Correlation, Cause, Pubic Confusion and Diet

Newsprint reporters are probably a dying breed and should be treated gently. So, I will simply note that Nanci Hellmich of the USA Today in an article today seems not to understand the basic principle of scientific experimentation.

She leads this article with the following. "Dieters who write down everything they eat each day lose twice as much weight as those who don't, according to one of the largest weight-loss studies ever conducted. This confirms the importance of keeping a food diary." The only real confirmation here is this passage confirms the error in the title "Using food diaries doubles weight loss, study shows.

This is simply wrong. The dieters were not randomly assigned, this was just the first phase of an experiment which in phase two will test, randomly assigned, subjects who already lost weight to a number of maintenance scheme. The design is to see what maintenance schemes work.

The doubling of weight loss, referred to in the article, is simply that the loss for those who by self selection used a food diary is twice that of those that did not. Obvious confounds include the possibility that those who kept diaries were more motivated to lose the weight, or those who were losing weight were proud of their achievements and motivated to keep a diary.

The article is by J. Hollis and others, published Aug 2008 in the American Journal of Preventive Medicine, Vol 35 (2) pg 118-126.

Clearly stated in the article itself is: "A limitation of Phase I of WLM was that it was an uncontrolled observational study." Exactly.

Cause and correlation are difficult to separate. That is the entire reason for random assignment controlled experimentation. Well controlled experiments are still sometimes wrong, correlations picked out by observation are much much worse. Confusion between the two is disastrous in the attempt to explain scientific work to the public.

Lack of Feedback Means Supreme Court is Literally Out of Control

The first concept taught in system control is feedback. It is via a feedback arrangement that a system can be brought under control. The systems that are most associated with elegant responsiveness are all under feedback control (e.g., free markets; electronic circuits; "Darwinian" control in evolution)

With this in mind, it is interesting to think about our three branches of government. The legislature and executive face significant feedback in the form of elections. A lot of people are not happy with the electorates judgment but it does judge and sometimes harshly. The judiciary, except the rare local or state judge facing election, receive little controlling feedback. The Supreme Court, which cannot be reversed, faces none.

The decision prohibiting the death penalty for child rape, unintentionally ironically called "Kennedy v Louisiana", has been roundly blasted by some legal authors, (e.g., Scott Johnson, Andy McCarthy, Jim Lindgren, Orin Kerr).

To this layman, the Justice's argument that there was a consensus in society against the death penalty for child rape seems risible. Louisiana clearly, and the nation probably, sport a consensus in exactly the opposite direction. The court here, and often, seems simply to rationalize its own values and hold the intent of the constitution (underlying the actual words) reflects values that (imagine the coincidence) match their own. In other words, a court which does a lot of what it simply wants. A court out of control.

An Anti-Poverty Program That's Working: Millionaires Rise

An annual report from the consulting company, Capgemini, about the wealthy in the world was just released. Interesting stuff. They make estimates of the number of millionaires (excluding housing) and ultramillionaires. (all numbers here are real dollars, i.e., corrected for inflation)

For me, the fascination in these numbers is their pattern of growth. This is hard to measure, but two estimates are available from this company and they agree. One estimate is to simply average the recent growth of the number of millionaires (having the advantage of using a fixed methods and definitions). Over the last couple of years that average is a 7%. The other way is to look at the growth of the total wealth of the wealthy. Here, future growth is easier to estimate since this is among the consultant's specialties and they have access to (private) data. They estimate that wealth will grow about 8% per year over the near future, which also produces a estimate of about 7% for the increase in millionaires (based on past history which suggests millionaire number in the US grows slightly more slowly that total wealth of the wealthy).

There is little agreement on the number of U.S. millionaires, partly because the data are not good and partly because it depending upon definitions. Reasonable estimates range from 1% to 4% of the population. Nonetheless, what is clear is that a 7% growth in the number of millionaires (which causes a doubling in 10 years) applied several times is a tremendous program to eliminate material hardship. Could the median U.S. household have a worth of a million in 2050?

Across the world, the economic growth of China and India has been, by far, THE success story in alleviating world poverty. In the U.S., economic growth will have massive, human benefits too. Millionaire status does not provide opulence, but it's not a bad start or average.

Mainstream Media Death .... Ruled Suicide

The biggest story on the front page (paper version, above the fold and all that) in the USA Today this morning was "Utilities cut off more homes." Examples of sloppy thinking are fairly common, but this was still remarkable. It hits on all cylinders. In the first three sentences, it manages to pull three major logical errors, misdirection, data selection and a biased source.

The silliness begins with the title which defines the thrust of the article. Any actual interest in such an issue would, one hopes, want to find an underlying cause, essentially an independent variable. The utilities are turning off power because the bills are unpaid, same as always. The utility companies are essentially irrelevant in this. The writer does mention 'skyrocketing' food and fuel costs but makes no attempt to sort out income, outflows, and utility costs in people's budgets.

To 'prove' the case, the reporter (Paul Davidson) brings out the other two corkers. First he writes "Electricity and natural gas shutoffs are up at least 15% in several states. Totals for some utilities have more than doubled." ummmm ..... data selection is your friend, eh? Why not just give the U.S. avg? The intelligent reader, far from being convinced, is left wondering why he selected that data .... does the rest fail to support his case?

For his third logical error, the writer pulls a quote from the head of the National Energy Assistance Directors' Association. Is there any earthly reason to suppose this person unbiased? Energy assistance organizations advocate for and help fund those who are, or are in danger of, power cutoffs. This can be thought of as a noble effort, but there is little possibility of an neutral viewpoint.

Now, perhaps the author simply wanted to bash the utilities, seeing greedy plutocrats lighting cigars with $100 bills squeezed from the blood of the poor. Fine. Just write and editorial. Editorials on the front page, unlabeled, should be an embarrassment

.

Casualty Rates, Fishing and Fighting

A recent measure of Iraq War coverage shows the TV Networks are no longer covering the war. The obvious presumption is that the lack of coverage is because casualties are down. Many blogs suspect political motivations (e.g., Glenn Reynolds and John Hinderaker)

While it is obviously impossible to know what future casualties will be, the steep reduction in Iraq makes it an interesting time point to stop and try to assess.

Such comparisons must be done carefully, both because it must not be seen to denigrate even a single soldier's death as trivial and it must hold up to scrutiny, unlike some claims. Nonetheless, comparisons of casualty rates is critical to provide perspective and the total human cost of the war. In the following, the focus is on death rates as they are both most important and not subject to definition and category problems.

There have been about 4,000 U.S. Armed Services people killed in the Iraq War and about 450 killed in Afghanistan. These come from among the slightly greater than one million armed services personnel, so the annual death rate over the 5 years of the War on Terror is about 100/100,000.

Occupational death rates are available from the National Institute of Occupational Safety and Health. They list forestry and fishing as the most deadly (major) occupation with an annual death rate of 112/100,000.

One may argue, of course, that the 'proper' base group for war death rate is the troops in Iraq and Afghanistan rather than the entire armed forces. With this subset of the armed forces, the annual death rate is about 400 or 500/100,000. There is, however, a deadlier subset of fishermen too. The crab fishermen of Alaska (featured in the documentary "Deadliest Catch"). These fishermen are subject to a similar, 400/100,000 annual death rate.

It is the nature of the world that all men die. Some lose their lives fighting for a cause. One can argue that the Iraq and Afghan wars are or were mistakes, but one must keep a perspective on the risk. The mainstream media, in their obsession with casualty rates, never achieved perspective.

Salmonella by the Dashboard Light

Sometimes you just have to whistle at the irony. I was doing a Google search to find how much attention was being paid to safety in the rage to discuss small cars and gas mileage.

I had the idea that there would be more media focus on the salmonella tomatoes (killing perhaps 5) than on the move to small cars (killing thousands). So, I did a search on "$4.00 gas" and "safety." A healthy 50K results came up and I dug in to read the first few. The first listing was useless, having the "safety" entry in a inane flame among the comments. The second produced the whistle. The main article discussed the usual techniques for maximizing gas mileage. The irony was in the location of the "safety" hit. It was in a side list of categories of previous articles, the group was "food safety."

Somehow, I suspect if I did a search on "salmonella" and "safety", I wouldn't get much on the effect car size has on survival chances in car crashes.

Regardless of the lack of media interest, don't be foolish. Buy a mid-size or larger car. In an average driving life in America, you have about an even chance of getting in a potentially fatal car accident.

"When hunting rabbits in lion country, beware of lions. When hunting lions in rabbit country you may ignore the rabbits." Eat tomatoes and buy big cars.

Charter Schools, Experimentation and Free Markets

The USA Today has an editorial debate on the value of charter schools. To their credit they seem willing to examine the charter experience carefully and try to encourage success where it occurs.

A (actually, this) scientist, viewing the intellectual landscape of the education debate might despair. There are too few controlled experiments and too few calls for them. Both are outrageous. Despite that, we occasionally learn something with controlled analysis (e.g., teacher certification is worth little or nothing) and despite its importance, the lesson is not applied.

If the governmental or central approach to schools was going to work in improving schools, it would be by the above, scientific approach. Pretty clearly that path isn't working. We probably teach no better than the classical Greeks.

But, this is America. The land of free markets, independent thinkers and doers. There is another way. We can "Let a thousand flowers bloom, let a hundred schools of thought contend". Apt, ignore the quote's source or remember that lovely flowers grow in manure.

This perspective makes particularly amusing the invited Opposing View to the USA Today editorial by Alfie Kohn. His lead point is "Will charters strengthen public education — or pave the way for vouchers and other privatization policies? As superintendents George and Mary Garcia warned, "The law of supply and demand, where winners make all the money and losers go broke, is a tragic idea to introduce into an institution whose purpose is to transmit democratic values and ensure equity for all." "

Wow. Do these fools think the winners and losers here are the children? The losers will be the bad charter schools, and a good thing that will be. The winners will be the good charter schools, and all the children after the bad charter and bad public schools close for want of "customers".

Stop Light Coalition: Red, Green, and Yellow

The long primary fight among the Democrats has pushed the winner, Barack Obama, left, or more accurately, prevented him from tacking right, toward the political center. This was caused, of course, by the need to appeal to the American left which has a dominant place in the Democratic Party and has coalesced into a stop-light coalition.

The three legs of a modern political alliance are economic policy, social policy and foreign policy. The left is currently Red on economics (well, pinkish actually) advocating a strong redistribution ("from each according to his ability; to each according to his need"), Green on social policy (major sacrifice to cut CO2, strict, even strangling, regulation of industry and commerce) and Yellow on foreign policy (talk to anybody, retreat from anywhere).

John Kennedy is rolling in his grave.

"Let every nation know, whether it wishes us well or ill, that we shall pay any price, bear any burden, meet any hardship, support any friend, oppose any foe to assure the survival and the success of liberty." - John F. Kennedy

Old People Got No Reason to Live

Anyone who goes to a gym regularly has shared the experience. Gyms are filled with young people. Some of us have been shaking our heads over that for years; young people should work out, old people must work out.

The value of exercise for older people is widely reported. It seems likely to delay osteoporosis, fight cardiac disease and (perhaps via improved circulatory function) slow dementia. Hard to argue with the benefits.

Yet, unfortunately, few seniors exercise. Perhaps worse, recent publicity campaigns and the spread of the culture of exercise hasn't budged the exercise habits of the seniors. According to CDC surveys published in a USA Today column by Kim Painter today, about 20% of seniors exercise regularly. A number that is essentially unchanged over the last 10 years.

What does it mean? Old dogs really don't learn new tricks. Exercise appears to be a habit that the seniors won't learn. Sure, the amount of pub on 'exercise and aging' could be doubled or redoubled, but the message actually has been delivered, particularly of the last 10 years. It just hasn't been generally heard. There is no earthly reason to suppose that will change.

For the vast majority of people, it looks as though the only practical solution is to start young. Easy to say, hard to do. Perhaps, just not as hard as starting when old.

Tactics May be the Issue

The Republicans are bemoaning, and the Democrats celebrating, the third straight Democratic victory in a strong Republican congressional district. Three does make it look like a significant trend. The nature of the trend is debatable. While most pundits are arguing for a change among Republicans of message or viewpoint, I suspect too little is made of effective Democratic tactics. In each case, and in many of the seats the Democrats picked up in 2006, the Democrats ran a candidate who had many conservative aspects, often using a service veteran.

The juxtaposition of the most recent congressional triumph of the Democrats, running a DINO, with a recent Wall Street Journal piece from Pat Toomey defending RINO hunting among Republicans is telling. Either party can ably contest districts more sympathetic to the main message of the other, if they run candidates appropriate to the district. As long as the Republicans shun RINOs and the Democrats woe DINOs the US Congress is likely to remain Democratic.

Grading Teachers; Should Student Results Count?

There is a lot to laugh about in this title's question. Laugh-so-you-don't-cry kind of laugh. The Wall Street Journal published an article by John Merrow a few days ago covering the basics of teacher evaluation based on student test performance. My laughter comes from the subsequent letters.

Today's WSJ had a set of these. A couple are logical, straightforward comments saying that results must be the basis of any evaluation. The laugher was caused by a reply from a teacher with 20 years of experience, Karen Fisher, in New York. She writes: "Blame the teachers and the unions--how often do we have to hear the same old tired arguments as to why the American educational system is failing? ... Sorry parents--when your kids don't do well in school, it is usually due to lack of discipline at home."

What is, sort of, funny is that one of the folks engaged in teaching America's young how to think flunks elementary logic. Karen, even if, as you suggest, home-discipline is an important part of the problem, it does not follow that teachers are not an important part of the problem too. This analysis suggests that teachers account for 15%-30% of the variance (that's a lot in a social science setting).

Now, I strongly suspect home discipline is important. I also think innate talent is important as well as peer culture. There is not a lot of good data, but what there is (summarized nicely by Judith Rich Harris in The Nurture Assumption) suggests that peer culture and genetics are both more critical than home environment.

Sadly, teachers and their unions seem to be in step with Karen on all this, sailing down deNial.

What the Libertarian Party and Barr Should Do

BThe entry of Bob Barr, former congressman from Georgia, as a candidate for the Libertarian Party's Presidential nomination may have generated record coverage for that party. The exposure provides the Libertarians with a chance to sell the "product."

Here is what I think they should do. Run only in Georgia. That's it. Focus all effort on Georgia where the results of the (probable) McCain/Obama contest is not in doubt. Voters there will feel free to vote Libertarian. The advantages of such a focus are: Barr is best known in Georgia; this, novel, one-state strategy will accrue further attention to his candidacy; the limited resources available will go much farther spent in one state only.

The disadvantages are slight. The party would pass up the opportunity to battle Nader and the Greens for the "Top of the Irrelevant" crown. Instead, the Libertarians should try to enter the national political consciousness by getting some 5% or even 10% of the voters in Georgia. That would get them, and their ideas, talked about.

Why is This Man Alone?

Patrick Moore is a co-founder of Greenpeace. In a recent speech in Boise, he re-affirmed his support for a major expansion of nuclear power. The Idaho Statesman's Rocky Barker reported:

Greenpeace founder Patrick Moore says there is no proof global warming is caused by humans, but it is likely enough that the world should turn to nuclear power - a concept tied closely to the underground nuclear testing his former environmental group formed to oppose.

The chemistry of the atmosphere is changing, and there is a high-enough risk that "true believers" like Al Gore are right that world economies need to wean themselves off fossil fuels to reduce greenhouse gases, he said.

Ignoring the silly "...nuclear power - a concept tied closely to underground nuclear testing ..." from the reporter, Barker, (about as closely tied, incidentally, as ordinary munitions testing is to hydrocarbon power plants), this should be the main policy position of sensible greens. He is right about anthropogenic global warming, "unproven but likely" is the actual scientific consensus. Nuclear power is safe (another scientific consensus), and is the only practical substitute for coal/gas/oil (economists' consensus as I read it).

Support for Patric Moore's position, somewhat oddly considering that it has a strongly environmentalist aim, is from folks like Lemuel Calhoon on the right rather than from the left. The fact that Moore is something of a lone voice among greens, suggests to me that the main motive of most supporters of severe greenhouse gas limitations is to punish industry or capitalism rather than to limit warming.

Double Digit Numerology: Clinton won by 9%

Hey, I don't much care, but the standard passing around as a marked victory for Clinton in Pennsylvania was 10%. I don't care, that is, about Clinton much. I do care about numbers.

The way to maximize accuracy in performing arithmetic is to retain all the accuracy practical and then round off at the end. If you need to convince yourself of that, take a few examples and experiment. You will find if you round off first then add (or whatever) you will often have some inaccuracy.

According to the Pennsylvania state election site, with 99.51% of precincts reporting Clinton has 1,238,351 and Obama has 1,030,805 votes. That's 54.5732% for Clinton 45.4268% for Obama. That is a difference of 9.1464% or 9.146% or 9.15% or 9.1% or 9% depending upon how you want to express it.

If we want to engage in numerology and set a 10% threshold, she didn't make it. She won by 9. I haven't seen this pointed out elsewhere. Wonder if I am in the first 100,000 sites to report it, do I get anything?

I attempted to criticize science journalists yesterday for their groupthink regarding global warming. I should probably have been more understanding as the enforcement of this groupthink can get rough.

About a week ago NPR had a feature on a 16 year old blogger, Kristen Byrnes, who has a (currently flooded) website, PonderTheMaunder, which is quite sophisticated for a 16 year old. Amazingly, NPR has been attacked rather viciously from the left because the website is strongly skeptical of the evidence backing much of the global warming idea.

David Appell is a science journalist who is a prominent blogger (QuarkSoup.com) who leads the way with this post that was the first Google search hit with opinion on the subject (third overall). He writes:

Yesterday NPR published one of the most atrocious, absolutely embarrassing pieces of scientific journalism I have ever witnessed. [wow, reminds me of the "simplify then exaggerate" dictate of sensationalism, not the scientific approach -jim]

David Kestenbaum -- who I thought was hired by NPR because he had some scientific training -- profiled a girl in Maine with a Web site that questions the canonical view on global warming:

Ms. Byrnes' Web site is an absolute joke, full of errors, entirely unscientific. For example, her graph of CO2 levels in the atmosphere showed it flat until about 1950, which is simply wrong. I corresponded with her for months and months about this, as did many other bloggers, until she finally hid her error without acknowledgement of what was correct. [wow again, Mr. Appell spends months harassing some teenager about her web site, actually gets some change but wants acknowledgment. Sounds a little like school yard stuff -jim]

.....

Profiling a high school girl as if her science stands up to the best peer-reviewed results from professional scientists and academic journals, as if she matters one iota, is a joke and reflects very, very poorly on NPR, and especially on their science desk. ...
Kestenbaum should be ashamed. Personally, I will never be able to believe anything he ever reports again, and, for that matter, the entire NPR Science Desk. This is pathetic.

The emotionalism in the comments is even worse.

This is a 16 year old kid. She got some stuff wrong. So did Al Gore despite serious scientific advisers. So, NPR wants to do a human interest piece on a kid's science site, focused much more on the kid than on global warming ideas, and Appell concludes "Personally, I will never be able to believe anything he reports again, and, for that matter, the entire NPR Science Desk." The imagination boggles at the thought of the reaction if NPR actually did a piece featuring the serious global warming skeptics.

Earth Day, Food Riots from "Fighting" Global Warming

Earth Day. Trees, photons and electrons are being sacrificed today to catalog the damages that may follow a 2 to 4 deg (C) warming over the century. Did the environmentalist lobby buy the biggest ad spread in history? If there is any discussion of the cost of fighting global warming it will likely take the form of "less than 2% of world income" or some such offhand attempt to downplay it.

Meanwhile, there are food riots across the world. No, this is not caused simply by the ethanol-for-fuel effort but that effort has strongly contributed.

My interest here is to observe just how selective the global warming coverage is. The mainstream media's coverage of the food riots often mentions (as it obviously should, summarized here) the contribution of alternative-fuel use bidding up the price of crops. It would be appropriate to see it mentioned equally in discussion of global warming. Do your own research. I have not seen much (actually none in the 5 pieces I picked to sample from a Google search on "global warming" and "earth day"). Bloggers, of course, have not been reticent (Robert Rapier, John Ray, Southchild have all been caustic).

We have an early read on the success, or not, of a tentative baby-step in the US effort to fight global warming. The step was typical ham-fisted government action. One evident result was higher food prices and probably famine. On this "Earth Day" the possible costs of doing nothing about global warming are dwelt upon constantly. The costs of doing something are not. These latter costs need far more attention from the public.

Science Debate 2008

Yesterday was the day the "Science Debate 2008" was supposed to be held. The lack of interest was marked. I find two ways of looking at this. On one hand, it would be nice to see Presidential candidates focusing on concrete science-related issues which a science debate might help. On the other hand, any politician on the right would be a fool to step into anything run by the "Science Debate 2008" group.

The group labels itself nonpartisan and trots out a few Republicans to make that point. Token Republicans. Window dressing. The ScienceDebate2008 site seems to be under deconstruction now, but my random sampling of the bloggers listed on the site (e.g., Coturnix, Kriswager, Stoller & Bowers, David) shows a strong left lean. The science Professoriate, which is the core of American science, is also heavily weighted to the political left. The agenda will be set by their beliefs. The questioning would have been nonpartisan in the sense that these folks would focus on what their concerns are, not necessarily the talking points of the Democratic National Committee. Unfortunately for the right, there is not that much difference.

The issues chosen, and the emphasis within the issues will reflect these biases. If you were a candidate from the right would you trust the ScienceDebate2008 "nonpartisans", when discussing the inadequacy of US science education, to focus on the results and education experimentation allowed by vouchers, charter schools and home schooling, or would you suspect the emphasis would be on spending more money at public schools?

Another example is embryonic stem cell research which is widely mentioned as an important debate topic. Here a key point will be how much we could help sick and dying patients if only this research were allowed. This is a big winner for the left. When focus is on the research, most opinion polls show Americans want stem cell research. Instead, why not focus on the moral issue causing not just this fight but the larger one on abortion? When does life begin? Science has a lot to add to that conversation.

By rough analogy one could discuss the issue of organ transplantation. Medical science is expanding the range of possible organs and improving the outcomes for those transplants currently done. The biggest hold up is a lack of donors. A (partly) free market for donations would bring many thousand forward, saving many lives and improving the quality of many others. Here, however, the moral hesitancy preventing this is (mostly) from the left and so a Science Debate controlled by the left would not judge this to be a health issue but a moral one. Not, coincidently, there is little mention of organ transplantation as a ScienceDebate2008 issue.

Given the political bend of the scientists and journalists likely to control a Science Debate, a candidate from the right side of the political spectrum ought to be very wary indeed before straying into that den.