Steve Kass

In scary news today: Stroke Patients Getting Younger, Stroke Rising Among Young People, and so on.

Apparently some researchers reported that some number went up recently, and the number that went up had something to do with stroke and something to do with 15-44 year-olds.

Sounds like a good excuse for a rousing chorus or two of Fire in the Theater! Obesity! Diet Soda! You’re Gonna Die!, no?

No.

The number reported to have gone up recently is not the total number of strokes among 15-44 year-olds, nor is it the rate (per 10,000 people, for example) of strokes among people that age.

The number that went up recently is the rate of strokes in 15-44 year-olds as a fraction of all hospitalizations for that age group. Not an easy quantity to conceptualize. But when a quantity is hard to conceptualize, you aren’t automatically allowed to grab a “you may pretend it’s something else.” pass and lie with impunity. (Do these same journalists give up and write “Boxer” if they can’t spell “Feinstein”?)

Maybe the stroke rate among 15-44 year-olds is not going up.

It could be that hospitalizations of 15-44 year-olds for reasons other than stroke are going down. Maybe hospitals are more and more likely to list multiple reasons for hospitalization than in the past. Maybe many former headaches are now deemed strokes (thanks to the proliferation of imaging tests). Either of these trends would make the numerical rate of stroke diagnoses per 10,000 hospitalizations go up without reflecting an increase in stroke.

Maybe a lot of things. Maybe the rate of stroke is going up among young people. Which might be scary. Or not. It’s possible more and more diagnoses of stroke are insignificant — no worse than a bad headache. Just because “stroke” sounds scary doesn’t mean there can’t be innocuous kinds of stroke.

Unfortunately we don’t know from today’s irresponsible scramble to turn numbers into fear.

A theretofore charming day, spoiled in the blink of an eye, because I caught sight of the link “David Brooks: Obama’s Very Good Week” in the Times. Brooks isn’t known for sarcastic titles, and it’s been nobody’s good week in Washington. A $25,000,000,000 tax windfall (yes, 25 billion) for the richest 0.3% (yes, less than one third of one percent), and DADT, in particular.

I clicked, but I didn’t read the article. I’m trying hard not to read everything from start to finish that I know in advance will be crazymaking. I wasn’t about to backslide with the likes of Brooks, so I applied the How To Read a Ross Douthat Article approach.

The HTRaRDA approach is this: Type Control-End (preceded by SINGLE PAGE, if available) to reach the bottom of the article, and then read the last screenful. Do not pass Go! You will discover the offensive conclusion without wading through the invidious, lame, emetic screensful that invariably, insidiously, precede it.

Unfortunately, sometimes the insidious run-up extends into that last screenful, and it did today. It was impossible to ignore, because it included numbers.

Warning: The David Brooks quote below is a lie.

According to the most recent Gallup numbers, 67 percent of independents and 52 percent of Democrats support extending all the tax cuts.

Warning: The David Brooks quote above is a lie.

It was painful to follow up on this quote, but I’m still holding onto the belief that it would have been worse to have read Brooks in toto.

On December 1, Gallup reported that “Forty percent [of Americans] want Congress to maintain the tax cuts for everyone, while 44% support setting limits on how much of wealthy Americans’ income is eligible for the lower rates.” In particular, 18% of Democrats preferred keeping the tax cuts for everyone (when offered two alternatives: setting limits or letting the cuts expire entirely). Those can’t be the “most recent Gallup numbers” to which Brooks refers, because 18 doesn’t equal 52, even with rounding. (Maybe at the Times it does, but not on my blog.)

Here are the actual Gallup numbers to which Brooks refers. Over the past weekend, Gallup asked this question: “Suppose that on Election Day you could vote on key issues as well as candidates.  Please tell me whether you would vote for or against a law that would do each of the following.  Would you vote for or against a law that would extend the federal income tax cuts passed in 2001 and 2003 for all Americans for two years?” Fifty-two percent of the Democrats, 67% of independents, and 85% of Republicans surveyed said they would vote for such a law.¹

Quiz. Given three choices similar to those before Congress (A. extend cuts for all; B. extend cuts but only up to a dollar limit for the wealthy; C. let tax cuts expire), Democrat’s least preferred option is to extend all the tax cuts. (Breakdown: 18% answered A, 21% answered C, and 55% answered B). Is it fair to say that 52 percent of Democrats support extending all the tax cuts?

The correct answer is NO.

Kudos to Gallup for the careful and precise phrase to describe the option before Congress: “limits on how much of wealthy Americans’ income is eligible for the lower rates.”

Shame on Brooks for vulpigerating.

¹ I apologize for previously calling the Republican party the “Party of No.” I should have more precisely called them the “Party of No, except when it will make the rich richer.”

Last night, as a few of us not backing up Ray Davies in Philadelphia gathered for dinner in Hoboken, I spotted this holiday mispostrophe (ssp. dyspostrophe).

It probably wasn’t intentional on Trader Joe’s’s part, but the mispostrophe distracted me from the numbers on the box — especially 24 and 50. But only briefly; the pressing question quickly loomed.

If 24 Milk Chocolates weigh 50 grams altogether, aren’t they too small?

The appropriate comparison was obvious: M&M’s®. Little did I suspect it would be something of a challenge to find out the true weight of one regular M&M.

Disregarding outliers like “I think it is about 15g; 15 grams is perhaps the answer,” answers on the web (to the question of an M&M’s weight) generally fell into two camps. There was a handful of a-bit-less-than-a-gram answers, like “There are about 500 Plain M&M’s per pound,” and there was also a handful of around-2-grams answers, like “After an experiment, of weighing M&M’s, here were the results. 1) 2.208 g 2) 1.882 g 3) 1.904 g 4) 2.438 g.”

After considerable “research,” but no direct measurement, I’m swayed, not by any attestations of milligram precision, but by the preponderance of evidence [and 1] that one regular M&M weighs a bit less than a gram. Which conclusion is consistent with my personal experiences as a candy sorter (when I can find an uncluttered flat surface, which isn’t very often).

From the web’s many M&M Q&A (or should I say Q&“A”?) a few examples:

• Q: What is the weight of one M-and-M candy? [link]
  A: I think it is about 15g; 15 grams is perhaps the answer
   
• Q: How much does an M and M weigh? [link]
  A: When we counted the number of M&M’s in a 12.6oz bag, we got 404, which means there are 32.06 M&M’s/oz, which means that each M&M weighs 1.13 grams. [SK: If you divide backwardsly, perhaps. Otherwise, each M&M weighs (on average) about 0.88 grams.]
   
• Q: How many m&m’s do you reckon are in 7oz? I’m ordering custom m&ms, and they come in 7oz bags. I need about 1000 m&ms, total. how many bags should I order? [link]
  A1: [Best Answer] 10 bags, maybe around 75 or 100 in each bag. [SK: Better safe than sorry.]
  A2: 2 or 3.
   
• Q: How much does a single plain m&m weigh? [link]
  A: After an experiment, of weighing M&M’s, here were the results. 1) 2.208 g 2) 1.882 g 3) 1.904 g 4) 2.438 g.

As for the pressing question, I’ll cautiously answer it “No” and hope Toby and Theo agree. Two or three M&M’s-worth of chocolate every day for most of a month — for those endless days, those sacred days, believe me — is not so bad. Despite anyone’s opinion that one serving of M&Ms comprises 208 grams (and 1023 calories).

“Screaming performance” is a label journalists seem to apply judiciously. The vast Google News Archives records only 28 outbursts having met the high bar. However, a trend is clear. Did Howard Dean kill screaming? Is technological singularity just around the corner?

CNN doesn’t get nearly as much attention as it should, given that they’re easily as irresponsible as their buddies at [expletive deleted]. Today they’re fanning the fires about “illegal immigration,” the current euphemism for people we don’t like because they’re brownish and speak another language especially Spanish. Writer Arthur Brice devotes a big chunk of a 900-word article on CNN.com today to a discussion of “anchor babies,” the current not-so-euphemism for babies of people we don’t like because they’re brownish and speak another language especially Spanish. Here’s my brief rant on the article, “Report: 8 percent of U.S. newborns have undocumented parents.”

Before ranting, though, let me be one of the first to greet all these new and beautiful U.S. citizens: “¡Welcome, and bienvenidos!”

This rant has two parts. First, let’s see what “have undocumented parents” means, so we know more about this 8% on whom the goons will be spreading their invective. The phrase shouldn’t mean anything other than “have undocumented parents,” but somehow it does, and not just because of headlinic license. It means “has at least one undocumented parent.” Here’s the relevant wording (emphasis mine) from the Pew report Brice describes:

A child has unauthorized immigrant parents if either parent is unauthorized. A child has U.S.-born parents if all identified parents are U.S.-born.

Well, that’s stupid. The asymmetry reminds me of the definition of Colored, as in for the purpose of what school you can go to, what train car you can sit in, and what drinking fountain you can use, and, before the 14th amendment was ratified, as in whether you were a U.S. citizen, more or less.

Next thing you know, today’s goons who want to abridge the Fourteenth Amendment will find a way to damn not only these youngsters but sus hijos y nietos también, no matter what, probably because fuck the Constitution and Bill of Rights, God tells them to.

Not to mention that “[s]ome pregnant women from other countries are traveling to the United States to give birth and then taking their babies back home to raise them as terrorists that would return to attack America,” a concern raised by Texas state representative Debbie Riddle, “a Republican,” that Brice thought fit to pass on.

Tattooing the letter U on them to start, maybe? (You can bet they’d have no problem paying for that medical procedure with government dollars.)

Part 2: The word “anchor babies” doesn’t appear in the Pew report, but instead of leaving it out of the article entirely, Brice fills us in. He knows that more people will read an article if it’s about anchor babies.

“Babies born to illegal alien mothers within U.S. borders are called anchor babies because under the 1965 immigration Act, they act as an anchor that pulls the illegal alien mother and eventually a host of other relatives into permanent U.S. residency,” says an organization called The American Resistance, which has described itself as “a coalition of immigration crime fighters opposing illegal and undocumented immigration.”

Minor partial credit to Brice for using the past tense when mentioning The American Resistance, but he forgot to mention that they are “no longer an active – or updated – Website or effort,” and haven’t been since 2006, according to — well, themselves, in a message they left on the web four years ago. The fact that Brice names them at all is goofy, to put it kindly. There are dozens of non-moribund organizations he could have called up. A Youtube link to a [expletive deleted] broadcast from within the last week, maybe.

That’s all. Have a nice week.

Internet news aggregator robots never leave me alone. Internet news aggregator robots, never leave me alone.

Every day or more, one of the news aggregator robots gets both my attention and my goat. Here’s one of today’s missiles: “CDC: Most Teens Choose to Abstain,” at cbn.com. The first paragraph:

A recent study shows that most teenagers are virgins, contradicting claims from family planning groups that most young people do not abstain from sex and more sex ed should be taught in schools.

Image by Kelley Boone, some rights reserved (CC-BY-SA 2.0)

This kind of blabbery drives me nuts. They might has well have said, “A recent study shows that the earth is flat, contradicting claims from Unitarians that the planets revolve around the sun and astronomy should be taught in schools instead of the Bible,” when in fact a recent study showed no such thing, and even if it had, it wouldn’t contradict what the Unitarians supposedly said. Maybe if I’d been on the debate team I’d know how to respond more effectively.

If I were a fundamentalist Christian who wanted to justify abstinence education, I wouldn’t quote or misquote studies, nor would I attempt to use logic. I’d be honest: “According to my church, the world is flat, most young people abstain from sex, and abstinence should be taught in schools right after study hall and before creationism. That’s what I believe, because faith in the church is my guiding light.”

Studies be damned, science be damned, the church is the ultimate authority. I might have more respect if they put it that way more often. (I would still object if it got to the point of the Constitution be damned and laws be damned.) Why should fundamentalists care a whit about the fact that science is consistent, well-founded, and predictive? Why should they care about evidence from studies and measurements, if faith, not intelligence, is their life’s compass? I can disagree, disapprove, and be dismayed, but I have no appeal. We live on different planets; we grew up in different universes.

Anyway, for readers who might appreciate facts and figures, let me explain the CBN’s vulpigeration.

What is “sex,” anyway? For its study, the CDC defined “sex” to be heterosexual vaginal sexual intercourse¹ only (though the boy need not stay on top). Many English speakers would call a bunch of other things people do naked with others sex, but the CDC’s restrictive definition should suit the Christian Broadcasting Network in two ways. First, this definition doesn’t infringe on the way CBN might define another word, “sodomy.” They might prefer it for that bunch of other things people do naked with each other. Second, it yields higher virgin percentages. As far as the CDC and CBN.com are concerned, you’re a virgin if you haven’t been part of any penis-in-vagina hanky-panky, even if you’ve gotten plenty naked and nasty with one or more hims or hers.

Fact: Most young people do not abstain from sex. (Or “sex.”) Not during their entire youth, which is what CBN.com suggested. According to the CDC study, most (65% of) boys aged 18-19 and most (60% of) girls in the same age group have had heterosexual vaginal sexual intercourse. The CDC numbers suggest that most young people do abstain from sex “sex” until about age 17 or 18, but abstaining until you stop abstaining is not the same thing as abstaining. Using the CBN.com logic, you could say that all people abstain from sex, ’cuz they all do — until they stop, and most stop, as we know from all the babies being born and abortions being performed. Few babies (or aborted fetuses) are incarnate nowadays.

¹ Additional information available on the internet.

Published research studies usually drive me nuts, but this one less than most. Social Psychological and Personality Science just published “Real Men Don’t Eat Quiche: Regulation of Gender-Expressive Choices by Men,” by David Gal and James Wilkie, and it’s fabulous awesome.

According to the abstract,

Our findings suggest that men experience a conflict between their relatively intrinsic preferences and gender norms and that they tend to forgo their intrinsic preferences to conform to a masculine gender identity (when they have sufficient resources to incorporate gender norm information in their choices). Women, on the other hand, appear to be less concerned with making gender-congruent choices.

The authors found that men, when asked to choose between two foods, one with a straight macho masculine description corresponding to American societal “norms” for guy food and the other that was gay sissy feminine more what Americans might think of as girly food, they picked the guy item almost two-thirds of the time. If, that is, they had plenty of time to ponder their choice. If they were rushed to decide, though, they picked girl food choices more often — about 55% of the time, on average. Women, on the other hand, choose girl food about two-thirds of the time, regardless of whether they’re rushed to decide.

The authors conclude (in more precise language than my paraphrase) that men, unlike women, are cognitively self-regulating their decisions according to societal norms of gender expression. In other words, while men like girly food as much as girls, they’ll decide not to order it (forgoing food they like in order to look like “real men”) if they have time to think things through.

It doesn’t surprise me that men put energy into “behaving like men,” even when it sometimes conflicts with their intrinsic desire. (I’m not so convinced that men are as unlike women as the authors say.¹)

The authors describe the effect they saw as “making gender-congruent choices,” but I might envision it another way: men put a lot of energy into avoiding anything they think will make them look gay. How different is what the authors call “threats associated with gender-norm transgression” from fear of being labeled a fag?

In any case, special thanks to the authors for their menu of “feminine” and “masculine” menu items, which was half the fun of the paper. Here are a few selections. I want all of them, but hold the shredded American cheese.

• Martha’s Vineyard Salad Mixed baby greens and fresh spinach with toasted pine nuts, dried cranberries, cucumber, red onion, and a warm Vermont goat cheese crouton with a balsamic vinaigrette
• Chunky Fudge Cake Ice Cream Vanilla ice cream, smothered in hot fudge with chunks of chocolate fudge cake, whipped cream, and peanuts
• Vitello Carciofi and Asparagus Beef medallions sautéed with asparagus and artichoke in a light demi-glace sauce
• Damon’s Specialty Pizza Ground hamburger, red onions, roasted peppers, and mozzarella cheese
• Western Salad Chunks of barbequed chicken with shredded American cheese served on greens with a side of Ranch dressing

¹ The authors’ findings suggest that men do this, but women don’t (or do to a much smaller extent). But the authors only studied university undergraduates at (I assume from their affiliation) a largish private Midwestern university. For that population, it’s fairly reasonable to generalize, and perhaps for that population this in fact is a guy-only thing. I’d speculate, but with no support from the study, that the effect is present among men across most segments of the U.S. population. But it wouldn’t surprise me to find the “I better order something gender-appropriate” effect in women, too, in some places (richer white populations in the South?). Studies in populations other than undergraduates would be nice to see.

The danger in generalizing from undergraduates, who are readily available to university researchers, to the general population, has fortunately been getting some press lately. It’s a real danger.

There are plenty of other interesting angles to explore. To what extent this effect is expressed ought to depend on the environment. Do men (with time to think) pick guy foods more frequently when they’re dining with several guys as opposed to when dining with a single woman? (I’d put money on yes.) Are there differences between straight men and gay men? (I’m not sure I’d bet on this one.)

Thanks to Edward Tufte, millions of people have seen Charles Minard’s remarkable chart of the French Army’s losses during its Russian campaign in the winter of 1812-13. Minard’s chart is a joy to behold. It’s the acme of data presentation — magnificent, spectacular, inspiring. So it kills me that Gene Zelazny, who wrote the you-know-from-the-title-it’s-bad book “Say it with Charts” FUCKING SHAT ALL OVER MINARD’S LEGACY.

I learned about Zelazny’s desecration here, though Andrew Abela, who reported it, failed to call it that. “Zelazny notes that the graphic is difficult to read, and proposes that there might be better ways to convey the same information.”

Sure, there might be, just like there might be better ways to decorate the ceiling of the Sistine Chapel, but what Zelazny offers is an epic numerical fail, because IT NOT ONLY DEPICTS DIFFERENT (AS IN WRONG) INFORMATION, IT ALSO DEPICTS THE WRONG INFORMATION BADLY (as in we can’t even tell what wrong numbers he wants us to find and what they are supposed to quantify).

Zelazny might as well propose that “there might be better ways to clean up the Gulf of Mexico,” trot out a broken doorknob and a bent bicycle tire, and ask us to wonder with him. Ugh, ugh, ugh. And ugh.

Here are a few of the gory details (of which there can be but few, given how little actual stuff there is in Zelazny’s chart).

Ok, so the spirit levels in the cutesy clipart thermometers don’t match the numbers, only their absolute values (sort of).  But the numbers are wrong, too. Five of Zelazny’s six data points are wrong — misread from Minard’s original. Five out of six. That’s almost all wrong, for those of you who aren’t counting. Badly, differently, and horrifyingly wrong.

Minard reports that there was rain on October 24, and that the temperature was about zero*. Zelazny misread the day of the month (24) as a temperature, then used the only other written figure at that spot on Minard’s chart (8^bre, for octobre) for both the month and the day. No explanation short of “Who gives a fuck?” works for this slop.

*Minard’s figures give the Réaumur scale temperature, which detail Minard, lest future readers misconstrue his chart, tells us. Minard cared deeply about communicating. (Zelazny’s figures are wrong in every known temperature scale. He cares less. Much less, like not at all.)

Remarkably or not, almost nothing is correct in Zelazny’s “presentation.” The border between Poland and Russia is misplaced, and all the graphical scales are wrong. I’m no PowerPoint guru, but I assume you have to work very hard to incorporate numbers into a slide this wrongly (as was famously done here, and better).

Even Zelazny’s title is wrong.

Things got bad on the retreat from Moscow. And it’s not clear how many died. Minard charted the number of troops, not deaths. Some who didn’t return were captured. Others may have deserted. And the overall message isn’t “the colder … the more.” The biggest declines were early in the campaign, when the temperatures aren’t given. So the title is all wrong, but hell, IT FIT ON TWO LINES. Shit like this matters. It’s our planet’s fucking history.

Sure, Minard’s correct title (Carte figurative des pertes successives en hommes de l’ Armée Française…, and penned more beautifully than any web typography can be rendered) won’t work projected at WXGA resolution or on your favorite eReader or phone.

THIS IS WHY WE PRINT STUFF ON PAPER. If you don’t have a copy of Minard’s chart, buy yourself one. Fuck, if you’re one of the first five people to ask me, I’ll buy you one.

Now turn off your computer and pick up a beautiful book. Or go to the library. Or write. On paper. Thanks for listening.

Web sites about mathematics should help people understand and appreciate mathematics, not confuse the crap out of them with misinformation. Unfortunately, Wolfram Mathworld does the latter.

Example 1. MathWorld explains here that “The numbers of palindromic numbers less than a given number are illustrated in the plot [below].”

So the left plot tells us that there are about 100 palindromes less than or equal to 20. But there are only 21 nonnegative integers less than or equal to 20, so there can’t be 100 palindromes among them. In fact, there are 11 palindromes less than or equal to 20: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, and 11. My guess is that the left plot illustrates the n-th palindromic number as a function of n. In any case, it’s not what MathWorld describes.

MathWorld begins its list of the “first few palindromic numbers” with 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 (these 10 numbers are palindromes and are all less than 10), but in the next paragraph, MathWorld states that the number of palindromic numbers less than 10 is 9. There are 9 if you don’t count zero for some strange reason, but if you don’t intend to, give a definition that excludes it (MathWorld’s definition is less than clear), and then don’t list it.

Still confused? Read the Wikipedia article.

Example 2. Pascal’s Triangle shouldn’t be hard to screw up, right? Wrong. Here’s MathWorld’s Pascal’s Triangle:

This triangle needs to go to the shop for an alignment. The numbers are neither lined up in columns nor staggered (the latter being the usual presentation). What are the numbers in the column containing the rightmost 4? What numbers are along the diagonal through the top? (1, 1, 1, 1, 1, 5, 6?) As shown, MathWorld’s anyway-ill-worded “each subsequent row is obtained by adding the two entries diagonally above” is meaningless.

Example 3. In its article on Mersenne numbers (numbers that are one less than a power of two), MathWorld attempts to explain why “[i]n order for the Mersenne number [2ⁿ-1] to be prime, n must be prime.” MathWorld’s justification: “This is true since for composite n with factors r and s, n = rs. Therefore, 2ⁿ–1 can be written as 2^rs–1, which is a binomial number and can be factored.” That’s sloppy to say the least. First, if a composite number n has factors r and s, it’s not necessarily the case that n = rs. Furthermore, the fact that a number can be factored doesn’t prove it’s composite. Every Mersenne number 2ⁿ–1 can be factored. It’s just that when n is composite, there’s definitely a factorization into positive integers neither of which equals 1. Explaining it isn’t hard: In order for 2ⁿ-1 to be prime, n must be prime. For if not, n = rs where r and s are integers greater than 1 and less than n; then 2ⁿ–1 = 2^rs–1 has a factor between 1 and 2ⁿ–1, namely 2^r–1.

Example 4. MathWorld describes prime numbers as “numbers that cannot be factored.” Prime numbers, like all integers, however, can be factored, and elsewhere, MathWorld gives the factorization of several prime numbers, such as 7: 7 = 7×1.

Example 5. Any of MathWorld’s articles on statistics.

In the article on the Central Limit Theorem, what is lowercase n? What is f? The “limiting cumulative distribution function” of X_norm is limiting in the sense of what approaching what? (It’s not clear to me that MathWorld’s statement of the theorem is even correct, but it’s clearly unclear.)

The article “explaining” the p-value has perhaps the worst definition of p-value I’ve ever seen when not grading exams. MathWorld says it’s “[t]he probability that a variate would assume a value greater than or equal to the observed value strictly by chance: P(z > z_observed)” (wrong). Wikipedia says “In statistical hypothesis testing, the p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true” (right).

Renewed concern remained high for the third consecutive month, setting a record in March, according to first-quarter figures released today. The data came as a surprise to analysts, many of whom expected 2009’s historically low levels to continue for at least 12-18 months.

“We’re surprised, but we’re still forecasting a positive outlook for the year,” said Trend Analytics’ Sandeep Singh. “Don’t forget that increasing optimism, though off its January peak, remains strong.” In his widely-read newsletter, “Pulses,” Singh calls for renewed concern to drop by 10-15% in Q2 and for increasing optimism to remain steady.