TED Talk Tuesday: Part 2

How statistics impacts jury trials and can lead to incorrect convictions.

Measuring Ministry Success without Worrying About Numbers

A nice piece on effective youth ministry from Kurt Willems today. (HT: Rob Martin) Willems talks about his experience early in his career as a youth pastor where he was basically told that the worth of his ministry was related to how many students showed up. Willems suggests a different economy:

Many people ask me what we should measure if it’s not numbers and attendance. I think I’ve finally begun to figure this out. I recently took a course on church planting. In this class, one of the things we talked about was the need to change the ministry scorecard (to borrow Reggie McNeil’s language). Instead of measuring success by numbers and quotas, what if we measure success by stories of how God is at work through various signs of the kingdom that we see in our context? In other words, in ministry and in any church function, what is our primary goal? Is it to meet quotas? Is it you rant and rave about how many showed up? Nope, it’s not all about the numbers! It’s all about the kingdom!

If we want to know if a particular ministry is being “successful” we should ask the following question: What signs of the kingdom have we seen or experienced during the past week? All other measurements of success fall subservient to that single question.

Check out the rest here. I think we do a pretty good job of that at McBIC, though I know the numbers have come along as well. Our focus is on what God is doing among the youth.

Thoughts on the Criminal Justice System

In this TED Talk, Bryan Stevenson challenges his listeners to evaluate whether the current way that our criminal justice system is run is working. We spend billions of dollars, but rarely are willing to ask the hard questions about whether it is working. In short, I don’t think it is. Perhaps his best challenge: rather than asking whether the criminal deserves to die, why aren’t we asking whether we deserve to kill? Please take the time to watch this. I think we can do better for victims and criminals.

We’ve Got Great Alumni

Here is the latest example of our great alumni in the Department of Information and Mathematical Sciences here at Messiah College. In the controversial rankings of 4th through 8th grade teachers in New York City, class of 2005 alumna Kelly (Toolan) Hudson is listed as the 10th best mathematics teacher in the city. She’s also the second 8th grade teacher listed. Like all rankings, the methodology is fair to discuss. The methods seemed to be an attempt to model the benefit a teacher gave their students on standardized testing. Other variables were controlled for. While I’m not a fan of basing everything on standardized tests, I do think that this methodology does not take away from the honor that this ranking bestows. Congratulations, Kelly!

How to Bomb the AP Stats Test

To those of us who grade AP Stats, this is hilarious. If you are not one of those, you might learn something about how not to do statistics.

Interesting Statistics from 60 Minutes

I can’t embed the videos here at this point, but here are a series of videos from recent 60 Minutes episodes that reference statistics in the news.

First, on a recent case of forged credentials and misrepresented data at Duke University.

• The main piece can be found here.
• A bit about the researcher and his overstated credentials.

Secondly, a study of the placebo effect that seems to indicate that anti-depressant medication may be nothing more than a very expensive investment in the placebo effect.

• The main piece about looking at all of the research over the years and finding a stunning lack of significance.
• The journalist reflects on the story.

As a statistician, the ability to get a drug approved if two studies show an effect and ten studies show none is ridiculous. This is not good scientifically or statistically.

On Fracking, the Science is Clear

An Illustration of a Fracking Well

Unfortunately, while the science on natural gas fracking is clear, the public debate does not reflect this fact. Why? Here is a take from an op-ed piece in the NY Post by Jon Entine:

The academic face of the anti-fracking movement — Cornell marine ecologist Robert Howarth — increasingly looks like he’s willing to turn science into farce.

Last spring, the once-obscure professor became the go-to expert for anti-fracking journalists and lawmakers when he published a report claiming shale gas pollutes more than coal. The New York Times featured his study in two uncritical articles in one week, he was interviewed on dozens of talk shows — and the media echo chamber did the rest: He was a star.

Since then, other scientists have almost universally challenged his findings — but now he’s doubled down.

Last week, Howarth released another scientifically questionable study, now warning that fracking could push the world over a tipping point, sending temperatures irreversibly higher — an inflammatory and demonstrably incorrect assertion.

After some backstory, Entine turns to the science and Howarth’s motivation:

In an interview, Howarth told me his goal was to make the anti-fracking movement mainstream and fashionable. He said he met with the Ithaca-based foundation two years ago, agreeing to produce a study challenging the conventional wisdom that shale gas is comparatively clean.

The polluting impact of shale gas revolves around one key issue: how much methane gas is released during extraction. Methane has more short-term global-warming impact than any other fossil fuel. Howarth emerged from academic nowhere when he claimed shale-gas wells leak like sieves, venting methane half the time, spewing 7 percent to 8 percent of reserves into the atmosphere.

“That’s absurd,” says Michael Levi, director of the Program on Energy Security and Climate Change at the Council of Foreign Relations. “Most methane gas is either ‘delivered to sales’ with no leakage, or it’s burnt off through flaring, which diminishes its greenhouse impact.”

Renowned geologist Lawrence Cathles, also at Cornell, who published a scathing deconstruction of Howarth’s paper this month, says that he “doesn’t document venting but what the industry calls ‘capture.’”

Almost every independent researcher — at the Environmental Defense Fund, the Natural Resources Defense Council, the Energy Department and numerous independent university teams — has slammed Howarth’s conclusions. At a minimum, the evidence suggests he either acted in bad faith or is ignorant of gas technology.

The core problem: Howarth uses Environmental Protection Agency estimates dating to 2007 — ancient data, given how quickly the technology is evolving.

Crucially, he fails to account for innovation. Gas lost through leakage is money lost, literally into thin air. For that reason, new wells are now “green completed” — meaning most leaking gas is captured and sold rather than vented.

Cathles notes the latest Devon study, now being verified by the EPA, documenting that shale gas is vented in only 5 percent of wells. The Energy Department estimates only 1 percent to 2 percent of methane is now lost during production.

Bottom line, almost all nonindustry-linked researchers believe Howarth exaggerates the impact of shale-gas leakage by 10 to 20 times. “His conclusions are more a politically charged articulation than a balanced scientific assessment,” Cathles says.

Howarth hired an aggressive PR firm, the Hastings Group, to promote his politicized viewpoint. Scientists aren’t buying it, but many journalists fall for the fear-mongering.

Howarth doesn’t have to convince anyone he’s right to devastate New York’s budding shale industry and put tens of thousands of jobs into question. He wins if he muddies the waters enough to give cautious Albany bureaucrats reason to stall.

Almost every news story now frames this issue as a standoff between equally valid scientific experts. In fact, it’s really a debate between science and ideology.

You can read the rest of the piece here.

It Is Exam Day Again …

In honor of the Intro Stat exam I gave from 1:00-2:00 or so today:

Stat/Probability Quiz of the Day

Good luck with this one!

Demography: Predicting the Downfall of China?

A very interesting look at where all of the earth’s billions of people are now, and where the population is heading. Projections place the occurrence of the world population of humans passing 7 billion at some point between 31 Oct 2011 (UN estimate) and March 2012 (US Census Bureau estimate). The most interesting piece of this article is the pointing out that the rate of increase is now slowing. (The second derivative is now negative, for my math inclined readers.) Where the fertility rate is high, and where it is low, helps to predict the economic future of countries around the world. The pertinent section follows this graph of fertility rates (live births per woman).

The low-fertility countries face the biggest demographic problems. The elderly share of Japan’s population is already the highest in the world. By 2050 the country will have almost as many dependents as working-age adults, and half the population will be over 52. This will make Japan the oldest society the world has ever known. Europe faces similar trends, less acutely. It has roughly half as many dependent children and retired people as working-age adults now. By 2050 it will have three dependents for every four adults, so will shoulder a large burden of ageing, which even sustained increases in fertility would fail to reverse for decades. This will cause disturbing policy implications in the provision of pensions and health care, which rely on continuing healthy tax revenues from the working population.

At least these countries are rich enough to make such provision. Not so China. With its fertility artificially suppressed by the one-child policy, it is ageing at an unprecedented rate. In 1980 China’s median age (the point where half the population is older and half younger) was 22 years, a developing-country figure. China will be older than America as early as 2020 and older than Europe by 2030. This will bring an abrupt end to its cheap-labour manufacturing. Its dependency ratio will rise from 38 to 64 by 2050, the sharpest rise in the world. Add in the country’s sexual imbalances—after a decade of sex-selective abortions, China will have 96.5m men in their 20s in 2025 but only 80.3m young women—and demography may become the gravest problem the Communist Party has to face.

Many countries with intermediate fertility—South-East Asia, Latin America, the United States—are better off. Their dependency ratios are not deteriorating so fast and their societies are ageing more slowly. America’s demographic profile is slowly tugging it away from Europe. Though its fertility rate may have fallen recently, it is still slightly higher than Europe’s. In 2010 the two sides of the Atlantic had similar dependency rates. By 2050 America’s could be nearly ten points lower.

But the biggest potential beneficiaries are the two other areas with intermediate fertility—India and the Middle East—and the high-fertility continent of Africa. These places have long been regarded as demographic time-bombs, with youth bulges, poverty and low levels of education and health. But that is because they are moving only slowly out of the early stage of high fertility into the one in which lower fertility begins to make an impact.

At the moment, Africa has larger families and more dependent children than India or Arab countries and is a few years younger (its median age is 20 compared with their 25). But all three areas will see their dependency ratios fall in the next 40 years, the only parts of the world to do so. And they will keep their median ages low—below 38 in 2050. If they can make their public institutions less corrupt, keep their economic policies outward-looking and invest more in education, as East Asia did, then Africa, the Middle East and India could become the fastest-growing parts of the world economy within a decade or so.

Demography, though, is not only about economics. Most emerging countries have benefited from the sort of dividend that changed Europe and America in the 1960s. They are catching up with the West in terms of income, family size and middle-class formation. Most say they want to keep their cultures unsullied by the social trends—divorce, illegitimacy and so on—that also affected the West. But the growing number of never-married women in urban Asia suggests that this will be hard.If you look at the overall size of the world’s population, then, the picture is one of falling fertility, decelerating growth and a gradual return to the flat population level of the 18th century. But below the surface societies are being churned up in ways not seen in the much more static pre-industrial world. The earth’s population may never need a larger island than Maui to stand on. But the way it arranges itself will go on shifting for centuries to come.

Here are a couple more helpful graphs of the world population and world fertility rate.

World Population Graph: Note the change in concavity in the last few years.

The Declining World Fertility Rate

You can read the rest here.