Speed of light exceeded (astounding!)? Or was it measurement error?
This morning I read this NY Times news that European physicists measured neutrinos at 0.0025 percent above the speed of light. If so, it may be only a matter of time before you can send yourself a telegram to not do whatever you did that you’ve always regretted and, by the way, to please invest a thousand dollars in Microsoft, Facebook or the like (depending on the timing).
Years ago I visited Mount Wilson Observatory in California with my son Hank. See me pictured by their two domes that house 60 and 100 inch telescopes; respectively. This was the center for landmark experiments on the speed of light as detailed in this Wikipedia article. Obviously measurement error made this a very difficult.
Being a skeptic, and seeing that a similar experiment* found neutrinos whizzing about at the speed of light, but not beyond that, I was going to advise caution. However, Hank gave me the heads up to today’s xkcd cartoon (click the image to make it bigger and more readable). I think this guy has got a better idea.
*Done with a group at the Soudan Underground Laboratory here in Minnesota. They first did physics experiments there, in an abandoned iron mine, in 1980. I featured this in a retro young-adult techno/adventure/mystery/thriller called The Secret of the Wolf Ring (Amazon, Kindle Edition).
Blended learning for math & stats
Posted by mark in Basic stats & math on September 18, 2011
Check out this intriguing YouTube video by Khan Academy proving the Pythagorean Theorem:
Now imagine grade schoolers being lectured like this at home and then spending their time in class following up one-on-one or in small group sessions with the teacher. See this report from a 7th grade math teacher in California who takes advantage of this “blended learning” approach. As face-to-face time with educators becomes ever-more expensive, expect more-and-more use of asynchronous web-based training like this. That’s what I foresee. Don’t you?
Proofing Blackbeard’s rum
Posted by mark in pop, Uncategorized on September 11, 2011
Being only about a week from this year’s Talk Like a Pirate Day this Atlantic Monthly article (read belatedly from a backlog of magazines) about Gunpowder on the Rocks caught my eye. I like the idea of setting a drink on fire and then drinking it, as Blackbeard did to impress his pirate crew.
It turns out that this is a practical test of rum to ensure it hasn’t been watered down by a ne’er-do-well hornswoggler, as you can see in this video by experimental archaeologist Jeff Lindow. After watching this, I decided not to try this at home as it would no doubt shiver my timbers. However, if it gets cold enough this winter, I might consider a swig of this gunpowder-infused Man O’War rum. Yo ho ho!
It may pay to make your product less than perfect!
Posted by mark in Consumer behavior, Uncategorized on September 4, 2011
I once analyzed data from a designed experiment that quantified consumer distaste for flaws in chocolate-covered cherries. This was a very rewarding project – lots of free candy! It also produced a counter-intuitive result: People preferred boxes with a few upside-down morsels. I figure this is akin to a beauty mark adding to the enticement of a model or actor. This article on “When Blemishing Leads to Blossoming”, published online by the Journal of Consumer Research confirms that under specific circumstances, a flaw makes a product more attractive. For example, in one experiment (highlighted in the July 16 issue of Wall Street Journal) the researchers (Danit Ein-Gar, Baba Shiv, Zakary L. Tormala) offered either perfect or slightly flawed chocolate bars to several hundred relaxed (strolling around) or stressed (rushing to exams) college students. I searched out the results and reproduced them in this interaction graph from Design-Expert® software. It seems to me that this surprising effect, presuming it’s real, provides yet another devious opportunity for marketing mavens to make us buy stuff. One thing I might advise is that you never buy anything when you are in a hurry.
Mind-reading fish know I am out to catch them
Last week I enjoyed a relaxing sojourn up in the north woods of Wisconsin. The resort encompasses its own pristine pine-ringed lake featuring a 26-foot fishing hole. Just before I headed off for my vacation I read this Scientific American report on The Mind-Reading Salmon: The True Meaning of Statistical Significance
. Although I think they meant to be disrespectful of p-values in this case, my feeling, based on empirical evidence from a large sample size – hundreds of unsuccessful casts of my lure around the shore and over the hole, is that some fish living in isolated areas have developed mental telepathy. How else do they avoid being caught?
PS. Here’s a picture of me in happier days at a different lake last summer. My brother-in-law insisted that the first one to catch a crappie would have to kiss it. Evidently this fish thought it might be fun to try, knowing I’d then release it back into the lake.
The chaotic, yet regular sounds of weeping waters and lapping waves inside Lake Superior sea caves
Earlier this month I enjoyed a wonderful sail out of Cornucopia, Wisconsin to Lake Superior’s Mawikwe Sea Caves — described nicely here in a pictorial blog by the Howder family.
Mawikwe means “weeping woman” in Ojibwe. Due to heavy rains in the days leading up to our voyage, the caves were weeping steadily, as you can see and hear in my video. I was fascinated by the cacophony of dripping water combined with the galumphing of the waves into the baby caves at water level. It provided a pleasing mix of randomness and rhythm. Turn up your volume and listen for yourself.
PS. By the way, I learned that by yelling into a sea cave you can (pun intended!) duet yourself.
A Santa Claus machine
Someone just sent me this amazing video of a 3D printer copying a crescent wrench – moving parts and all. The company featured, Z-Corp, is a Stat-Ease client. See this case study showing how their engineers used response surface methods (RSM) to discover a small window of operability.
Another client of Stat-Ease, Stratasys, also offers rapid-prototyping machines, but they make use of another technology called fused deposition modeling (FDM) – explained nicely by this schematic from Ferris State University. I’ve seen these machines at work. They run from plastic line similar to what’s used in a weed whacker. Based on a computerized blueprint, this material (suitable for functional parts, not just prototypes) is melted layer-by-layer into complex shapes. Check out this full-scale turboprop engine produced by FDM.
The next step will be the development of machines that can make whatever you need out of whatever you happen to have nearby that can be shoveled in the hopper. Then people who really want to get away from the crowds can rocket off to any old unoccupied planetismal and set themselves up with house and home.
“It’s possible to imagine a machine that could scoop up material – rocks from the Moon or rocks from asteroids – process them inside and produce just about any product: washing machines or teacups or automobiles or starships. Once such a machine exists it could gather sunlight and materials that it’s sitting on, and produce on call whatever product anybody wants to name, as long as somebody knows how to make it and those instructions can be given to the machine. I think the name Santa Claus Machine for such a device is appropriate.”
– Physicist Theodore Taylor (1978)
Commuting by car or bike: Studies by UK statisticians
This month’s issue (June 2011, volume 8, issue 2) of the Royal Statistical Society magazine Significance [motto “statistics making sense” 🙂 ] features two intriguing articles on commuting.
The first one details Martin Griffiths, a math lecturer at U Manchester, “trying to pull out of the drive” (pp 89-91). This poor fellow must wait upwards of 2 minutes just to get on the road from his property. Imagine the frustration of a somewhat random stream of cars blocking your way out. Then, just as you see a gap, another car comes along to fill it. Griffiths provides a very impressive formula for average wait time based on a Poisson distribution. He factors in the average number of cars passing by as well as the time taken to pull out into the flow. The bottom line: It does not pay to be timid. You’d best mind the gap (inside joke for anyone who’s traveled London’s subways) and make a move!
The second study* comes from a biker, Dr. Jeremy Groves, who spends up to 2 hours or more commuting to his work at Chesterfield Royal Hospital. Thankfully his ride gets considerably shorter in summer when he needn’t wear baggy outerwear, which creates a real drag (Groves estimates 30% more wind resistance). This cycling enthusiast bought a new bicycle recently – one that featured a carbon frame, as opposed to the steel one he’d bought second-hand. Being a fan of randomized (“randomised” in UK spelling) trials, Groves completed a series of runs with one or the other of his bikes – measuring the times taken for the ride from his home in Sheffield to his work at hospital. Seeing his run chart starting off very raggedly at the high end in January, I transcribed only the latter 28 runs (14 of each) for the chart shown. 
Obviously from the overlap in the least-signficant-difference (LSD) bars the results remain inconclusive. (If you have trouble seeing this, click the graph for a larger view of it.) The difference is less than 1 minute in favor of the very costly carbon-framed bicycle. Given a 3.5 minute standard deviation under summer conditions, it would take 400 total runs (200 each) to resolve whether this is a true advantage, according to a power calculation I did with the aid of Design-Expert® software. Dr. Groves has moved on to another experiment for this summer – he plans to randomly load a 4 kg weight on his bicycle (the carbon one, I presume). Aside from being a glutton for punishment, I suppose this fellow wonders how much it slows him down when he must carry in his laptop.
* “Bicycle weight and commuting time: A randomised trial,” pp95-96.
Fun summer-time experiment: Super-cool beer so it instantly freezes solid
It turns out that if a bottle of beer is put in the freezer for long enough, and then removed while it is still liquid and, lastly, given a sudden shock, the beer will instantly freeze solid. I saw this confirmed by the pop-science TV show Mythbusters in their episode 153, originally aired on 11/10/10. Based on trial-and-error experimentation their Build Team found that 3 hours of cooling time sufficed to create the delightful phenomenon. See the instant beer-freeze demonstrated by EasyBarTricks.com here. For detailed instructions on how to try this at home or in a classroom, plus a nerdy explanation (think PVNERT) by physics and math teacher Daryl Taylor, check out this website.
Of course I had to try this for myself. However, not being one who ever leaves well-enough alone, I tried light beer (Miller, bottled in clear glass) side by side with the recommended Corona – two of each. Could this be a factor (light versus regular brew)? After being careful to wait at least 3 hours for the quartet of brews to super-cool, I brought them out for a family party. Two of the beers crystallized when smacked on our kitchen cutting board, but the other two did not. Here’s a twist, though: None of the bottles were uncapped first, so how does that gibe with the PV-nerd’s explanation?
Alas, one of the light beers crystallized and the other did not – ditto for the Corona, so my results, albeit semi-successful, were indeterminate on the issue of light vs regular brews. The good news is that we salvaged two bottles of beer (the frozen ones become undrinkable).
Feel free to weigh in with your theories and experimental results from this beer trick. One thing I learned from my first try – a lot more beer would be good, along with a walk-in freezer (or the backyard in mid-winter).
Strategy of experimentation: Break it into a series of smaller stages
Posted by mark in design of experiments on June 20, 2011
Tia Ghose of The Scientist provides a thought-provoking “Q&A” with biostatistician Peter Bacchetti on “Why small is beautiful” in her June 15th column seen here. Peter’s message is that you can learn from a small study even though it may not provide the holy grail of at least 80 percent power.* The rule-of-thumb I worked from as a process development engineer is not to put more than 25% of your budget into the first experiment, thus allowing the chance to adapt as you work through the project (or abandon it altogether). Furthermore, a good strategy of experimentation is to proceed in three stages:
- Screening the vital few factors (typically 20%) from the trivial many (80%)
- Characterizing main effects and interactions
- Optimizing (typically via response surface methods).
For a great overview of this “SCO” path for successful design of experiments (DOE) see this detailing on “Implementing Quality by Design” by Ronald D. Snee in Pharm Pro Magazine, March 23, 2010.
Of course, at the very end, one must not overlook one last step: confirmation and/or verification.
* I am loathe to abandon the 80% power “rule”** but, rather, increase the size of effect that you screen for in the first stage, that is, do not use too fine a mesh.
** For a primer on power in the context of industrial experimentation via two-level factorial design, see these webinar slides posted by Stat-Ease.