When all you have is a hammer…

…everything looks like a nail.

Daniel Lakens, the 20% Statistician, takes a rare but easy shot at statisticians and null hypothesis significance testing.

Our statistics education turns a blind eye to training people how to ask a good question. After a brief explanation of what a mean is, and a pit-stop at the normal distribution, we jump through as many tests as we can fit in the number of weeks we are teaching. We are training students to perform tests, but not to ask questions

He defines

…the Statisticians’ Fallacy: Statisticians who tell you ‘what you really want to know’, instead of explaining how to ask one specific kind of question from your data.

My favorite is the two-tailed test of the difference of two means, which can provide evidence that the two are different, but not that they are (nearly) the same.  My runners up are goodness-of-fit tests, which do no such thing.  Sometimes I feel like I’m selling the researcher’s version of Snake Oil, rather than teaching sound data analysis and interpretation.

Lakens closes with an excellent addendum, a reference to David Hand’s Deconstructing Statistical Questions,  which goes into much more detail.

Advertisements

A Tinned Oyster Treat

Some time ago, I promised I’d report on my attempts at recipes from Barbara-Jo McIntosh’s Tin Fish Gourmet.  As usual, I didn’t read the cookbook so much as fixed recipes, but as more of a guide.  So I combined elements from two different recipes, “Christmas Eve Oysters” (p 82) and “Shrimp and Spinach-Stuffed Tomatoes” (p 133).  The result is delicious.

TinFishGourmet

So here’s my first offering: Oyster-Stuffed Tomatoes.

  • 8 Campari tomatoes
  • one mushroom
  • one green onion
  • one tablespoon capers
  • 1/4 cup grated parmesan cheese
  • one 3 1/2 oz can smoked oysters

With a paring knife, cut off the tops of the tomatoes, removing the stem.  Then use a melon baller to scoop out most of the flesh of the tomatoes (save this for your soup or sauce stock).  Dice the mushroom to pea size, and slice the green onion finely.  Mix mushroom, onion, capers, cheese, and oysters in a bowl, using the oil from the tinned oysters to moisten the mixture.  Spoon mixture into the tomatoes.  Place stuffed tomatoes in a shallow, foil-lined pan, and broil for 10 minutes.  Serve immediately.

My wife’s only complaint was that the tomatoes should have been bigger, with more stuffing.

 

Seven Pillars

Wisdom hath built her house, she hath hewn out her seven pillars.  –Proverbs 9:1

I just finished Stephen Stigler’s The Seven Pillars of Statistical Wisdom, and I’m daunted–and embarrassed that I waited so long to read it.  Stigler gives us a structure and taxonomy to statistical thinking* that gives us the “big picture” of statistics.

StiglerSevenPillars

Quite a difference from the descriptives-to-inference-to-models approach that most textbook authors follow.  This is making me rethink how I approach my introductory courses, especially those for statistics majors.  I’m starting with a baby step: adding the (inexpensive, paperbound) book as a required reading in my statistical research methods class.

*the 7 pillars: aggregation, information, likelihood, intercomparison, regression, design, and residual (and that’s just the table of contents!)

Deus ex Machina, on steroids

…is the tagline I’d use to describe Stephen Baxter and Alastair Reynolds’ The Medusa Chronicles, the startling sequel to Arthur C. Clarke’s short story “A Meeting with Medusa.”

TheMedusaChronicles

Baxter and Reynolds are up to their usual tricks of piling wonder atop wonder in their usual over-the-top scenarios, while cleverly maintaining Clarke’s style and tone, AND sneaking in episodes strongly reminiscent of 2001, A Space Odyssey.  An added bonus is the introduction of a “new physics” based on the Mach Principle, which is still puzzling serious researchers today.

…local physical laws must be shaped by the large-scale structures of the universe.  And it is meaningless to talk of the behaviour of an object in isolation, without relation to the rest of the universe.  This was 90’s insight.  From that beginning, 90, and a group of others, developed a new kind of physics–from first principles, based only on observation and philosophy.  (The Medusa Chronicles, p. 99)

Houston, We Have a Solution

heb-disaster

Long-time south Texas residents swear by the H-E-B grocery chain for value, selection, quality, and always being well-stocked.  These guys are supply-chain ninjas; we see groceries, they see a logistics network.  And they always step up in emergencies; Houston may be their finest hour to date.

Tip from American Digest.

Stanford Invents AI Gaydar, Flubs Write-Up

Yilun Wang and Michal Kosinsksi, researchers at Stanford’s Graduate School of Business, have developed a neural-net classifier that purportedly detects sexual orientation (in caucasians).
FacialRecognition
The authors report an avalanche of experimental results, and claim the classifier can “correctly distinguish between gay and straight men 81% of the time, and 74% for women.”  OK, that’s the sensitivity of the gadget.  What about specificity, i.e. how well does it correctly distinguish folks who are not-so-gay?  Without that second number (as well as an estimate of prevalance), it’s not possible to estimate the false positive and false negative rates for this thing.  Very important, if some of the more Orwellian applications mentioned by the authors come to pass.
I give the authors a “C,” for incomplete work.
Update: Dan Simmons, writing at the Andrew Gelman blog, writes a rambling, fascinating takedown of this “research,” from both the scientific and MSM points of view.  Based on just the statistical problems, I’m changing the grade to a “D-.”

You can see a lot just by looking*

Any lawyer or successful bureacrat will tell you to never ask a question you don’t already know the answer to.  Some folks aren’t that smart:

It was a strange moment of triumph against racism: The gun-slinging white supremacist Craig Cobb, dressed up for daytime TV in a dark suit and red tie, hearing that his DNA testing revealed his ancestry to be only “86% European, and … 14% Sub-Saharan African.”

The studio audience whooped and laughed and cheered. And Cobb — who was, in 2013, charged with terrorizing people while trying to create an all-white enclave in North Dakota — reacted like a sore loser in the schoolyard.

“Wait a minute, wait a minute, hold on, just wait a minute,” he said, trying to put on an all-knowing smile. “This is called statistical noise.”

Does this make Cobb–to use a white nationalist phrase–a “self-hating race traitor?”

I’m reminded of Holmes in “A Scandal in Bohemia” saying “It is a capital mistake to theorize before one has data.”

Tip from Tom Knighton at PJ Media, by way of Sarah Hoyt at Instapundit.

*the Yogi never disappoints.

UpdateThis cuts both ways.

I found out I was White. Not just 13% White, my husband’s percentage when he too completed the ancestry composition report. Not just 25% White, since the average amount of DNA in an African American’s genome traced back to West Africa is about 75%. I was damn near 1/3 White. That’s significant.

Of course, this nice lady fellow**can always fall back on the “one drop” principle, that standard promoted by 19th century white slavers and 21st century African Americans.

Tip straight from the Instapundit himself.

**Speed reading is one of my more egregious faults.