Gapminder, a great website

One of my very first posts was about Hans Rosling's "200 Countries in 200 Years" video, which shows the growth in income around the world across two centuries. Now here's a link to Rosling's own site, Gapminder.com. It contains links to many of his famous TED conference videos, including the "Magic Washing Machine," embedded below.

But Gapminder has a lot more than that. One of the best pages is the "Health and Wealth of Nations" chart that lets the viewer compare all sorts of metrics around the world. Interested in seeing carbon emissions per person by country? It's there, and also available by region and world wide. Health, education, population, energy, environment work - it's all there, and it's all accessible. There's even a guide for teachers. It's creative and useful. And fascinating.

More anecdotal evidence of global warning

As if any were needed: the New York Times sports section has an article today describing high altitude glaciers melting. If it weren't such an inappropriate adjective I'd call it chilling. But I suppose scary is the operative adjective. Glacierworks has a graphic photo illustrating the difference in one glacier. Here's how it looks now:

and here's how it used to look:

Source: Glacierworks

If you haven't seen one of the "Rivers of Ice" exhibits (underway now in London and Beijing) there's another US one slated at the MIT Museum in Cambridge, MA, opening in April.

Football and grades

You may have seen the story, since it's being widely reported, about a new study by three University of Oregon economists finding a decline in grades, especially among male students, when the school's football team is doing well. You can get the full study ($5) here.

Two things about the article: first, according to the NY Times, the study, while large, looked only at Oregon. Second, the article reports that the study found a correlation, though the comment in the article by the University of Oregon spokesman refers to causation. They're not the same thing. But the study points to yet another manifestation of the negative effects of big-time college sports. If you haven't read it already, read Taylor Branch's careful and thorough article in The Atlantic, "The Shame of College Sports."

Perspective matters - interesting chart

The McKinsey Quarterly Newsletter sent this interesting chart this morning. It appears to be telling us that the US, Russia (I think), France, Germany, India and Japan provide a large proportion of box office revenues. And that domestic filmmakers consume more than 40% of market share in the US, Russia, India and Japan. What I can't tell from this is whether McKinsey means US domestic or each country's own domestic film industry. The answer may be obvious to an industry insider, but the chart needs to do better to reach more people.

McKinsey's larger point, that a map can "correct the misimpression that a viewer's vantage point doesn't influence the way things look" is a valid one.

10 Years of Student Assessment

Michael Winerip's column in today's NY Times describes a series of statements, retractions, and, well, mistakes in New York State and City's attempts to assess public education. I've written about the City's efforts before, for example here (about the City's school rating system) and here (about teacher ratings). All I want to add now is that coming up with metrics is an iterative process, with lots of starts and do-overs. And, as Regents Chancellor Merryl Tisch continues to point out, the numbers have to mean something. I think that assessment of public education is something that can be done. I'm not sure that relying entirely on test scores is the way to do it.

One way to start fixing primary and secondary education is to provide good preschool education. If you haven't ever done so, take a look at the Highscope/Perry Preschool research material. It's pretty interesting, and, combined with the recent article about the declining number of "universal" pre-K spots (not to mention the inadequate numbers of child care spots) pretty compelling. If even some of the money spent on testing and test prep in the last decade had been spent instead on expanding high quality preschool education, New York's test scores might be showing a different trend.

I will be taking a little time off over the coming weeks for the holidays and don't expect to be posting as often. See you in January!

"All models are wrong but some are useful"

UPDATE, Feb 9: Kaiser Fung's blog has a useful illustration of the variability point discussed below in a post about representations of LA's annual rainfall, here.

Kaiser Fung, a statistician and blogger (also here - I've written about his excellent site, "Junk Charts," earlier) has written a terrific book about what he calls "the statistical way of thinking." In it, he explains how to think about what happens when numbers don't lie. Here's a screenshot of the cover, from Amazon.

In five easily digested chapters, using events from newspaper reports or even everyday life to illustrate, he makes five basic points about statistical analysis that everyone should know.

1. The average isn't as important as the variability around the average - how large the variation is, how often it happens, and why are all more useful information. Fung illustrates this principle by discussing the wait times for rides at Disney theme parks. Think you have a long wait ahead of you? The "Fast Pass" allows you to come back at a set time, when numbers are lower. You feel as if you are not waiting, but of course you've gone and done something else while you waited.

2. Often, variability doesn't have to be explained by causes - correlation can be enough to make the information useful. Fung contrasts an epidemiological search for an outbreak of food-borne disease with consumer credit ratings. In the latter, correlation is a good enough reason to extend or deny credit. Epidemiologists need to be more cautious, and Fung reiterates Bradford Hill's nine aspects of cause and effect.

3. Not everyone or everything can be aggregated, and aggregation can mask differences. Fung points to differences in SAT scores among black and white students, which are masked when all scores are analyzed together. "Black students of high ability score the same as whites; the scores for low ability students of both races are also the same. And yet the average white student score  . . . is higher than the average black student score. Due to superior educational resources, 60% of white students have high ability compared to only 20% of black students."  The averages mask the weights, or the greater number of white students in the high-achieving group and of black students in the low-achieving group.

4. Making judgments based on statistics can force you to balance two types of error, over-inclusion and under-inclusion. Unfortunately, these two types are inversely related, so by lowering the chances of one you increase the chances of the other. Fung's example in this case is drug testing for athletes, and he shows that by setting the cutoffs for positive tests very high, there a lot of false negatives among tested athletes. Remember that next time you read a statement saying an athlete has never failed a drug test.

5. Statistical testing can help us decide whether available evidence explains an event. When sellers of Canadian lottery tickets turned up as winners at a rate considerably greater than chance would have us expect, Fung explains, it was unlikelihood of it happening (one in a VERY large number) that set authorities down a path to correct the wrong and change processes so that it was harder for vendors to cheat.

This is a clearly written, lucid book. After carefully setting out and illustrating the five main concepts, Fung demonstrates how statistical tools are often used together, going back to his original illustrations with a deeper explanation. It's very satisfying, and very clear. I highly recommend it.

The pitfalls of evaluation

Who knew? The most e-mailed article on the Stanford Social Innovation Review website is dated  2006, and titled "Drowning in Data." Whether you're a funder or a service provider it's still a useful article to read five years after publication. Data collection, analysis, and reporting is hard, and the article outlines several of the reasons. One is that terminology is not yet standardized - my outcomes may look like implementation to you. Another is that organizations are often over-ambitious, wanting to know about outcomes that can occur only several years down the road without providing funding to develop that kind of information.

The main focus of the article is the disparate data requests generated by funders - and often, each requires its own form for reports even if data are similar. I've had to gather and report this kind of data, and it can be a problem, not least because the process may not generate data that's useful for managing a program.

The article distinguished what it calls "summative" evaluations: did the intervention "work"? from formative evaluations: does the evaluation help the organization "improve"? You can see where I'm going - it's not clear, from this context, what it means to say that an intervention worked, or that an organization improved. One thing I always tell clients is that the process of developing measures is important, and that you're going to be doing it over and over again. (I know, that's really two things.)

I don't think that there's an argument here for not trying to evaluate. But there is very good reason to be thoughtful about doing so.

Earthquake magnitudes visualized

The Richter scale for measuring earthquakes is logarithmic, so, as the USGS earthquakes website puts it, "each whole number increase in magnitude represents a tenfold increase in measure amplitude. . . " It's hard to imaging, so here's a startling visualization of the difference in energy by Nathan Becker, a government geologist:

I found this video thanks to Alexis Madrigal's blog on TheAtlantic.com

From Silicon Alley Insider: ATOS, a France-based international IT services company, is switching all its internal communications from email to IMs and an internal social networking interface.

As ATOS puts it on its website:

The volume of emails we send and receive is unsustainable for business. It is estimated that managers spend between 5 and 20 hours a week just reading and writing emails. Furthermore, they are already using social media networking more and spend around 25 per cent of their time searching for information.

A Facebook-like (or Google +) interface makes sense for an office - you can post information that now you would send in an email to umpteen different people. With a better interface, you can make the information readily available when it's needed, and the internal company newsletter can become more like a blog or a wiki.

If you go to the SAI article, notice two things: first, the very interesting chart showing the change in web-based email usage by age. Web-based email is only part of the issue, but still the chart shows an interesting dynamic. Second, and perhaps I am late noticing this, the green links in the article are to advertisers.

SAI calls the ATOS move a bombshell, but the news has spawned an interesting public debate on LinkedIn. What do you think? Brilliant move? Marketing move? Mistake?

Some horrifying statistics on children living in poverty,

in two senses of the word.

Charles Blow's column in Saturday's NY Times contains a scathing deconstruction of Newt Gingrich's statements about poor children, in which Gingrich said that poor children do not have people around them who work so they don't develop work habits. Blow's take:

This statement isn’t only cruel and, broadly speaking, incorrect, it’s mind-numbingly tone-deaf at a time when poverty is rising in this country. He comes across as a callous Dickensian character in his attitude toward America’s most vulnerable — our poor children. This is the kind of statement that shines light on the soul of a man and shows how dark it is.

Gingrich wants to start with the facts? O.K.

First, as I’ve pointed out before, three out of four poor working-aged adults — ages 18 to 64 — work. Half of them have full-time jobs and a quarter work part time.

Furthermore, according to an analysis of census data by Andrew A. Beveridge, a sociologist at Queens College, most poor children live in a household where at least one parent is employed. And even among children who live in extreme poverty — defined here as a household with income less than 50 percent of the poverty level — a third have at least one working parent. And even among extremely poor children who live in extremely poor areas — those in which 30 percent or more of the population is poor — nearly a third live with at least one working parent.

So far so good. It's clear that Blow is mad that Gingrich is ignoring a lot of readily available facts. And that the facts are so damning.

But here's a screenshot of (part of) the accompanying graphic. (The link takes you to the full graphic.)

(Source: NYTimes.com)

Can you figure out what is going on with these numbers? Neither could I. The colors distinguish whether there are working parents in the household, I get that. And the relative size of the squares tells you how many people are considered "poor" or "extremely poor." But what's with the detailed numbers? Rounding them to the nearest 10 thousand or even 100 thousand would be clearer. And the numbers running down the middle between the two sets of boxes? Notice that they overlap, so that areas with 10% poor households show up twice. I think the Times is trying to show that even in poor neighborhoods, there are lots of working parents. Right, there it is, buried in the text:

And even among children who live in extreme poverty — defined here as a household with income less than 50 percent of the poverty level — a third have at least one working parent. And even among extremely poor children who live in extremely poor areas — those in which 30 percent or more of the population is poor — nearly a third live with at least one working parent.

 These are important data. There's got to be a better way to show them graphically.

"Steve Jobs" by Walter Isaacson

Here's a link to my review of Walter Isaacson's biography of Steve Jobs pm the Brooklyn Bugle. I don't think this bio is going to be the last word on the subject of Jobs and his legacy, but it's a terrific beginning.

In honor of Durban, more on global warming

Here's a link to an article from New Scientist via Slate, about extreme weather events, global warming, and a new technique British meteorologists want to use that considers the odds of particular events to assess the extent of human-induced climate change involved. (You know: was Hurricane Katrina really a 100 year storm, or was it worse than it might have been because of global warming?) Peter A. Stott, a climate scientist in the UK, wrote the article. We don't have a good enough system yet, he says, but:

What we need is an attribution system, operated regularly like the weather forecast and made available to the public. Its purpose would be to deliver rapid and authoritative assessments of the links, if any, between recent extreme weather events and human-induced climate change.

In the event of, say, a severe flood, the system would provide estimates of the extent to which the event was made more or less likely by human-induced climate change. It would also take into account alternative natural explanations such as the El Niño Southern Oscillation, a large-scale climate pattern in the tropical Pacific Ocean that affects weather worldwide.

Such a system would be useful - it's hard to tell from one day to the next what is causing particular weather events. And it might even move the discussion in the US from whether climate change is happening to what to do about it.

Sometimes it helps to go outside US news sources, and today's BBC World News Service has a fascinating interview with David Attenborough about his new programs about global warming. I can't find a direct link to the interview, but you can listen to the BBC's webcast here. The Attenborough interview is in the Newshour.

And, FWIW, since I use Wikipedia a lot, here's a link to the Tech Crunch blog post complaining about the placement of Wikepedia's fundraising pleas.