Sociological Images

Billions of dollars are spent on christmas trees — real and fake — each year.  The data for 2009:

Most of the companies benefiting from this spending are small businesses:

Fake trees appear to be growing in popularity, but the sales of real trees do not appear to be slowing:

The states that benefit most from Christmas tree sales include Oregon, Michigan, North Carolina, New York, New Jersey, Ohio, Pennsylvania, and Washington, and Wisconsin:

Found at Intuit.

Dolores R. sent a link to a map created by Derek Watkins to show how the names given to geographic features reflect cultural patterns. Using a database of names officially accepted by the U.S. Board of Geographic Names, Watkins mapped generic terms used in the names of streams (excluding “creek” and “river,” which are commonly used throughout the U.S and were plotted in a gray that fades into the background):

The generic terms reflect historical immigration patterns. “Kill” appears in areas of New York originally settled by the Dutch; “cañada,” “arroyo,” and “río” indicate areas of Spanish exploration and settlement in the Southwest; of course, Louisiana and the surrounding area still reflects its French heritage through the term “bayou.”

The map reflects internal migration and cultural diffusion within the U.S., as well. For instance, Watkins suggests that the patch of red in southwest Wisconsin, indicating the use of “branch,” may be due to the lead mining boom in the early 1800s. Lead mining attracted Appalachian miners to the area, and they may have influenced local naming practices, bringing along terms common in Appalachia.

For more on the interconnections between geographic names or terms and larger cultural patterns, Watkins suggests reading Names on the Land: A Historical Account of Place-Naming in the United States, by George Stewart (2008). Another excellent source is Wisdom Sits in Places: Landscape and Language among the Western Apache, by Keith Basso (1996).

If I ran the Federal scary anti-smoking image warning program, I might show smokers the list of health-related terms that show up most in the states with the highest cigarette smoking rates.

If you take the smoking rates by state, and throw them into the Google Correlate hopper, you can see the 100 search terms that are most highly correlated with that reported smoking behavior. That is, the terms that are most likely to be used in high-smoking states and least likely to be used in the low-smoking states.

Is the result just a lot of noise? Maybe, but I don’t think so. Here are the smoking-related terms in the top 100:

• camel no 9
• cigarette coupon
• cigarette coupons
• marlboro coupons
• my time to quit
• safe cigarettes
• stopping smoking
• time to quit
• fire safe cigarettes
• ways to stop smoking

So that’s good for face validity — a list of random search terms isn’t likely to have all those smoking terms on it.

But after the smoking terms, the thing that jumps out is the health-related terms. We know from the Google flu tracker that people search for their symptoms. So these caught my eye.

Here is a screen shot of the first page of results:

I selected “stages of copd” as the term to map. The map on the left is the smoking rates; the one on the right is the relative frequency of searches for “stages of copd.” That is, chronic obstructive pulmonary disease, a nasty disease the most common cause of which is smoking.

Here is the complete list of health-related terms among the top-100 correlates with smoking rates:

• stages of copd
• abnormal mammogram [the association between smoking and breast cancer remains controversial]
• aneurysm symptoms [more common among smokers]
• hardening of the arteries [exacerbated by smoking]
• jaw swelling
• red rash on legs [could be psoriasis, which smoking increases the risk of]
• swelling in neck
• swollen lymph node in neck
• pilonidal cyst pictures

Lymph node swelling, which is implicated in the jaw and neck searches, most often reflects infection — which smoking causes.

How strong are the connections? They’re not the strongest I’ve seen on Google Correlate. The “stages of copd” search is correlated with smoking rates at .77 on a scale of 0 to 1. It’s not uncommon to find correlations of .93 (which is the relationship between “quiche” and “volvo v70 xc”).

But considering the smoking rates come from a sample survey (the National Survey on Drug Use and Health) which includes random error, and states are somewhat arbitrary geographic units, that correlation seems pretty high to me. Here’s the scatterplot:

What is the correlation causality story here? I can’t say. But the simplest explanation is that these are the terms smokers (and maybe those who know or care for them) are most likely to Google relative to non-smokers — not that they are the most common searches smokers do, of course, but the searches that differentiate them from non-smokers. The simplest explanation is the best place to start.

I like this list of conditions because in my experience smokers sometimes have the attitude of “you have to die of something.” But it’s not just the chance of dying that smoking increases — it’s a lot of possible forms of suffering along the way.

——————————

The Google Correlate tool is showing the great potential for using Internet search activity to investigate layers of behavior and meaning behind other observable social phenomena, such as race/ethnic composition, health behavior, and family patterns.

Dmitriy T.M. sent us a link to some images at the Brookings Institution, based on analysis by William Frey, illustrating the very uneven changes in average of of the population by state in the U.S. Overall, the U.S. population is aging, with rapid growth in the population over age 55 and individuals over age 45 surpassing those aged 18-44, according to the 2010 Census:

But this varies by region of the country. Here’s a map showing growth in the +45 population, illustrating the rapid growth in the Southwest and much of the South:

Nevada had the single highest growth in the 45+ population, with this group increasing by 50% between 2000 and 2010. West Virginia growth comes in last among this group (excluding Washington, D.C.), increasing by 15%. Of course, growth doesn’t tell you anything about the underlying numbers.

Many of the same states that had rapid growth in the 45+ population also saw significant gains in the under-45 range. But unlike with the 45+ population, where every state’s population was stable or growing, a significant number of states actually experienced a loss of the under-45 group:

Again, Nevada’s #1, with 28% growth. Michigan, on the other hand, had an 11% loss.

These patterns have significant implications for individual states — everything from estimating how many elementary schools they’ll need to build in the future, to how many health care workers they’ll need to educate or attract, to a state’s or region’s ability to attract different types of employers, and so on. And states will be grappling with these issues under very different circumstances. It’s one thing to, for instance, address the potential health-care needs of the elderly in a state where every age group is increasing; it’s another if your working-age population is fleeing.

Brookings has a much more detailed interactive map that includes information on aging; you can look at the dependency ratio (population under 18 or over 65 per person of working age) and look at age changes by major metro areas in addition to states.

Sociologists have shown that rates of “obesity” correlate with economic class. That is, the poorer you are the more likely it is that you will be overweight.  This is, in part, because healthy, low-calorie food tends to be more expensive that calorie rich, nutrient poor food; and also because poor neighborhoods have fewer grocery stores, forcing the poor, especially if they don’t have cars, to shop for groceries at corner stores, gas stations, and fast food restaurants.  When there are so many other things to worry about, like not going hungry, food quality is not prioritized.  Level of fitness, then, correlates with social class (and the time and money it affords you) and the things that correlate with social class, like level of education.

The American College of Sports Medicine has released data showing these correlations, if measured at the level of U.S. metro areas, as reported at The Atlantic and sent along by Tracie Hitter, a doctoral student at New Mexico State University.  First, fitness level is correlated with average income in these areas:

Second, fitness level correlates with average level of education (here called “human capital”):

And fitness level correlates with overall well-being, a measure related to both fitness and socioeconomic class:

Here are some selected metro areas plotted in relation to one another:

A new publication from the CDC, sent along by sociology professor Sangyoub Park, reports that only 13% of households in the U.S. are still cell phone-free; meanwhile, 27% of households have now abandoned their landline telephone altogether.  The data, however, varies pretty tremendously by state.  Rhode Island and New Jersey have the lowest proportion of wireless-only households at 13%, while Arkansas leads with 35%:

For more detail, here are the states in order:

Dr. Park wondered if part of what was driving the state-by-state difference was levels of poverty.  Perhaps poorer families can’t afford both a landline and a cell phone and so they drop the former.  A rough comparison of the data with rates of poverty in various states is suggestive (source):

So that’s interesting.  But why does the CDC care?  One way to collect survey data is to get a random selection of Americans (or some subset) through random digit dialings. These, however, tend to exclude cell phones.  So the technological change is creating a methodological challenge.  Now scholars using random digit dialing have to consider how the exclusion of 27% of households with cell phones only skews their data, perhaps by disproportionately excluding the poor.  It’s a much more difficult case to make than when such methods excluded only the 2% of households with no phone service at all.

For those keeping track of unemployment in the U.S., The Economist posted this graph showing job gains and losses by state, based on Bureau of Labor Statistics data:

According to the BLS, as of January 2011, the U.S. unemployment rate had dropped to 9.0%. The lowest unemployment rate is in North Dakota, at 3.8%, while the highest is still in Nevada, at 14.2%.

The BLS has detailed state-level employment data here.