Archive for January, 2016
Too many dogs at farmers markets?
Today’s “Gray Matter” column in the New York Times provides an exceptionally well-balanced report that casts doubt on the healthiness of food from farmers markets—read it here. What caught my eye is how the author, a professor at University of Minnesota (my alma mater), lays out a number of positive correlations (being careful not to conclude causation) between farmers markets and various food-borne illnesses, including one attributed to the ‘droppings’ from dogs and the like. But the thing I most admire is him admitting to “a number of dogs that did not bark”, i.e., a number of outbreaks that did not show a statistically significant connection to farmers markets.
This suggestion of possible health issues with farmers markets is heavily hedged—very possibly it will not be borne out by subsequent research. Nevertheless it would only be prudent to thoroughly wash locally grown and sold produce.
Sine illusion makes peaks and valleys on graphs look overly variable
An article in the latest Journal of Computational and Graphical Statistics (JCGS, Vol 24, Num 4, Dec 2015, p1170)) alerted me to a fascinating misperception called the “sine illusion” that causes misinterpretation of trends in variability. See it nicely illustrated here by vision researcher Micheal Bach. The JGCS, Susan VanderPlas and Heike Hofmann, detail “Signs of Sine Illusion—Why We Need to Care” and provide methods to counteract its misleading effects.
If you see a scatter plot that goes up and down with seemingly large scatter at the bends, get out a ruler to get the true perspective. That is my take home message for those like me who like to be accurate in their assessments of data.
“The illusion is explained in terms of a perceptual compromise between the vertical extent and the greater overall dimensions of the section at the turn of the sine-wave figure.”
– RH Day and EJ Stecher, “Sine of an illusion,” Perception, 20; 1991, 49–55.
A knotty problem—how to keep track of stuff without computers or even pen and paper
Posted by mark in history, Uncategorized on January 3, 2016
The New York Times reports today on the recent discovery of several knotted string records, called khipus, that ancient Incas used to record things such as the colorful potatoes I photographed at a Peruvian market. From what I saw on my travels there—see this blog on Incan agriculture experiments, a great deal of food must have been produced and stored.
Based on this Times picture I suspect these “mops that have seen better days”, as George Gheverghese Joseph, a mathematics historian at the University of Manchester, U.K., put it, must be a bit easier to untangle than Christmas lights. But then there remains the problem of deciphering them.
Thus far researchers have picked up on mathematical aspects of the khipus. However, the latest trove of colored strings provides a chance at figuring out the Incan scheme for identifying what was being counted. Here is where database capability and statistical methodology comes in handy.
I amazes me how all of the technology we now have at our disposable is challenged by methods developed 600 or so years ago. Hats off to Incan ‘thinken!
“Many now think that although khipu probably began as accounting tools, they had evolved into a writing system—a kind of three-dimensional binary code, unlike any other on Earth—by the time the Spanish arrived.”
— Cracking the Khipu Code Science magazine.