The 12 States of America, by Dante Chinni and James Gimpel, from “Patchwork Nation”
Over the past few months, this blog has explored a number of spatial facets of the American landscape: migration trends, re-regionalization, some new socio-demographic patterns as shown in the 2010 Census, issues of gentrification, urban inequities, megaregions, the national Twitter, criminal justice spending, and a host of others topics, including a number of historical-geographical look-backs on our recent and not-so-recent pasts, and how they inform our current lived experiences.
Continuing in this theme of putting the U.S. under the “spatial microscope” (giving props to Andrew M. for turning me on to this phrase), this posting introduces the reality of the segmentation of America. Starting off with looking at how fragmented (and unevenly distributed) our country is in terms of basic economic indicators, we move on to how this fragmentation is used by marketers/advertisers/corporations to categorize you, pigeon-hole you, to better enable them to sell you stuff (market segmentation), or to deny you stuff. And how, in general, our lives are being monitored and tracked to a degree never before imagined outside of 1984, and science fiction (and this if often with our own tacit or implicit permission! We are complicit in our own surveillance!). Well, it’s perhaps not the cheeriest of postings. But it is pretty interesting!
The 12 States of AmericaThe authors, Dante Chinni and James Gimpel, who also wrote the book associated with this work, “
Patchwork Nation,”
http://www.patchworknation.org/ say they assembled thousands of data points to categorize each county in the U.S. by one of 12 different themes, which presumably describes each specific county better than any one of the other categories.
The themes are:
§ Boom towns
§ Campuses and careers
§ Emptying nests
§ Evangelical epicenters
§ Immigration nation
§ Industrial metropolis
§ Military bastions
§ Minority central
§ Monied 'burbs
§ Mormon outposts
§ Service worker centers
§ Tractor country
This process of assigning socio-demographic profiles to counties is fraught with problems, as commenters on The Atlantic blog have duly noted. First of all, it reduces a heterogeneous and multi-faceted county to a uni-dimensional descriptor. Many counties would rightfully be able to lay claim to being in several of these categories, and despite the fact that the authors say they had a method for ascertaining which was the most appropriate category, it still feels a bit arbitrary and unsatisfactory. If you click on the category name, they do give some definitions and demographic graphs, fleshing out their classification process a little more. For instance, “service worker centers” are defined as “Midsize and small towns with economies fueled by hotels, stores and restaurants and lower-than-average median household income by county.” According to this, the service worker centers have a predominantly white population, with a lower household income than the country average, and place very high on the “Hardship Index.” Whereas the “Immigration Nation” category means “Communities with large Latino populations and lower-than-average incomes, typically clustered in the South and Southwest.”
Another downside to the 12 States approach is that some of the categories are so broken up geographically that there aren’t contiguous groups of counties formed but merely dots of categories scattered here and there throughout the landscape, which doesn’t appear very helpful in obtaining a snapshot of who and where we are these days. So rather than 12 contiguous “states,” as the geographical way we normally think of states, we get more like hundreds of city-states and vast hinterlands. The most problematic categories in terms of excessive geographical fragmentation seem to be Industrial Metropolis, Monied ‘Burbs, and Military Bastions. Only Minority Central, Tractor Country, and Service Worker Centers seemed to have any real geographic cohesiveness, and even those were pretty spotty.
Their website, however, is quite interesting. You can plug in your zip code and see how your county is classified. I tried some zipcodes around the country that I knew off the top of my head, and I have to say, the matching-up of zip code to descriptor was not always very accurate, in my estimation. Their methodology is explained fully in their book, but basically you have to take them at their word that they tried to pick the most definitive category based on their criteria, since each county could be assigned only one descriptor.
There are some cool datasets that allow you to map, among other things, numbers of Walmarts per county (an indication of blue-collar-ness?), counties containing a Cracker Barrel restaurant (presumably some metric of “redneck-edness”?), gun dealers per 100K pop (scary-ass-place-ness?), percent attending church regularly (the high in the data range is only 20%???), and various voting patterns.
The Federal spending per capita is an interesting one. The category of over $80,000 per year per capita occurs most frequently in the middle of the farm belt, and doesn’t seem to be an urban phenomena at all. Certain counties in North Dakota, South Dakota, and Nebraska, isolated other spots in Texas and Louisiana seem to top the charts for Federal spending per capita at well over $80K per person. The Bronx, by comparison, was $13K, Queens is 9K, and NY County (Manhattan) is 15K. Poor Brooklyn is only 6K! For some odd reason, Cape May, NJ has over 120K of Federal dollars coming in per person! Dare County, NC (which is listed as a “boom town” category) gets 122K per capita in federal spending. What are they doing down there? No wonder it’s a boom town! In fact a lot of the coastal counties in the south seem to be getting inordinate amounts of the Federal largesse. Could it be for hurricane protection or something? I can’t imagine. But even Orleans parish itself where New Orleans is located, doesn’t see close to these amounts. Maybe it's for defense? A disproprotionate number of elderly live in these counties?
So the data is fun to map and toggle around with, but the authors commit the cardinal sins of choroplething absolute numbers, and also overlaying two choropleth maps on top of one another, in order to visualize multivariable data (NOT!!!). Oh, well.
They also break the country down in a more geographical sense by “districts,” and here they have 9 categories, and these appear to have more geographical cohesiveness. They also bear some conceptual resemblance to the regions in Joel Garreau’s book Nine Nations of North America:
§ Established wealth,
§ The shifting middle
§ Booming growth
§ New diversity
§ Young exurbs
§ Old diversity
§ Wired and educated
§ Christian conservative
§ Small town America
Socio-Demographic Profiling for Marketing:
I have a general problem with these socio-demographic profiles, right from the get-go.
Years ago, when firms like Claritas started segmenting markets into categories of supposedly like-minded individuals for the purpose of pinpoint targeting for advertising/marketing/purchasing, I used to cringe.
Now it has become a common place with many such firms entering the fray.
Their categories are much more extensive and fine-tuned than the
Patchwork Nation ones, and include such distinctions as globetrotters, business class, golden agers, power couples, technovators, up-and-comers, middleburg managers, multi-culti mosaic, park bench seniors, money and brains, pools and patios, bohemian mix, upper crust, country squires, urban achievers, new homesteaders, big sky families, and so forth.
They have a number of different classification systems, one of which seems to be all about income and leisure activities and spending, and one is all about stage in the life cycle and lifestyle/housing/community/geographical choice and situation.
You can see all the different classification schemes and categories on the Claritas website at
http://www.claritas.com/MyBestSegments/Default.jsp?ID=30&id1=2000
Claritas has one scheme focusing on media connectivity, which has categories such as Antenna land, plugged-in families, satellite seniors, old time media, the unconnected, land line living, generation wifi, cyber-sophisticates, new kids on the grid, low speed boomers, analoggers, cinemaniacs, dish country, cyber strivers, internet hinterland, IM nation, kids and keyboards, and so forth. I don’t know about you, but I hate the idea of the unique and messy complexity of “me” being reduced to a uni-dimensional category, (what would I be in their scheme of things, anyway? Digital dreamer? Grey power?). But most of all, I hate that “they” know so much about me that they are able to fit me (or shoehorn me) into one of their categories. It reminds me of that scene in the film Minority Reports (2002) where Tom Cruise walks down the street and all the billboards and advertisements talk to him directly, because they can respond to exactly “who” he is according to the interactive information the advertisements can glean from his unique ID (in that case, a retinal scan). People, we are not far away from that at all, in fact the technology exists and is used in a limited fashion.
This potential for privacy infringement and Big Brother-like surveillance is actually one of the more objectionable things about the universality of GPS enabled devices, mapping and tracking capabilities.
We have recently learned how dangerous it is to post digital photos on Facebook and other social networking sites because embedded within them is a geographical marker of exactly where the photo was taken, leading virtually anybody right to your doorstep.
See short ABC news story “TMI – Picture Privacy” (Thanks for the link to this story, Kristen.)
http://abclocal.go.com/wabc/video?id=7621105
From GIS for the Urban Environment: “Geodemographic data now can pinpoint the geographic location of individuals and with it, previously intimate and private information attached to those locations and individuals. Much information about an individual based solely on an address can be gathered from public or private sources. Privately held data includes buying habits, income, vacation destinations, leisure activities, telephone and computer use, religious affiliation, banking and investment accounts, educational record, and medical information. Public data includes assessed value of a home, motor vehicle information, political contributions, and criminal record. Many people clamor for full access to data until they realize what that might entail for the dissemination of private data about themselves. Where do we draw the line? What constitutes legitimate use of private data?
There is an enormous potential for abuse in the use of data about individuals and the concomitant surveillance possibilities. ‘The prospect of socioeconomic application of GIS permitting efficiently functioning organization such as insurance companies to develop ‘geodemographical’ insurance rate schedules based on the identification of zones and localities of high risk, the targeting of civil rights groups (the ‘politically militant’) for particular police or vigilante attention, or the extension of direct-mail solicitation to exact-market targeting based on recorded purchasing and general expenditure records (already a reality, of course) . . .’ is generally seen as objectionable when stated in terms of rights to privacy. All too often, however, these uses of GIS are seen as normal and neutral, as scientific uses of socioeconomic data (Pickles, Ground Truth: The Social Implications of GIS, 1995:16).
The construction of data profiles is particularly troubling in terms of privacy, since an individual can be profiled by someone utilizing individual level data, such as that available from credit card companies, governmental agencies, and so forth, and combining these with publicly-available aggregate data, such as sociodemographic data from the census or other data providers at the tract or ZIP Code level. This type of data profile may be an eerily accurate portrayal of an individual, and the potential for abuse of this profile is enormous,” (from GIS for the Urban Environment, 2006:285).
This, of course, has come true, with a vengeance. Just the other day in The New York Times, there was quite an amazing piece about a German politician in the Green Party, and his quest to find out how much the phone company tracked his whereabouts by his cell phone location. After much legal wrangling, he obtained 6 months of data pinpointing his whereabouts, with over 35,000 of his locations having been tracked (and stored!) by the phone company! Then Mr. Spitz decided to map out the locational data points to obtain a visual narrative of the phone company tracking. You can see it illustrated in a very cool interactive website, mapping where he (or more correctly, his cell phone) was during those 6 months. Interactive map of 6-months of his locational data:
http://www.zeit.de/datenschutz/malte-spitz-vorratsdaten
A visualization of data collected by Malte Spitz's mobile phone
“The data were contained in a massive Excel document. Each of the 35,831 rows of the spreadsheet represents an instance when Spitz’s mobile phone transferred information over a half-year period. Seen individually, the pieces of data are mostly inconsequential and harmless. But taken together, they provide what investigators call a profile – a clear picture of a person’s habits and preferences, and indeed, of his or her life.
This profile reveals when Spitz walked down the street, when he took a train, when he was in an airplane. It shows where he was in the cities he visited. It shows when he worked and when he slept, when he could be reached by phone and when was unavailable. It shows when he preferred to talk on his phone and when he preferred to send a text message. It shows which beer gardens he liked to visit in his free time. All in all, it reveals an entire life.
To illustrate just how much detail from someone’s life can be mined from this stored data, ZEIT ONLINE has 'augmented' Spitz’s information with records that anyone can access: the politician’s tweets and blog entries were added to the information on his movements. It is the kind of process that any good investigator would likely use to profile a person under observation,” From “Data Protection: Betrayed by our Own Data,” in Zeit Online
“In the United States, there are law enforcement and safety reasons for cellphone companies being encouraged to keep track of its customers. Both the F.B.I. and the Drug Enforcement Administration have used cellphone records to identify suspects and make arrests.
If the information is valuable to law enforcement, it could be lucrative for marketers. The major American cellphone providers declined to explain what exactly they collect and what they use it for.
Verizon, for example, declined to elaborate other than to point to its privacy policy, which includes: “Information such as call records, service usage, traffic data,” the statement in part reads, may be used for “marketing to you based on your use of the products and services you already have, subject to any restrictions required by law.”
AT&T, for example, works with a company, Sense Networks, that uses anonymous location information “to better understand aggregate human activity.” One product, CitySense, makes recommendations about local nightlife to customers who choose to participate based on their cellphone usage. (Many smartphone apps already on the market are based on location but that’s with the consent of the user and through GPS, not the cellphone company’s records.)”
From the New York Times article, March 27, 2011
It’s Tracking Your Every Move and You May Not Even Know, by Noam Cohen.
It’s a brave new world.
Just what is the “clod” and why is it cheaper? The business model is based on reducing infrastructure cost by getting computing resources for free. As explained by startup vendor Chip-On-A-Shingle, “A clod is simply a cluster of unwanted computers and peripherals all hooked together to create some really awesome processing.”
