Thursday, May 31, 2012

Distribution of adults with college degrees around the US

This graphic comes from today's New York Times; it illustrates a story about the changing distribution, nationwide, of college graduates. Increasingly, college graduates are concentrating in urban centers. The graphic is interesting for two reasons. First, the shape of the curve is changing: in 1970, 12% of adults in US metropolitan areas had college degrees - but the curve was steep, and most areas were within 5 percentage points of that percentage. in 2010, 32% of adults in metro areas had college degrees - but the shape of the curve has changes, with many more cities falling outside 5 percentage points, both above and below.

Second, notice that I have said that "some percentage of adults in US metropolitan areas" to describe the chart. The Times says "In 1970, 12% of adults had college degrees in US metro areas." This awkward language is confusing, I had to read it several times when I read the print version this morning, and it was only when I looked at the data the Times provides, a list of the metro areas with share of population with college degrees, that I understood what the paper was trying to tell me. This confusion could have been avoided with better wording. Am I right? Let me know in the comments if you think I've misinterpreted here.

Wednesday, May 30, 2012

Chicago and traffic fatalities

The City of Chicago has expressed an ambitious goal: to eliminate all pedestrian, bicycle, and overall traffic crash fatalities within 10 years. How does it intend to do this? According to one report, by increasing biking as the mode of transportation in trips of less than five miles, by 20 mph zones in residential areas, improving street maintenance and a focus on and address problem intersections.

A lot of programs and organizations are reluctant to specify performance measures, often for fear that they may not be able to accomplish them. But the point about performance measures is that, although it may have taken an organization a while to get to them, they are the starting point for analysis, not the end point. When you look at your performance, you don't just want to check to see whether you got there or not. You want to think about how close you came, what went right, and what didn't go so well. Performance measure reviews shouldn't be an exercise in casting blame. They should be an effort to make things work better.

Chicago may not make its ambitious goal. Ten years out is a long time; technologies will change; politics will change. Rahm Emanuel may not longer be mayor. But there are a lot of reasons to think that this is a well-thought-out goal. The City of Chicago released a detailed plan earlier this month - you can get a full (pdf) copy of the plan, Chicago Forward, here. The traffic crash fatality goal is the first goal set out under an overarching theme, Safety is Paramount. It's got a time frame. It does not stand alone, but is the first among several related goals, which include reducing pedestrian and bicycle crash injuries, reducing total roadway crashes and injuries, and increasing safety education for adults and children. And it's followed by a detailed list of research findings, policies and planned actions.

I plan to check back regularly, to see what updates are published. It will be interesting to see how Chicago progresses, and how it reports that progress.

Tuesday, May 29, 2012

McKinsey on Women in the Workplace

McKinsey has recently published its updated research on the advancement of women in the workplace, particularly large for-profit corporations. Among the  key findings are that two types of structures that advance women successfully have emerged in practice: either a lot of women start (more than 50%) and so more are available to move up, or women are retained at the same rate as men, and so are available to move up. In social services work, the first scenario is probably more common than the second, though that may be changing in recent years.

And what happens in the higher echelons? Again, McKinsey is reporting on what its research among large for-profit corporations found: "many women opt to take staff roles, get stuck in middle management, or leave their organization without giving the company a chance to address their concerns."

Sound familiar? It does to me. And in fact the report identified some familiar barriers:

* structural obstacles (ie, few women at the top)
* lifestyle choices (women reported being both the primary caregiver and the primary breadwinner much more often than men did)
* institutional culture and individual behavior (women don't knock on doors asking for advice the way men do; men will assume that a pregnant woman will not want to move to take up a new position)

And the report offers some specific suggestions, including hands-on leadership, diversity leadership with the clot to make things happen, and robust talent management - all of which can (and often are) be done by not-for profits.

It's an interesting look at an important picture. You can get a pdf of the full report here. (McKinsey reports are free if you sign up.)

Update: This post was edited for clarity.

Thursday, May 24, 2012

More good charts on health care costs

As you know, I keep an eye on Kaiser Fung's blog "Junk Charts." I'm also interested in the rising costs of health care. Here's a link to Fung's blog post "Look What I Found: Two Amazing Charts" about . . . health care costs. Screenshot of one is below, but read the full post for Fung's quick reprise about what makes the charts good.

via Junk Charts

Wednesday, May 23, 2012

Recipe organization

Organizing recipes, especially the kind clipped out of newspapers and magazines over many years, can be a challenge. Here's how one (perhaps obsessive? what do you think? let me know in the comments) New York Times staffer managed it. In case you are not ready to program your own recipe database, in addition to the options the article lists, you might also check out Evernote - I like it because it syncs across platforms, letting you link to recipes at your desk but look up ingredients on your phone while you're grocery shopping. And for another approach, try Cookstr, which allows searches by ingredients and has a separate gluten-free index.

Photo via

Tuesday, May 22, 2012

Admitting cognitive mistakes

In case you missed it over the weekend, here is a link to the NY Times story about how Dr. Robert Spitzer, author of a 2001 study that supposedly showed that self-reports of change in sexual orientation after therapy were credible, has now changed his mind. More than that, in a letter to the editor of the journal that published the initial study, he says, "I believe I owe the gay community an apology for my study making unproven claims of the efficacy of reparative therapy." You can read the text of the letter here.

In its exploration of a scientist's decision to admit he'd made a mistake it's quite a moving story. It's also a nice illustration of substitution, one of the shortcuts in thinking Daniel Kahneman describes in his fascinating book "Thinking Fast and Slow." As Kahneman puts it, "If a satisfactory answer to a hard question is not found quickly . . . [we] find a related question that is easier and will answer it." According to Spitzer's letter, he substituted an easier question (how do individuals undergoing reparative therapy describe changes in sexual orientation) for the hard one (can some version of reparative therapy enable individuals to change their sexual  orientation from homosexual to heterosexual).

Spitzer then compounded the study's flaws by persuading himself that the self-reports in the studies were credible. In Kahneman's terms, he "focuse[d] on existing evidence and ignore[d] absent evidence." We all make these kinds of mistakes, and Kahneman not only shows how, he offers some useful correctives. I highly recommend the book.

Sunday, May 20, 2012

Twitterati and Facebook's trading price

Update, May 26: Check out this chart from Bloomberg News, via Derek Thompson of The Atlantic comparing Facebook's IPO performance to other IPOs.

Update, May 24: Here's an article on Business Insider claiming that the correct valuation for Facebook is $29/share. 

In case you were tempted to buy Facebook on Friday, the day of its IPO: be glad you didn't. Via Alexis Madrigal's blog at, here's a look at predictions on Twitter for FB's closing price on its first day of trading.
Notice anything about the shape of the distribution? And remember where Facebook closed Friday? At just above $38, about where it started.

Thursday, May 17, 2012

Two good stats articles from the NY Times

OK, I spend a fair amount of time on this blog making fun of the NY Times' statistical reporting. So today, two articles where the Times does a good job explaining, reporting, and illustrating statistics.

I thought the Times did a great job explaining the census data released this morning, which show that 50.4% of the US population younger than age 1 are minorities as of July 1, 2011. Interestingly, when I read the Census Bureau press release, I learned also that there are four majority-minority states (Hawaii, California, New Mexico, and Texas). The District of Columbia is also majority-minority. I did think that the Times' graphic illustrating Generational Gaps in Minority Births was confusing (screenshot below). A better title might have been "States with Largest Generational Gaps in White Population" because that is what I think the graphic shows.

Source: The New York Times
I would also quibble with the "tipping point" characterization of the births graphic illustrating the article.
Any thoughts about this? Let me know in the comments.

In a different kind of article, I thought the Times did a very thorough job of explaining new research on HDL, "good cholesterol" and why simply raising HDL may not prevent heart disease. The story is a good illustration of the point that I've made and reiterated about using statistics: correlation and causation are not the same thing. I've just begun reading Jim Manzi's book "Uncontrolled," (I'll be writing more about that later). The episode is a good illustration of the progress of scientific knowledge he describes in his third chapter.

Wednesday, May 16, 2012

Slime mold and highways

You may have read about the experimenters in the UK who have put slime mold into boxes shaped like various countries or cities, put food at the nodes, and then watched what happened. . . . the slime mold recreated the US highway system, or the Tokyo transit system, to name just a couple. n case you find it hard to believe from the still pictures, here's a video of slime mold recreating the Canadian highway system.

Tuesday, May 15, 2012

How to lose weight

FWIW, here's a link to the human weight simulator designed by a mathematician at the National Institute of Diabetes and Digestive and Kidney Diseases. I've been playing with it today. That's a screenshot of one of my scenarios: energy intake and expenditure for a 30 year old male weighing 200 pounds who wants to bring his weight down to 165 in 6 months, and keep it there. My assumption is that he'd increase his physical activity by 20%. Oh, and the starting point is a diet of 2,669 calories per day. That has to be brought down to 1904 calories for six months - but if my guy can keep up the increased physical activity permanently, he can increase  his consumption to 2456 calories a day.

Here's the Times interview, worth reading for its insights that, a) we're getting fatter because food is cheaper and b) weight change takes a long time.

Monday, May 14, 2012

Paying for college and student debt

I was surprised to read in yesterday's New York Times that "94% of students who earn a bachelor's degree borrow to pay for higher education." That number is an increase from 45% in 1993, but it includes loans from relatives as well as the federal government and private lenders. I do wonder whether all those loans from relatives are paid back. Also, the Times isn't clear what's included in that 94% figure - the Project on Student Debt, which the Times says is its source, says that two-thirds of college seniors graduated with loans in 2010.

The Times  included a useful interactive graphic that allows the reader to look at debt-at-graduation data by enrollment size, graduation rate, share of graduates with debt, and athletic conference. Here's a screenshot of the 2010 data for all colleges (the browns are private schools, the grays public).

There's some interesting school by school data there as well. For example, even students at schools that do not charge tuition graduate with debt. (For example, 2010 graduates of Cooper Union averaged $10,400 in debt, about the same average debt as that year's Harvard graduates.) I suspect the reasons for the increase in debt is the increase in private loans in the mix of scholarships, work/study grants, and private and federal loans that most people now use to pay for college.

Update: Here's a link to a Derek Thompson column in the Atlantic framing the issue slightly differently: first, that most students graduate with less than $50,000 in debt, and, second, that 50% of 21-year-olds have dropped out of college or off the college track entirely. Take a look.

A few years ago I researched how to pay for college. Here, updated, is what I learned:

1. Make sure you understand the full costs of college
At first look, college costs can seem daunting -- $35,000, or $40,000, or more, per year, for a private college. For four years. But not every school costs that much. State schools often charge far less in tuition and fees, and if your student qualifies for in-state tuition, costs can drop dramatically. According to the College Board, costs come in five distinct components:  tuition and fees, room and board, books and supplies, personal expenses, and travel. The problem is that college web sites report costs differently.

The lesson: make sure you understand all the potential fees for each school your student is considering. Think about possible lab fees, and travel, yours and your student's. Will five members of your family be going to Parents’ Weekend? That will cost a lot more for a New Jersey-based family if the child is in college in Arizona than if the child is in college in Connecticut. Think about how many vacations the school has, and include them in your budgeting. Also think about allowance, and how much your child will need for a social life, laundry, or other interests like off-campus music lessons or culture. What if your child has a car?  Travel costs may decrease, but parking, gas, insurance and service expenses need to be budgeted.

2. Start saving early
According to the US Department of Education, the best time to start planning for college finances is in middle school: 6th, 7th, or 8th grade, or even earlier. Remember, the more you save, the more you will have, as compound interest means your savings will mount rapidly. 529 savings plans allow families to set aside money to pay for college; the income is free from state and local taxes so long as it is used to pay for higher education.

But even if your child is already in high school, it’s not too late to start saving! The College Board suggests that families think of it the way you would plan for buying a house:  the more you save now (for the down payment) the less you’ll need later from current income or borrowing. Take a look at the College Board's "Financial Aid 101" for more detailed information.

3. Understand the different types of Financial Aid
Financial aid, which can take the form of loans, work/study, or scholarships, is an option for every family. According to the College Board, students and their families are expected to contribute to the cost of college to the extent that they can, but if a family is unable to contribute the entire cost, financial aid is available to bridge the gap. The government will calculate an “expected family contribution” based on family circumstances, and assume that families will meet their share through a combination of savings, current income, and borrowing. A college may also calculate an expected family contribution using its own--different--methods. The expected family contributions takes account of savings, salaries, housing, and other children who are in college or secondary school.  

Much financial aid comes from the Federal government, usually in the form of low-cost loans.  The College Board says “Students apply for . . .  all federal, state, and institutional financial aid programs (except scholarships) by completing the Free Application for Federal Student Aid (FAFSA).”  You can complete the form online at  The application process is designed to be simple, so that families can complete the application easily, based on readily-available documents like their most recent tax returns.  (Be consistent!  The Department of Education will compare the FAFSA application with your tax returns.)  Remember, the system is designed to be easy for parents to complete; you should not need to pay for assistance with this application.

Financial aid can also take the form of grants. Many colleges will provide merit scholarships that cover some or all of a student’s tuition. The advantage of these grants is that, unlike loans, they do not have to be repaid. For more information about different types of aid, look at the Federal Government's Financial Aid Resource Publications, available here.

Once the student is admitted, take a hard look at the financial package; make sure you understand how much is a grant, and how much is a loan. The College Board web site quotes Mariko Gomez, Director, Student Financial Aid, Southwest Texas State University, who says, "When comparing aid offers,  look at both quantity and quality. Figure out quantity by adding up how much each college is giving you and what you must pay. Figure out quality by comparing how much gift aid to the amount of loan and job." American Student Assistance, a non-profit organization, has good advice here, as well as a form you can use to compare awards that offer different combinations of awards.


In addition to merit aid the college provides, your students might be eligible for one or more privately available scholarships. Scholarships, of course, do not have to be repaid. But that means it’s all the more important to be honest and straightforward in the application. Don’t apply, or allow your student to apply, unless he or she is truly qualified. Find out about any scholarships your state offers – most states do. Check the membership organizations and employers of parents and of your student, if she has one. Often colleges or universities provide financial aid to the children of their employees. Then think of the national awards, like ROTC or the National Merit Scholarship. Remember, a private scholarship might reduce your financial aid package.

Use a free service to help you search.  Some free scholarship search services are :Scholarship Search --'s scholarship search tool, and

Be cautious!  If you have to pay for information, or are asked to pay to apply for or obtain a scholarship, it’s probably a scam, so be wary.  It should cost NOTHING to apply for federal – or any – student aid.

Facebook and Big Data

Today's column by Nick Bilton in the Bits blog of the New York Times is a good reminder of the power of huge amounts of data. Harness it, and it can give you a sense of what your users want. But it's also a reminder that every time you click on 'like' or 'share,' well, you're giving Facebook (or Microsoft of Google or numerous others) yet another piece of information about what you like, use, or want.

Friday, May 11, 2012 and climate change

Update 1:50 pm: Here's a link to Derek Thompson's blog on, with charts from Hansen's paper showing areas where temperatures are more than 3 standard deviations above recorded normals. There's no question that it's human caused global warming.

Yesterday's New York Times had an Op-Ed by James Hansen, director of the NASA Goddard Institute for Space Studies, arguing that by using tar shale, or tar sands, for fuel, instead of finding other, less damaging energy sources, we make what is now a risk of of keeping earth's carbon concentrations below 550 ppm (parts per million) a certainty. (Hansen addresses tar shale in the US and Canada, though there are also tar sands in Venezuela.) It's a depressing prediction, and one worth thinking about.

As Hansen points out, we are already seeing extreme weather events caused by our heating planet. The photo at the top, from Climate Central, shows waterspouts (essentially tornadoes over water) in the Gulf of Mexico off of Louisiana on May 9th, 2012. You can see more at the Climate Central's website, here.

But there are some grounds for at least minimal hope. The author Bill McKibben and a team of 'university friends' has started a website called intended to build grassroots understanding of the need to reduce the carbon concentration to, you guessed it, 350ppm. Here's their introductory video:

And I know it's behind a paywall, but this New Yorker article by David Owen about a search for an artificial photosynthesis, is interesting, first because of the science (an artificial leaf) and second because of the approach (limiting the increase in energy needs of developing countries while bringing up the standard of living, instead of focusing only decreasing demand in developed countries).

Thursday, May 10, 2012

Obama on gay marriage

OK, I admit this isn't about numbers in the sense I usually use. But it is about a politician making a brave and important statement (some people are suggesting that it's in defiance of numbers, those in Tuesday's vote in North Carolina). Here's President Obama speaking out in support of gay marriage:

Any thoughts? Let me know in the comments.

Popular Posts