BONUS UPDATE: The statistician Jan de Leeuw, dwindling in old age into philosopher of science.
Follow up to: this comment and to a lesser extent this post.
You’ve probably read Steven Pinker or someone else write about how everyone should learn statistics in school rather than trigonometry (grr, I’m still mad at those transcendental functions). I nodded along, but the fact is I’ve never taken any myself. I took probability in college, but that’s not quite the same thing. Fortunately, there is an online introductory statistics textbook which can be found, appropriate enough at OnlineStatBook.com. From what I checked out, it seems like it would be fine for middle-schoolers without any knowledge of statistics. It has nice simulations and exercises/quizzes that have an explanation available whenever you give a wrong answer. The one failing on their part is that those explanations pop up in windows that cut off the upper portion of the explanation, so you have to copy-paste the off-screen text if you want to read it. This happens in both Firefox and Internet Explorer and resizing does not help. If you want another option, there’s SticiGui, the online text for the first and last statistics class most students at Berkely take. Internet Explorer does not apparently work reliably for it.
In completely unrelated matters, it recently occurred to me to combine something like the Atkins diet with alcohol. This was partially inspired by Agnostics post extolling the virtues of high-fat low-carb diets. Apparently that was just the beginning for him, you can find his later carb posts here. It got me to thinking that my diet has long been carb heavy and perhaps I should make some changes (along with eating vegetables, exercising and getting enough sleep blahblahblah). I’m not interested in losing weight, in fact I could use some more insulation, so I thought some more fat, grease or oil couldn’t hurt. Lactose and peanuts aren’t my friends, so that limits things a bit. Agnostic is lactose-intolerant as well (in addition to being gluten intolerant, perhaps reflecting or causing his non-docile nature), so I appreciate his tips. At any rate, despite what you may have heard in Seinfeld, chicken (unfortunately) does not ferment. Fermentation comes from carb-rich sources (wheat, rice, potatoes, grapes, apples, molasses) and it is precisely the sugar content that motivates the addition of hops to beer. So I would expect alcoholic beverages to have lots of carbs. Judging from chemistry, an alcohol is distinct from a ketone or aldehyde (two types of carbs) so there shouldn’t be anything inherently incompatible with high alcohol and low carb content. The highest alcohol concentrations are found among the distilled beverages, but if you’re an out of practice drinker those aren’t especially pleasant straight. Solution: combine a non-alcoholic high-fat low-carb drink with a more concentrated alcoholic drink to have the same alcohol content per volume as beer but with less carbs and hopefully without a telltale “healthy” taste. I recall the necessity of downing irish carbombs quickly before the Baileys curdles, so perhaps dairy-products aren’t the best bet there. There’s already baconified bourbon, so I don’t think it’s a completely crazy idea. I still need to think of some names for it better than Protein Poteen or Lipid Lunch.
Entitled to an Opinion 
February 26, 2009 at 1:23 am
Electric Lemonade
Ingredients:
Blender almost full of crushed ice
3/4 tub container of Crystal Light Lemonade powder
6 oz. Vodka
How to Prepare:
Mix in blender until slushee consistency. Enjoy!
This goes down much easier than you think it would, so be careful.
FYI, alcohol doesn’t knock you out of ketosis, but it does put ketosis on hold until you burn all the alcohol calories off. Since you are not worried about weightloss, this concern should not be an issue for you.
February 26, 2009 at 2:46 pm
I find it’s really hard to do high intensity exercises without carbohydrates in the tank. If you work out or something similar I think you’ll get sick of it pretty quick.
February 26, 2009 at 4:56 pm
i have done the low-carb (<5g of carbs daily) diet before and i am a big drinker. they say that if you’re going to drink you should drink something hard, tequila works well for me. you can always buy the ketostix at a pharmacy and test yourself to see if you’re on track.
bacon infused vodka makes a great bloody mary.
and i would look at the MIT OpenCourse stuff if people want to try learning about statistics.
February 28, 2009 at 2:17 pm
Thanks for the suggestion, theOther. I don’t know if automatically generated links are permanent, on the off-chance that they’re not I’m preserving the one to Online/free statistics and econometrics texts.
March 3, 2009 at 3:30 pm
I don’t know why one would want to mix concentrated protein and booze when one could simply have roast beef or lamb and claret – the perfect combination.
Do you know why red wine goes with red meat? It isn’t just an arbitrary convention. The tannin in the red wine bonds with the iron in the meat, rather than with that in your tongue. For this reason, when drunk by itself red wine will often have an astringent taste (puckery, like alum). When drunk with red meat this characteristic is moderated, and you enjoy the other flavors of the wine.
It seems to me that the present importance of statistics owes itself in part to the advent of quantum physics and to the necessity of using it in the social sciences.
Statistics isn’t really very important to the parts of physics developed before quantum theory, e.g., Newtonian mechanics, optics, or electricity. Analytic geometry and calculus, in which trigonometry plays a significant part, are. In chemistry, yields of reactions often do not amount in practice to what stoichiometry predicts. The old custom used to be just to say that “yield is X% of theory” and leave it at that. Such results are more correctly explained in statistical terms, but this is purely incidental to understanding how the reactions proceed, and chemists got along without it for a long time. These disciplines always used to be framed in terms of unerring laws.
Social sciences do not, of course, operate according to anything like the unerring laws of physics or chemistry. When a physicist applies the gas laws, for example, he is able to predict real-world results within a much smaller margin of error than an economist does in applying the techniques of econometric modelling. Statistics provides a respectable allowance for the inaccuracy of the economist’s predictions, and gives him a pretext to claim a place under the mantle of science along with the physicist.
March 3, 2009 at 4:23 pm
Gas laws are the result of a statistical process. Thermodynamics is statistical. The online stat book site links to a page about the results of Galileo’s experiments. Due to human error we will have uncertainty about many things and statistics helps us deal with it.
I think statistics owes its present stature less to quantum mechanics than marketing and other fields that rely on demographic research.
March 4, 2009 at 3:06 pm
Of course it is now the view that gas laws are the result of a statistical process. It is now the view that all scientific knowledge is the result of such a process. Historically, however, that is not how the gas laws, or the laws of motion, Ohm’s law, the various theories of elective affinity in chemistry leading up to the present Redoxreihe (or whatever it is called – I forget – in English), etc., were conceived, understood or stated.
For example, here is a statement from an approximately sixty-year old text on organic chemistry:
“The Modified Rule of Crum Brown and Gibson. If the atom attached to the aromatic nucleus is attached to some other atom by an unsaturated linkage (i.e., by anyu bond which we commonly write as double or triple), then the next entering gruop takes the meta position; otherwise it takes the ortho and para positions.”
Nothing statistical about it at all; it’s as absolute a statement as any of Euclid’s. This was the common character of the physical sciences until relatively recent times. Quantum mechanics, because it requires some statistical assumptions, perhaps made ‘respectable’ the use of statistics in the physical sciences, but I agree with you, and already stated, that the present stature of statistics derives from the desire of social scientists, whose disciplines are necessarily dependent on it, to place them on the same elevated level of intellectual esteem the physical sciences have occupied since the time of Newton, Fontenelle, etc.
March 4, 2009 at 6:58 pm
I mentioned marketing in part because it makes no aspirations to “intellectual esteem”.
March 5, 2009 at 8:36 am
Biology and Agricultural applications drove many of the applications of statistics. RA Fisher who came up with the “p-value” and developed the theory of the randomized experiment was also a geneticist and agronomist. I do not think that most of the standard “tools” of inference (ANOVA, regression, p-value, etc) work very well in the social sciences outside of psychology.
March 5, 2009 at 12:54 pm
I disagree that marketing makes no aspirations to ‘intellectual esteem.’ It is now taught in universities, as is journalism. A century ago, to suggest that either of these activities deserved the dignity of academic disciplines would have been laughed to scorn.
Yet, in the absence of any other, the university has become the fons honorum of our ‘democratic’ society, and rich parvenus today aspire to its recognition in the same way that, ninety years ago, war profiteers who knew Lloyd George aspired to baronetcies.
March 5, 2009 at 8:19 pm
A lot of things are taught in universities, including agriculture, which is as humble as it comes.
Statsquatch, is psychology different because of experiments?
March 6, 2009 at 8:33 am
Yes, I think psychology is different becuase of experiments. Of course psychology also drove most of the development of factor analysis and associated methods like principle component analysis for “differntial” psychology.
March 6, 2009 at 1:53 pm
You folks might enjoy this:
In a recent talk by Google’s chief economist Hal Varian, he says this:
“I keep saying the sexy job in the next ten years will be statisticians.
People think I’m joking, but who would’ve guessed that computer engineers
would’ve been the sexy job of the 1990s? . . .
The ability to take data—to be able to understand it, to process it,
to extract value from it, to visualize it, to communicate it—that’s going to
be a hugely important skill in the next decades, not only at the
professional level but even at the educational level for elementary school
kids, for high school kids, for college kids. Because now we really do have
essentially free and ubiquitous data. So the complimentary scarce factor is
the ability to understand that data and extract value from it.
I think statisticians are part of it, but it’s just a part. You also
want to be able to visualize the data, communicate the data, and utilize it
effectively. But I do think those skills—of being able to access,
understand, and communicate the insights you get from data analysis—are
going to be extremely important. Managers need to be able to access and
understand the data themselves. ”
The full article is at
http://www.mckinseyquarterly.com/Hal_Varian_on_how_the_Web_challenges_managers_2286
March 6, 2009 at 9:08 pm
The cynic in me says managers don’t really want to understand the data. The optimist in me says we will someday find a way to avoid that element of human error. The factory of the future will have just two employees: a dog and a man. The job of the man will be to feed the dog. The job of the dog will be to stop the man from touching anything.
March 7, 2009 at 9:32 am
Predictions economists have made:
1) Buy stocks for the long run
2) Derivatives will improve market
efficiency
3) Abortions will lower the crime
rate
4) Statisticians will one day be as
sexy as computer engineers
If you make enough predictions at least one will be true.
March 7, 2009 at 4:00 pm
In the long run, a market index of stocks is about as good as the average schmo can get. The abortions thing was not a prediction at all, it was an after-the-fact explanation after a bunch of over stuff (improved economy, demographics, tougher sentencing) was accounted for. I don’t know much about finance, but I believe futures (a kind of derivative) are still thought to improve efficiency.
March 7, 2009 at 11:39 pm
I should have known not to be glib on this blog. I doubt statistics will ever be very sexy. The adoption of new methods occurs at a glacial pace. Last week there was a discussion of Akiake’s Information Criterion on GNXP. That statistic has been kicking around for 30 years and it is only now starting to get used regularly. Those plots the geneticist show of the gene expressions of various ethnic groups are constructed with principle components that Karl Pearson developed 100 years ago. Statistics is too conservative to be sexy.
March 8, 2009 at 12:26 am
Perhaps statistics was sexy all along, and we just didn’t realize it until the glasses and pigtails were removed like in a cliched movie. How to deal with the conservatism? Follow Schumpeter’s lead, perhaps it should take up wearing a cape, riding horses and fighting duels.
March 10, 2009 at 4:40 pm
Agriculture is not quite as humble as it seems. It has a long and distinguished literary history, from Cato the Elder’s ‘De agri cultura’ and Vergil’s Georgics (respective archetypes of the hard and the soft pastoral) to Tull’s “Horse-Hoeing Husbandry” (a volume studied by George Washington) and John Taylor of Caroline’s ‘Arator,’ the hard pastoral of the American South.
The ‘improving landlord’ who wanted, or rather, wanted his tenants, to practice enlightened farming methods, was a widespread phenomenon beginning in the seventeenth century, and in their ways both Thomas Jefferson (who designed a ‘scientific’ plough, deprecated by his cousin Randolph of Roanoke) and Lord Emsworth of Blandings Castle, with his prize pumpkins and pigs, are reflections of it.
Still, when ‘agricultural colleges’ were first established in the United States, they attracted just the sort of derision that Mencken later applied to journalism schools, and which the elevation of marketing to an academic discipline certainly deserved, and probably received from someone. One nineteenth century supposed the primary effect of agricultural schools would be to teach their pupils to speak more elegantly to their mules. Rather than ‘gee’ or ‘haw,’ they’d say ‘halt, Maria; pivot and proceed.’ On a somewhat different note, the Nashville Agrarians, who were mostly academics associated with Vanderbilt University, were derided by some contemporaries as impractical romantics living in the past, who moreover would not know one end of a mule from another.
March 10, 2009 at 7:12 pm
I think your last paragraph makes my point.
October 12, 2009 at 3:09 am
ITS CRAP!!!!
October 12, 2009 at 3:10 am
you all lick ball sack!!! Hardcore!!!
October 12, 2009 at 3:10 am
Go get a life!!!
October 12, 2009 at 3:11 am
I want to lick you
October 12, 2009 at 8:15 pm
For some reason I suspect your last nym is no more accurate than your first. It’s nice and generic sounding though.
October 13, 2009 at 1:51 am
Thank you i would like to find you and your family then make you lick my hairy ball sack!!!
So fuck off you sad little man!!!!