Archive for December, 2013
Must we randomize our experiment?
Posted by mark in design of experiments on December 31, 2013
In the early 1990s I spoke at an applied statistics conference attended by DOE gurus George Box and Stu Hunter. This was a time when Taguchi methods had taken hold, which engineers liked because the designs eschewed randomization in favor of ordering by convenience–with hardest-to-control factors changed only once during the experiment. I might have fallen for this as well, but in my early days in R&D I worked on a high-pressure hydrogenation unit that, due to risks of catastrophic explosion, had to be operated outdoors and well away from any other employees. (Being only a summer engineer it seemed that I was disposable.) Naturally the ambient conditions varied quite dramatically at times, particularly in the Fall season when I was under pressure (ha ha) to wrap up my project. Randomization of my experiment designs provided me insurance against the time-related lurking variables of temperate, humidity and wind.
I was trained to make runs at random and never questioned its importance. Thus I was really surprised when Taguchi disciples attending my talk picked on me for bothering to do so. But, thank goodness, Box had already addressed this in his 1989 report Must We Randomize Our Experiment. He advised that experimenters:
- Always randomize in those cases where it creates little inconvenience.
- When an experiment becomes impossible being subjected to randomization
- and you can safely assume your process is stable, that is, any chance-variations will be small compared to factor effects, then run it as you can in non-random order;
- but, if due to process variation, the results would be “useless and misleading” without randomization, abandon it and first work on stabilizing the process;
- or consider a split-plot design.
I am happy to say that Stat-Ease with the release of version 9 of its DOE programs now provides the tool for the compromise, as Box deems it, between randomizing or not, that is—split plots. For now it is geared to factorial designs, but that covers a lot of ground for dealing with hard-to-change factors such as oven temperature in a baking experiment.* Details on v9 Design-Expert® software can be found here http://www.statease.com/dx9.html along with a link to a 45-day free trial. Check it out!
*For a case study on a split-plot experiment that can be easily designed, assessed for power and readily analyzed with the newest version of Stat-Ease software, see this report by Bisgaard, et al (colleagues of Box).
Fake knee surgery shows it not really being needed
As reported here in today’s Wall Street Journal, Finnish surgeons split 146 patients with meniscus tears into two groups and ‘scoped them all, but only half had their cartilage removed. The remainder—the control group—underwent all the same post-operative processes and thus remained in the dark that they really did not get the full procedure. The end results showed that any advantages to this ‘partial meniscectomy,’ which purportedly accounts for $4 billion in annual medical costs on the USA alone, are relatively small and short-lived.
Naturally, an independent orthopedic surgeon asked by WSJ to assess these results did not agree that the arthroscopic procedure might be overdone, even though a previous study showed physical therapy to be just as effective for patients with somewhat similar knee problems.
Without strong affirmative evidence from double-blind studies such as this one, I myself am leery of just accepting any given surgeon’s advice to press ahead with a procedure. As Teppo L.N. Järvinen, co-author of the Finnish experiment, says:
Doctors have a bad tendency to confuse what they believe with what they know.
Cyclists wearing more visible clothes just make it easier for motorists to target them
Tuesday’s Health & Wellness section of Wall Street Journal passed along distressing news for folks like me who like to take a spin on their bicycle. New research by scholars at University of Bath and Brunel University* suggests that wearing noticeable clothes not only did nothing for getting motorists to back off, but when cyclists wore a “POLITE notice, Pass Slowly” vest, they were more likely to be harassed.
Perhaps wearing camouflage like this fellow shown here this fellow shown here might be the way to go.
*Detailed here by IrishCycle.com
White squirrel on fresh snow
On my triple-long commute into town this morning I enjoyed my scenic route along County Road B2 in Roseville. One may as well make a savory slurpee out of the snowstorm as sit in traffic stewing. This route paralleled the gob-stomped Highway 36 so moving along the stopped traffic made this bypass all the more satisfying. I happily paid heed to the calming advice of the mellifluous public radio announcer that “you will get to where you going in due time.”
While admiring the flakes falling on the white-bedecked urban forest I was startled by a clump of snow scurrying down a tree trunk and then disappearing when it hit the deck. It was an albino* squirrel who managed to survive standing out all summer amongst all the greenery. Cool!
You really can see a lot just by looking—to paraphrase Yogi Berra.
*Not being close enough to look it in the eye, I cannot be sure it wasn’t a plain old white squirrel (black eyed) but from the map posted at this website , my guess is it’s an albino. If so, that’s very rare—only 1 out of 100,000 squirrels exhibit this trait, according to my research.