Quotulatiousness

January 18, 2012

First they came for the smokers, then the drinkers, and now the meat-eaters

Filed under: Britain, Government, Health, Liberty — Tags: , , , — Nicholas @ 10:52

Rob Lyons on the flimsy case for declaring that “eating meat causes cancer”, and the rising tide of buttinsky government and their nudge, hector, prod, and persecute urges:

Meat causes cancer. It’s been said so many times that you’d have to be an idiot not to believe it, right?

The latest confirmation of this apparent common sense was a report published last week in the British Journal of Cancer Research. The authors, from the Karolinska Institute in Sweden, brought together 11 studies — published between 1993 and 2011 — that assessed the risk of pancreatic cancer from eating red meat and ‘processed’ meat. From this meta-analysis, the authors found that red meat increased the risk of pancreatic cancer for men, but not for women, and that the risk of pancreatic cancer rose by 19 per cent for every 50 grams of processed meat consumed.

The simple claim that ‘processed meat causes cancer’ was widely reported after the study was published. However, it would be wrong to assume that such claims about risk are all they are cracked up to be.

[. . .]

There are so many ways in which the crude tools of epidemiology could screw up the result of studies like this that it is normal for fairly small risks — like the 19 per cent increase in this case — to be treated with a massive pinch of salt. The authors of this study even note: ‘All studies controlled for age and smoking, but only a few studies adjusted for other potential confounders such as body mass index and history of diabetes.’

[. . .]

So, to sum up: the association between processed meat and pancreatic cancer is so weak it might well be a mirage; the increased risk might not be caused by the processed meat itself; and even if it is, the risk is so low that it’s really not worth bothering about. Yet still we are advised to consider cutting down on our red meat and processed meat consumption. Life is, frankly, too short to miss out on such tasty foods on the slim chance that we might lose a few years of life in old age.

[. . .]

Now that the precedent has been set for the government to lambast those who engage in unapproved habits, it’s open season on any habit that a campaigner or columnist disapproves of. Ban it! Tax it! Make them get a prescription for it! Deny them medical care! Ellen’s article is objectionable but it only follows the remorseless logic of so many others.

There is another lesson from the meat-and-cancer story: at a time when all sorts of dubious claims are made based on junk science and dodgy statistics, only some panics get wide publicity; others just pop up and disappear again in a matter of hours. The difference is that some play to an existing political or media agenda and some do not. The idea that meat causes cancer appeals to health busybodies, politicians scrabbling around for a sense of purpose, vegetarians who can’t win a moral argument about animal rights, and environmentalists who have failed to convince us that increasing the ‘human footprint’ — by wanting to eat more meat, for example — is killing the planet.

January 15, 2012

As you’d expect, healthcare costs are not evenly distributed

Filed under: Economics, Health, USA — Tags: , , , — Nicholas @ 12:20

Jordan Weissmann in The Atlantic:

When it comes to America’s spiraling health care costs, the country’s problems begin with the 5%. In 2008 and 2009, 5% of Americans were responsible for nearly half of the country’s medical spending.

Of course, health care has its own 1% crisis. In 2009, the top 1% of patients accounted for 21.8% of expenditures.

The figures are from a new study by the Department of Health and Human Services, which examined how different U.S. demographics contributed to medical costs. It looked at the $1.26 trillion spent by civilian, non-institutionalized Americans each year on health care.

The top 5% of spenders paid an annual average of $35,829 in doctors’ bills. By comparison, the bottom half paid an average $232 and made up about 3% of total costs.

January 5, 2012

The MPAA over-cooks their numbers to support SOPA

Filed under: Economics, Law, Media, Technology, USA — Tags: , , , , , — Nicholas @ 09:53

Techdirt reports on the work done by Julian Sanchez at the Cato Institute to actually scrutinize the “loss” numbers used by the MPAA:

One of the things we’ve noticed in the debate over SOPA and PIPA is just how the other side is really lying with statistics. We’ve done a thorough debunking of the stats used by the US Chamber of Commerce to support both bills, as well as highlighted the misleading-to-bogus stats used by Lamar Smith in his support of the bill.

But every day, more bogus stats are rolled out. Julian Sanchez, over at the Cato Institute, has decided to dig into one specific bogus number, the supposed claim of $58 billion in “losses,” and to show how the numbers don’t hold up to any scrutiny. In fact, using the details of where the numbers came from, Sanchez makes the case that SOPA won’t save a single net job for the US economy. Read on to find out how.

First off, the $58 billion comes from an absolutely laughable report for the Institute for Policy Innovation, done every year by Stephen Siwek at a firm called Economists Incorporated. We’ve challenged this ridiculous number in the past, but not to the level of detail that Sanchez has here. He starts out by bringing up (as we have many times), Tim Lee’s excellent debunking of the ridiculous “ripple effects” that Siwek/IPI always use, despite them being a trick to double, triple, quadruple, etc count the same dollars [. . .]

January 3, 2012

Blog statistics for non-statisticians

Filed under: Administrivia, Technology — Tags: , — Nicholas @ 10:01

I’m not a big stats nerd — being mathematically challenged means I’m less willing to devote time to things that require extra math. However, most if not all bloggers do care about a few statistical measurements: how many people are visiting their blogs. I’m no exception to that rule.

I don’t have a complete series of annual numbers, as the tools under MovableType (the old site) and WordPress (the current site) don’t provide quite the same slices of data. I installed SiteMeter on the old site a couple of months after I started blogging and it shows 414,416 unique visits from 17 August, 2004 to today (and it still gets around 100 visits per day, even though I haven’t been posting there for more than two years).

Since I switched to the current site the traffic has been going up, although the big blogs don’t have to worry that I’m drawing too much of their readership:

  • 2009: 58,121 unique visits, 131,825 hits (site went live in July, stats date from mid-August)
  • 2010: 328,374 unique visits, 825,381 hits
  • 2011: 413,463 unique visits, 1,118,497 hits

That concludes our occasional dip into the statistics. Thanks for coming by, and especially thanks to folks that link to my blog.

Update: I happened across this bit from July, 2009 on the old blog that still seems relevant:

I’m not sure why I’ve been blogging for five years … it’s certainly not the money, booze, and groupies! I’ve thought about stepping away from the keyboard every now and again, but I don’t actually write as much as I once did, so large chunks of my “blogging” time are actually copy-paste-and-code sessions, rather than writing.

The blogroll has certainly diminished in importance over the last couple of years. The Red Ensign bloggers, my primary affiliation, has diminished to about a dozen active blogs, of whom perhaps 5-6 produce the vast majority of posts. Other blogrolls I’m on have similar profiles of activity. Blogrolls don’t matter compared to when I first started blogging back in 2004.

I remember worrying about SiteMeter and the Ecosystem, as they showed me what my visitors were reading, where they came from and where they went. Time has also not been kind to the ease of gathering that sort of information, as more readers come in from search engine results, RSS feeds, and goodness knows what other channels. If/when I move the blog over to the new site, I may not bother including the links for those tools. They’re no longer all that useful or informative.

I do miss the cameraderie of the early blogging years … but as more of the early blogs go dark, the replacements are less likely to be bloggers and more likely to be Twitterers, Facebookers, YouTubers, Farkers, Slashdotters, and all the other Web 2.0/New Media options that are now available. What was that old expression about the only constant being change?

December 21, 2011

Barbara Kay: Spousal abuse is remarkably gender-balanced

Filed under: Cancon, Law, Media — Tags: , , , , — Nicholas @ 10:18

Everyone knows the old myth about a spike in wife-beating after major sporting events (most frequently referenced is the Superbowl, but the same factoid is trotted out about every “big game”). Barbara Kay reveals the awkward truth that nearly half of all spousal abuse is by female partners:

One of first-wave feminism’s great achievements in the 1970s was to end the denial surrounding wife abuse in even the “best” homes. Resources for abused women proliferated. Traditional social, judicial and political attitudes toward violence against women were cleansed and reconstructed along feminist-designed lines.

But then a funny thing happened. The closet from which abuse victims were emerging had, everyone assumed, been filled with women. But honest researchers were surprised by the results of their own objective inquiries. They were all finding, independently, that intimate partner violence (IPV) is mostly bidirectional.

But by then the IPV domain was awash in heavily politicized stakeholders. Even peer-reviewed community-based studies providing politically incorrect conclusions were cut off at the pass, their researchers’ names passed over for task force appointments and the writing of training manuals for the judiciary. Neither were internal whistle-blowers suffered gladly. Erin Pizzey, who opened the first refuge for battered women in England in 1971, was “disappeared” from the feminist movement when she revealed what she learned in her own shelter: She committed a heresy by asking women about their own violence, and they told her.

[. . .]

(While the CDC survey does not reference Canadian data, our IPV statistics vary significantly from the U.S.’s in certain respects. “Minor” wife assault rates as measured on the commonly employed Conflict Tactics Scale are identical, but “severe violence” rates in Canada fall as the violence ratchets up. For “kicking” and “hitting,” Canadian rates were 80% of the American rate; for “beat up,” they were 25%; and for “threatened with or used a gun/knife,” they were only 17%.)

By now there is no excuse for the failure of governments at all levels to follow through on — or at least acknowledge — the settled science of bilateral violence. Yet just last week the Justice Institute of British Columbia issued a lengthy report on “Domestic Violence Prevention and Reduction,” and sure enough, it defines domestic violence as “intimate partner violence against women,” recommending only that government work “to bridge gaps in the services and systems designed to protect women and children.”

One area where the majority of abusers are female is child abuse: women are much more likely to batter their children than men.

December 15, 2011

Grim, crime-wracked, post-apocalyptic Toronto ranks … 52nd most dangerous in Canada

Filed under: Cancon, Law, Liberty — Tags: , , , , — Nicholas @ 09:29

Everyone in Canada knows that Toronto is a cess-pit of crime where the oppressed citizenry huddle in fear, while idyllic Victoria is a benign, peaceful enclave of happiness. But what we know just ain’t so:

Toronto ranks 52nd among cities and towns in the country for the label “most dangerous” according to Maclean’s. Victoria, BC? Far from being a peaceful place, ranks second in the country after Prince George, BC. In fact, BC has four of the top ten dangerous cities, while Ontario’s most dangerous place, Belleville, clocks in at number 11.

December 8, 2011

Health advocates argue in advance of the data in new cancer study

Filed under: Britain, Health, Media — Tags: , , , , — Nicholas @ 08:36

Rob Lyons on the latest cancer study, which makes unsubstantiated claims about the “lifestyle” causes of cancer:

The essential idea we are being sold here is that medical experts know that certain behaviours — like smoking, drinking alcohol, eating read meat and not eating enough fruit and vegetables — increase your risk of developing cancer by a certain percentage. So, all we need to do is work out how many people would have got cancer if no one did any of those things, take that number away from the number of people who do get cancer, and the remainder is how many people that ‘unhealthy living’ is killing. Simple, right?

According to the report, If you do all the ‘right’ things — if you are a cigarette-dodging, skinny teetotaller who avoids all red meat, barely goes out in the sun (except, perhaps, to take the prescribed 30-minute sessions of exercise five times per week), gets lashings of fibre, cuts down on salt, avoids infectious diseases and ionising radiation, and so on — then you can cut your cancer risk by over 40 per cent. On that basis, you may avoid cancer but die of boredom instead.

More specifically, even in this report there’s a huge gulf between the widely acknowledged risk of smoking — which is estimated here to cause 19.4 per cent of all cancers — and other risk factors. Smoking accounts for nearly half the lifestyle risk of 43 per cent claimed in the report. The next biggest factors suggested are overweight and obesity (5.5 per cent), lack of fruit and veg (4.7 per cent), alcohol (4.0 per cent), occupation (3.7 per cent) and sunlight (3.5 per cent). No other single factor, according to the report, is responsible for more than three per cent of cancers. Some oft-quoted examples like salt (0.5 per cent) and physical exercise (one per cent) have little effect at all. Even avoiding red meat altogether would only avoid 2.5 per cent of cancers, says the report.

November 16, 2011

The gender wage gap won’t go away

Filed under: Economics, Government, Politics — Tags: , , , — Nicholas @ 09:28

Kay Hymowitz explains that even with the best will in the world, the wage gap — often referred to as the 75-cents-on-the-dollar phenomenon — between men and women will persist:

Let’s begin by unpacking that 75-cent statistic, which actually varies from 75 to about 81, depending on the year and the study. The figure is based on the average earnings of full-time, year-round workers, usually defined as those who work 35 hours a week or more.

But consider the mischief contained in that “or more.” It makes the full-time category embrace everyone from a clerk who arrives at her desk at 9 a.m. and leaves promptly at 4 p.m. to a trial lawyer who eats dinner four nights a week — and lunch on weekends — at his desk.

I assume, in this case, that the clerk is a woman and the lawyer a man for the simple reason that — and here is an average that proofers rarely mention — full-time men work more hours than full-time women do. In 2007, according to the Bureau of Labor Statistics, 27 percent of male full-time workers had workweeks of 41 or more hours, compared with 15 percent of female full-time workers; just 4 percent of full-time men worked 35 to 39 hours a week, while 12 percent of women did. Since full-time men work more than full-time women do, it shouldn’t be surprising that the men, on average, earn more.

The other arena of mischief contained in the 75-cent statistic lies in the seemingly harmless term “occupation.” Everyone knows that a CEO makes more than a secretary and that a computer scientist makes more than a nurse. Most people wouldn’t be shocked to hear that secretaries and nurses are likely to be women, while CEOs and computer scientists are likely to be men. That explains much of the wage gap.

But proofers often make the claim that women earn less than men doing the exact same job. They can’t possibly know that. The Labor Department’s occupational categories can be so large that a woman could drive a truck through them. Among “physicians and surgeons,” for example, women make only 64.2 percent of what men make. Outrageous, right? Not if you consider that there are dozens of specialties in medicine: some, like cardiac surgery, require years of extra training, grueling hours, and life-and-death procedures; others, like pediatrics, are less demanding and consequently less highly rewarded. Only 16 percent of surgeons, but a full 50 percent of pediatricians, are women.

Don’t expect China to save your economy

Filed under: China, Economics, Government, Media — Tags: , , , , , — Nicholas @ 09:19

Jon, my former virtual landlord, sent along this link which should pour cold water on the notion that China will step in to save the economies of other countries:

China’s economy has a reputation for being strong and prosperous, but according to a well-known Chinese television personality the country’s Gross Domestic Product is going in reverse.

Larry Lang, chair professor of Finance at the Chinese University of Hong Kong, said in a lecture that he didn’t think was being recorded that the Chinese regime is in a serious economic crisis — on the brink of bankruptcy. In his memorable formulation: every province in China is Greece.

The restrictions Lang placed on the Oct. 22 speech in Shenyang City, in northern China’s Liaoning Province, included no audio or video recording, and no media. He can be heard saying that people should not post his speech online, or “everyone will look bad,” in the audio that is now on Youtube.

In the unusual, closed-door lecture, Lang gave a frank analysis of the Chinese economy and the censorship that is placed on intellectuals and public figures. “What I’m about to say is all true. But under this system, we are not allowed to speak the truth,” he said.

Despite Lang’s polished appearance on his high-profile TV shows, he said: “Don’t think that we are living in a peaceful time now. Actually the media cannot report anything at all. Those of us who do TV shows are so miserable and frustrated, because we cannot do any programs. As long as something is related to the government, we cannot report about it.”

China, for all its amazing growth and rising economic prospects for (some of) its population, is still not a modern economy. The government — specifically the military — is too deeply involved at far deeper levels than other governments and the reported economic figures may or may not have any relationship with reality. When your boss is a general, he has ways of ensuring that you report the “right” results that a civilian CEO cannot match. It’s not just your job you risk by reporting unwelcome results.

I’ve ridden this hobby horse, as Jon calls it, many times over the years.

October 31, 2011

QotD: Economics is not a “hard science”

Filed under: Economics, Quotations, Science — Tags: , , , — Nicholas @ 13:20

The problem at base is that economics is not a branch of mathematics or statistics, no matter how much economists wish it was. Never forget that the economics equations you see, the pretty graphs and charts, are just educated guesses that are wrong more often than not — economists love the gloss of the hard sciences, but the truth is that the field is firmly placed among the philosophical and sociological disciplines. Economics is a study of human behavior more than anything else, with all the uncertainties and confusion that entails.

“Monty”, “DOOM: I like that Doom Doom Pow”, Ace of Spades H.Q., 2011-10-31

October 29, 2011

More on Malthus

Filed under: Books, Economics, History, Media — Tags: , , , — Nicholas @ 10:57

Tim Harford poses and answers some common questions about the much-talked-about 7-billionth birth:

There will be 7bn of us come Halloween? You’re kidding.

Someone at the UNPFA has a sense of humour, it seems. This is a terrifying prospect to some and cheery for others, while most of us find it irrelevant. Halloween seems appropriate.

This is a statistical projection, and statistical projections don’t have a sense of humour.

But they do have margins of error. The UNPFA doesn’t even know whether the seven billionth person will be born in 2011. Even the UK is currently relying on 10-year old census data. There are places in the world where the data are decades old and we’re just guessing at the population. Still, such celebrations — if they are celebrations — do pass the time.

[. . .]

So Malthus was wrong?

Even Robert Malthus realised that there was such a thing as birth control. His more excitable successors, such as Paul Ehrlich, author in 1968 of The Population Bomb, have been fairly thoroughly rebutted by subsequent events. Dan Gardner, author of the fascinating Future Babble, summarises Ehrlich’s three scenarios for the 1980s as “everybody dies” (starvation-triggered nuclear war), “every third human dies” (starvation-triggered virus) and in the optimistic scenario “a billion people still die of starvation”.

October 26, 2011

Mis-perception of relative risks

Filed under: Football, Health, Randomness — Tags: , , , — Nicholas @ 00:03

Gregg Easterbrook provides a good example of how difficult people often find to discern the relative weight of risks:

The first consideration is that both absolute numbers of football deaths and rates of death compared to participants are in long-term decline — mirroring the decline in many forms of risk in society. Age-adjusted rates of all deaths in the United States have declined for 10 consecutive years. Auto fatalities have been declining for more than a generation. Winning the War on War, an important new book by Joshua Goldstein [. . .] shows that despite the impression created by cable news, exposure to violence is in decline both in the United States and worldwide.

[. . .]

Data from the National Center for Catastrophic Sport Injury Research reflects a steady decline in deaths caused by football. Table 1 of the center’s most recent report shows that in the past decade, 34 high school, three pro and two college football players have died as the direct result of games or practices, with the primary cause of deaths being heat stroke. That is entirely awful — but much lower than the rate of a generation ago. In 1968 alone, 26 high school players died as a direct result of football; last year, the number was two. Table 3 of the report shows the direct fatality rate from high school football peaked at 2.6 deaths per 100,000 players in 1969 and declined steadily to 0.13 deaths per 100,000 in 2010. That means a 1968 high school football player was 20 times more likely to die than a 2010 player. (The main reason for declining deaths was that football helmets were improved to eliminate skull fractures.)

[. . .]

How to compare the slight risk of a terrible football outcome to other common risks experienced by the young? Consider the risk of being in a car. About 3,000 teens die each year in car crashes. There are about 21.3 million Americans between 15 and 19 years of age. Teens average about 146 miles driven per week, roughly 150 hours per year of driving. These figures yield a roughly one in 1 million chance that a teen will die in an hour of driving. The National Federation of State High School Associations reports that 1.1 million boys (and a few girls) played high school football last academic year. A typical high school football season would include, in games and practice, perhaps 75 hours of exposure to contact. That’s about 80 million total hours of exposure to contact on the part of high school football players. The National Center for Catastrophic Sport Injury Research reports a recent average of three deaths per year directly caused by high school football. That’s a roughly one in 27 million chance of a high school player dying from an hour of football contact.

These are all rough estimates. Taking them together, a teenager has a one in 1 million chance of dying in an hour behind the wheel, compared to a one in 27 million chance of dying in an hour of football contact. Being in pads on a football field is less deadly than driving to high school for class. Many contemporary parents, especially moms, might say, “I don’t want you playing football because it’s so dangerous, but it’s fine for you to drive to the mall.” As regards mortality, this misperceives the risks.

October 5, 2011

The irony of Bletchley Park’s funding windfall

Filed under: Britain, History, Technology, WW2 — Tags: , , — Nicholas @ 10:47

Cory Doctorow has the good news about Bletchley Park’s recent grant:

Bletchley Park, the birthplace of modern crypto and the home of the WWII codebreaking effort, has received a £4.6m Heritage Lottery Fund grant to fund restoration efforts and new exhibits. Bletchley was broken up after the war and its work was literally buried as part of the Cold War climate of secrecy that prevailed. In the years that followed, neglect and time led to the near-destruction of many of the historic sites. The Bletchley Park trust has since done amazing work on a shoestring budget to restore and preserve Bletchley, creating a fabulous museum and rebuilding some of the most beautiful electromechanical computers I’ve ever seen.

[. . .]

Ironically, the money to restore Bletchley has come from the lottery, a government-run system designed to reinforce and exploit statistical innumeracy of the sort that Bletchley’s cryptographers overcame in order to help win the war.

August 8, 2011

ESPN’s new attempt at a more accurate Quarterback rating system

Filed under: Football — Tags: , , — Nicholas @ 17:25

The existing quarterback ranking is hard to understand and concentrates on the “passing” side of the quarterback’s job. ESPN is introducing a more broad-based ranking system:

The Total Quarterback Rating is a statistical measure that incorporates the contexts and details of throws and what they mean for wins. It’s built from the team level down to the quarterback, where we understand first what each play means to the team, then give credit to the quarterback for what happened on a play based on what he contributed.

[. . .]

Total QBR Basics

A quick primer on the fundamentals of Total Quarterback Rating:

Scoring: 0-100, from low to high. An average QB would be at 50.
Win Probability: All QB plays are scored based on how much they contribute to a win. By determining expected point totals for almost any situation, Total QBR is able to apply points to a quarterback based on every type of play he would be involved in.
Dividing Credit: Total QBR factors in such things as overthrows, underthrows, yards after the catch and more to accurately determine how much a QB contributes to each play.
Clutch Index: How critical a certain play is based on when it happens in a game is factored into the score.

Under the new ranking, Brett Favre’s performance gets a lot less impressive (if an average QB would score 50 points):

  • 2008 – New York Jets – 41.7
  • 2009 – Minnesota Vikings – 63.1
  • 2010 – Minnesota Vikings – 25.8

June 30, 2011

Does exposure to porn increase the incidence of rape?

Filed under: Law, Media — Tags: , , , — Nicholas @ 14:54

In brief, it appears not:

But while theorising is all very well, it is necessary occasionally to fine-tune such theories by looking at the empirical evidence. And the most obvious fact about porn and rape is that reported rape incidence — at least in the United States, where a National Crime Victimization Survey takes place every year — has been falling in recent decades as porn becomes ever more available.

[. . .]

Now yes, it is absolutely true that correlation and causation are not the same thing. But at first glance we’d have a hard time claiming that the greater availability of porn led to more rapes: simply because there are fewer rapes reported while there’s definitely more porn.

[. . .]

In D’Amato’s paper, he uses Freakonomics-style statistics (one of his colleagues wrote the Freakonomics abortion and crime paper with Levitt) to try to tease out evidence of something more than just correlation.

What he found is that the lower the internet penetration in 2004 in a US state, the higher the rape rate had risen and that the higher the internet penetration, the lower rate had fallen.

We expect, for those societal reasons, that the reported rape rate will have risen over the time period. And where there’s no or limited internet access, it has. Where there is high internet access it has fallen, the fall being greater than the general societal rise.

Thus we have an empirical connection between internet access and lower rape figures. Whether it’s porn or not is a different matter: they could all be playing Second Life instead. An unlikely way to bet though really.

« Newer Posts — Older Posts »

Powered by WordPress