Monday, January 28, 2013

Centenarians

There are more than 50,000 centenarians in the United States, according to the 2010 Census and a report on the 100-and-older population released last month by the Census Bureau.

Other interesting facts from the report:
  • More than 80 percent of centenarians are women.
  • The majority live in households (about one third live in nursing homes or other group quarters).
  • California has the most population age 100 and older (5,921) followed by New York (4,605), and then Florida (4,090).
  • Despite Florida's reputation as a retirement destination, South Dakota and Iowa have substantially higher proportions of centenarians as a share state population (0.0295 and 0.0278, respectively compared with Florida's 0.0218).

Image from Time magazine special section on "How to live to 100 years"


Friday, January 25, 2013

Data link roundup (week of January 25, 2013)

The week's top data analysis links...

VISUALIZING CONNECTED COMMUNITIES

By analyzing and mapping the end point locations of telephone calls, SENSEable City Lab identified closely connected communities within Great Britain and the United States. Some patterns that emerge:

  • Residents of the New York metro area do not have terribly strong connections with residents upstate
  • There is a Mason-Dixon-like line separating southern and northern California
  • Northwest Florida is much more closely connected with Alabama and Georgia than with the rest of Florida

Source: Xiaoji Chen's design blog


EQUALITY AND JUSTICE

For those who focus on issues of environmental justice and social equity, Schweitzer's piece on equality and justice is thought-provoking. The post is not about not data, per-se, but understanding the distinction between equality and justice can be crucial when developing appropriate measurements (for example) for Title VI analysis.


BEST CHART OF THE WEEK

These histograms, by Bill Rankin, showing population by latitude and longitude are excellent data visualizations: clever and informative:
Source: Bill Rankin's Radical Cartography

Friday, January 18, 2013

Data link roundup (week of January 18, 2013)

The week's top data analysis links...

APPLIED DEMOGRAPHY TOOLBOX

Have you ever had to build a median calculator from scratch? Or build a population pyramid in Excel (pesky axis formatting!)? Eddie Hunsinger's collection of links in the Applied Demography Toolbox might be a time-saver.

WHO'S CITIED?

Rafael Pereira summarizes the Most Cited Authors in Sociology. The takeaway points:

  • Foucault is cited more than four times as often as Durkheim
  • Analytical weakness: Google Scholar citation tracker only tracks registered authors
  • If you haven't registered with Google Scholar, you might want to consider it. (The metrics I found for my own citations were worth the 30 seconds it took to sign up.)


GRAYEST GENERATION

Thursday's edition of Fresh Air featured a segment on "The Grayest Generation" -- the demographic trend of delayed childbirth in the United States.
In a December article for The New Republic, "The Grayest Generation: How Older Parenthood Will Upend American Society," the magazine's science editor Judith Shulevitz points out how the growing trend toward later parenthood since 1970 coincides with a rise in neurocognitive and developmental disorders among children.
To be honest, the cognitive issues Terri Gross discussed with Shulevitz were less interesting to me than Shulevitz's sociological commentary on career path and delayed childbirth. As a whole, the interview is fascinating. (Full disclosure: Really, I'm just a fan of any headline news story that involves a detailed discussion of demographic trends...)

GETTING PUBLISHED

If you want to know the key to publication, here are two hints:

Add formulas to your abstracts and articles (even if they are irrelevant). New research in the Journal of Judgment and Decision Making shows that non-math-trained reviewers judged papers to be "of higher quality" when the abstract included a formula.

Stop using the word "paper" (also stop using the words and phrases: preliminary, seeks to, attempts, etc...). Start using the word "article" (and also consider including words: hypothesis, tests, and finds)... at least according to Scatterplot.

BEST CHART OF THE WEEK

Eric Fischer maps language communities among Twitter users.
Source: Eric Fischer



Thursday, January 17, 2013

Breach of research ethics

The Wall Street Journal published a piece by data analyst and online-dater Amy Webb.

Webb went "undercover" by posting fictitious male profiles on JDate. Webb then used those profiles to collect data on how female JDaters interacted with her "male" avatars in the online world. She later used her findings to improve her own real profile and subsequently found a mate.

While I'm sure the WSJ intended to highlight the use of analytical skills in real-life decision-making, all I could do was sit in shock at the breach of research ethics.

Has anyone heard of informed consent?

And what about the golden rule: Do no harm?

While I am sure no online daters were in any physical danger during the course of this "research," I would argue that any sort of fraud or misrepresentation in the online dating world represents the potential for real emotional harm. (The case of Manti Te'o, of Notre Dame football and online dating fraud fame being the most recent headline example. While questions abound as to whether or not Te'o was in on the hoax, plenty of online fraudsters have hurt millions. See the Warrior Eli hoax, for example...)

Anyone who spent time e-flirting with JewishDoc1000 and LawMan2346 wasted time, effort, and (quite possibly) emotion on Webb's fraud.

It is doubly insulting that Webb summarizes her findings with an admonishment that women should dumb-down their online dating profiles...
Women: Don't mention work, especially if your job is difficult to explain. You may have the most amazing career on the planet, but it can inadvertently intimidate someone looking at your profile. I realize this sounds horribly regressive, but during my experiment I found that women were attracted to men with high-profile careers, while the majority of men were turned off by powerful women.
While Webb's breach of research ethics isn't as severe as Laud Humphreys' work in the 1960s, it violates the same rules that make his work an example of "what not to do" when conducting research on human behavior.

Shame on Ms. Webb for her sham. And shame on WSJ for lauding this "research."

Tuesday, January 15, 2013

Has flu season peaked?

By the end of December nearly every state in the nation reported widespread cases of influenza.
Source: CDC
And by the first week of January, all states were reporting at least regional cases of the flu, with the vast majority reporting widespread outbreaks.
Source: CDC
However, flu season usually peaks in February or March, not in December, and while the spread of flu appears to have slowed, epidemiologists will not know for weeks whether or not the peak of flu season has passed. The slowdown may simply be part of the ebb and flow of this year's flu season.
Source: CDC

More about the data:

In addition, U.S. Centers for Disease Control and Prevention collect data from state health departments report. The first two maps below show data from the state-level reporting.

The maps are based on:
...the estimated level of geographic spread of influenza activity in their states each week through the State and Territorial Epidemiologists Reports. States report geographic spread of influenza activity as no activity, sporadic, local, regional, or widespread. These levels are defined as follows:
No Activity: No laboratory-confirmed cases of influenza and no reported increase in the number of cases of ILI.
Sporadic: Small numbers of laboratory-confirmed influenza cases or a single laboratory-confirmed influenza outbreak has been reported, but there is no increase in cases of ILI.
Local: Outbreaks of influenza or increases in ILI cases and recent laboratory-confirmed influenza in a single region of the state.
Regional: Outbreaks of influenza or increases in ILI and recent laboratory confirmed influenza in at least two but less than half the regions of the state with recent laboratory evidence of influenza in those regions.
Widespread: Outbreaks of influenza or increases in ILI cases and recent laboratory-confirmed influenza in at least half the regions of the state with recent laboratory evidence of influenza in the state.
CDC cautions, however, that the maps reflect the "geographic spread of influenza viruses, but does not measure the severity of influenza activity."

In addition, shown on map three, CDC monitors and reports on influenza through the Influenza-like Illness Surveillance Network (ILINet).
Activity levels are based on the percent of outpatient visits in a state due to ILI and are compared to the average percent of ILI visits that occur during spring and fall weeks with little or no influenza virus circulation... The map uses the proportion of outpatient visits to health care providers for influenza-like illness to measure the ILI activity level within a state. It does not, however, measure the extent of geographic spread of flu within a state. Therefore, outbreaks occurring in a single city could cause the state to display high activity levels. Data collected in ILINet may disproportionately represent certain populations within a state, and therefore, may not accurately depict the full picture of influenza activity for the whole state.  
For more information CDC also provides an interactive mapping tool, Flu View.


Friday, January 11, 2013

Data link roundup (week of January 11, 2013)

The week's top data analysis links...

MONEY CAN'T BUY LONGEVITY

Americans spend more, per capita, on healthcare, but have higher mortality rates from chronic illness (i.e. non-communicable, often preventable disease like diabetes) and injury. The U.S. also has the highest infant mortality rate among the wealthiest 17 nations analyzed by The Economist.


NOTHING GOOD HAPPENS AFTER MIDNIGHT

John Nelson analyzed traffic fatalities by day of week, time of day, and contributing factor. Perhaps unsurprisingly, mother is right when she says "nothing good happens after midnight." Fatal traffic accidents involving alcohol are highest in the early morning hours on Saturday and Sunday.
(Thanks to FlowingData for the link.)
Source: UXBlog

BEST CHART OF THE WEEK

A new, interactive tool from the UK Office for National Statistics shows the population age structure in England and Wales for every census from 1901 to the present.

Source: Office for National Statistics


Tuesday, January 8, 2013

Influenza

In case you had not noticed the coughing and sneezing of your neighbors (or the complaints of fever and chills posted by friends on Facebook), CDC confirms that flu season is upon us...

The U.S. Centers for Disease Control and Prevention monitor and report on influenza through the Influenza-like Illness Surveillance Network (ILINet). In addition, CDC staff collect data from state health departments report. The maps below show data from the state-level reporting.

Influenza activity in early December...
Source: CDC

... and at the end of December.
Source: CDC

The maps are based on:
...the estimated level of geographic spread of influenza activity in their states each week through the State and Territorial Epidemiologists Reports. States report geographic spread of influenza activity as no activity, sporadic, local, regional, or widespread. These levels are defined as follows:
No Activity: No laboratory-confirmed cases of influenza and no reported increase in the number of cases of ILI.
Sporadic: Small numbers of laboratory-confirmed influenza cases or a single laboratory-confirmed influenza outbreak has been reported, but there is no increase in cases of ILI.
Local: Outbreaks of influenza or increases in ILI cases and recent laboratory-confirmed influenza in a single region of the state.
Regional: Outbreaks of influenza or increases in ILI and recent laboratory confirmed influenza in at least two but less than half the regions of the state with recent laboratory evidence of influenza in those regions.
Widespread: Outbreaks of influenza or increases in ILI cases and recent laboratory-confirmed influenza in at least half the regions of the state with recent laboratory evidence of influenza in the state.
According to the most recent information, flu virus is active in every one of the contiguous states and in Alaska. Hawaii is reporting only sporadic flu activity.

CDC cautions, however, that the maps reflect the "geographic spread of influenza viruses, but does not measure the severity of influenza activity."

For more information CDC also provides an interactive mapping tool, Flu View.


Monday, January 7, 2013

Population Profile: Alabama

POPULATION GROWTH IN ALABAMA:

Population in 2010: 4,779,735
Population in 2000: 4,447,100
Growth rate 2000 to 2010: 7.5%
Sources: U.S. Census Bureau, Census 2010 and Census 2000

Place of Birth: Percent of resident population born in the state of Alabama:
Source: U.S. Census Bureau, American Community Survey 2007-2011

AGE STRUCTURE:

Median age in 2010: 37.9

Alabama Age Structure in 2010:
Source: U.S. Census Bureau, Census 2010 and author's calculations

PUBLIC HEALTH:

Birth rate: 12.4 per 1,000 population (compared with national rate 12.7)
Fertility rate: 61.8 per 1,000 women age 15-44 (compared with national rate 63.2)
Infant mortality rate: 8.28 per 1,000 live births (compared with national rate 6.39)
Sources: U.S. Centers for Disease Control and Preventionpreliminary birth data for 2011final death data for 2009

Life expectancy: 75.2 years (compared with national 78.6)
Source: Kaiser Family Foundation state health facts 2007

Adult obesity rate in 2011: 32.0 percent
Source: U.S. Centers for Disease Control and Prevention, Behavioral Risk Factor Surveillance System

OTHER FACTS:

Only 3.4 percent of the population in Alabama is foreign-born, compared with a national average of 12.8 percent.
Source: U.S. Census Bureau, American Community Survey 2007-2011

Despite the holiday's association with New Orleans, it is Mobile, Alabama that gets credit for hosting the first Mardi Gras celebration in North America in 1703.
Sources: CNN and History Museum of Mobile

Friday, January 4, 2013

Data link roundup (week of January 4, 2013)

The week's top data analysis links...

WHERE TO BE BORN

Back in 1988 The World in 1988 ranked nations according to the "best place to be born." The Economist Intelligence Unit crafted a more modern ranking scheme, based on a quality-of-life index, and published the results in The Economist.

The verdict: Switzerland is the place to be (born).
Source: The Economist
Best quote of the article:
Quibblers will, of course, find more holes in all this than there are in a chunk of Swiss cheese.
...but fascinating reading, nonetheless.


CORRELATION BETWEEN AGE, EDUCATION, AND BIRTH OUTCOME

Philip Cohen provides unique data to show that while pregnancy risks increase with mother's age, the relationship between educational attainment and birth outcome is much stronger.


BEST CHART OF THE WEEK

Brandon Martin-Anderson mapped every person (341,817,095) in the United States and Canada, according to the 2010 and 2011 Census counts, respectively.
Source: Brandon Martin-Anderson

The map is visually appealing, but, as an active user of small-area Census data, my favorite bit the author's response to FAQs:

Nobody lives in Central Park/Pier 12/County Lockup/Abandoned Themepark.
The census reported that someone lived there.
This says someone lives in the middle of a lake.
The census reported that someone lives in a block which includes a lake, and that's where their dot was randomly placed. Also, some people live in the middle of lakes.