site that starts with the Nightingale charts. Note that the thread started in March of 2002. I'm not sure why this is making headlines now.