Magnifisyncopathological

[Updates below.]

About 90 minutes ago, "Jan" from the U.S. Census Bureau knocked on my door. She handed me a flyer announcing the 2006 Census Test. The bilingual flyer isn't was concerns me. What concerns me is that she asked to take a GPS reading of "the exact location of your front door."

I asked her why and she didn't have an answer. She did, however, express serious worry about the government's desire to collect this information and what might motivate it. I asked her why, if she was so concerned, didn't she quit the job and do something else. She replied that she needed to pay her rent like everyone else and was planning on starting her own business. I left the opening alone and we instead talked about "alternative media" such as Alex Jones and his documentaries, 91.7FM KOOP/KVRX, Radio Free Austin, and so forth. If she was putting on a performance for me, it was convincing. She mentioned subjects that are mentioned on GCN and RBN such as government manipulation of viruses and food supplies.

She also brought up the feelings of some that arming ourselves might be a good thing in case something happens. She thought the danger came from plans within the government to crack down on independent voices and dissenters. I told her I thought that was a worry worth keeping in mind, but an economic crack-up was more likely. Nowhere in this conversation did I venture information beyond my first name, my opinion of political events, where I stood. She reacted a tad shocked when I said I'd rather not have government period, rather than reform it.

When I mentioned that having this locational data on file might be of use to federal agents stationed in Austin, specifically mentioning the IRS, she said she worked for them for a year and hated it, hated the IRS itself for what she saw.

I told her that I did not want a precise reading at the foot of my door. I also told her I wasn't going to forcibly stop her from taking a reading off my property. But I did tell her if was going to save the latitude and longitude of my address, I wanted it done out in the street. I'm well aware that the more accurate GPS technology the government uses has more resolution and a higher degree of accuracy and both the error of guided munitions and their blast radii make any potential effort to push the aiming point 25 feet or so from my house somewhat futile, but I absolutely did not want doorstop accuracy on file with the federal government.

Searching the Census Bureau website questions box for "2006 census test" gets you nothing. Searching for "GPS" gets you the same. Googling for "2006 census test" gets you a few hits. One of them, 2006 Census Test Operational Photos has this to say:

The 2006 Census Test is being conducted in portions of Travis county, Tex. and the Cheyenne River Sioux Reservation, S.D. The 2006 Census Test will help the U.S. Census Bureau evaluate new ways to count the population and plan a more accurate and cost-effective 2010 Census.

On that page's list of pictures, I recognize two: the badge and the PDA that Jan had while talking to me. She did not have the tote bag and was not dressed in any formal attire.

A hit on the Dallas Regional Office has this to say:

The U.S. Census Bureau is conducting a 2006 Census Test in Travis County, Texas, including parts of the city of Austin and its suburbs to test new technology, methods, and procedures for implementation for the 2010 Census.

Posted is a Fact Sheet (PDF), a test area map (PDF), and the following call for employees:

EMPLOYMENT OPPORTUNITIES

CENSUS TAKERS NEEDED
$14.50 PER HOUR, PLUS MILEAGE

CALL NOW 1-888-814-6711

I do indeed live in the test area, along with a sizable chunk of the people living in Austin. The western portion of the Austin metroplex is left out. I want to imply nothing by noting this, but that area contains the wealthiest citizens, people from places like Westlake.

I don't have time to dig into this further, but I will when I get back home in a few hours.

UPDATED 8/1/2005 10:11am
When I first posted this, I didn't read the Fact Sheet. I wanted to sit down and look at it yesterday, but that didn't happen until this morning. Here's what it says:

The Census Bureau is testing:

• How to improve methods for collecting data from respondents who speak a language other than English, including bilingual questionnaire delivery.
  
• How to improve data collection procedures that will enhance overall data quality for people, housing units and group quarters.
  
• How effective is the use of hand held computers with Global Positioning System technology for field data collection and support operations.
  
• The impact of different replacement questionnaire-delivery strategies.
  
• How to improve data collection methods in group quarters, dormitories, correctional facilities, nursing homes, and unconventional housing.

The language is benign and the goals seem reasonable.

The conversation I described above with Jan the Census Enumerator reads oddly, but that is how it went. I didn't get a "pitch" or a scripted statement. When she announced who she was and what she was doing, I expected the kind of hyper-bland, nonconfrontational, multicultural tone I've come to expect from the state's agents, especially when directly interacting with hundreds of members of the public. There wasn't a trace of that to be found with her; she backtracked almost immediately, acting more embarassed than bored.

I've sent the following message to the Dallas Regional Office public e-mail address and the Press Inquiries e-mail address for the main Census Bureau website:

TO: dallas.regional.office@census.gov
CC: pio@census.gov
Subject: 2006 Census Test & GPS

My name is Charles Hueter, I live in Travis County, TX, and I was contacted by a Census Bureau employee over the weekend regarding the '06 Test. She asked permission to take a Global Positioning System coordinate reading as close as possible to my front door.

I would like to know why such a specific measurement is necessary. Other than for simply testing equipment, for what purposes would the Bureau use these latitude and longitude readings?

Thank you for your time.

-Charles Hueter

If I hear anything, I'll post it.

According to the USAID website, "East Timor, the world's youngest country, is also the first country to successfully make comprehensive use of global positioning system (GPS) technology for its census." The date of that release is November 16, 2004. Might the Census Bureau be thinking about doing the same for the US?

Actually, yes: Census Bureau eyes GPS for 2010 U.S. effort

The U.S. Bureau of Census plans to equip half a million temporary field workers with GPS-capable handheld computers that will underpin the agency's efforts to re-engineer its data collection and processing operations for the 2010 effort. The field data collection automation (FDCA) initiative aims at reducing the paperwork-driven costs of staff and office space required to compile census-takers' reports using traditional methods.

More than 60 million addresses and field assignments will be handled out of 500 temporary local census offices.

In preparation for the 2010 effort, the bureau is conducting a census test between June 2005 and December 2006. As part of this test, the bureau will acquire and use up to 1,651 personal digital assistant (PDA) devices with the option of incorporating GPS hardware. Field interviewers or "enumerators" will collect household-level information using a Census Bureau-developed automated questionnaire.

The PDAs will be used to collect housing unit location information using a GPS receiver and mapping software that can display and create maps using data saved in ESRI's ArcPad shapefile format.

Census officials expect the test will determine the feasibility of using the GPS-enabled PDA devices for locating and navigating to assignment areas and housing units, collecting census and payroll data, and testing new census procedures and concepts. Bids to supply the equipment closed early in August.

The GPS equipment must output NMEA-0183 (v2.20 or later) SiRF standard protocols and possibly others and track L1, L2, or L5 C/A-code continuously. The agency's FDCA solicitation sets differential GPS as the priority signal, followed by WAAS and unassisted "pure" GPS. The test's accuracy requirement is three meters or less uncorrected, although it notes that data can be differentially corrected. Finally, the receivers will take advantage of any Global Navigation Satellite System (GNSS) signals available at the time of procurement, and specifically notes Europe's Galileo system now under development.

Looking ahead to a full-scale rollout in 2010, the Census Burea has established a website with detailed background information, reports, and other documents for potential FDCA contractors: www.census.gov/procur/www/fdca.

This (PDF) paper describes a Census Bureau GPS test conducted in 1998, so this has been in the planning for some time. I wonder how many news articles I've ignored since then that mentioned this desire to nail down structural locations (I don't have a Lexis-Nexis subscription).

The above article has a date of September 2004, so when I went to the FDCA website, the only links up were those to the final Requst-for-Proposal Solicitation No. YA1323-05-RP-0009, the various draft RFPs, and a history/library of documents related to the process that led to the vendor getting selected.

One of the documents is a P54-page DF titled Scope of FDCA Program. I don't have the time to read it all the way, but I did find this on page 6:

The Census Bureau also will enhance the existing geographic database—the Topologically Integrated Geographic Encoding and Referencing System (TIGER), which contains street and map features as well as political and statistical boundaries used for data collection and tabulation of census results. The Census Bureau will improve the accuracy of the street and map features by having them in Global Positioning System (GPS) alignment. [collect GPS coordinates structures containing living quarters]

This isn't for the 2006 Test; this is for the full 2010 Decennial Census Process.

Further down on pages 29-30:

Address Canvassing (AC) is a field operation for verifying and correcting addresses for all living quarters and street features shown on the listings and maps produced from the Decennial Census extract of the MAF/TIGER. It is a dependent field operation that requires the listers to compare what is on the ground to what is on the MAF/TIGER. The field listers will verify existing living quarters, identify duplicates on the address list, add new living quarters that are not on the address list, identify duplicates on the list, and delete from the address list living quarters that do not exist on the ground. They will also capture GPS coordinates for every structure containing living quarters. The goal of this operation is to ensure the accuracy and completeness of the Census Bureau address file and spatial database used to control and facilitate the delivery of census questionnaires.

Listers will knock on every door to attempt to contact respondents to verify address information as well as inquire about any additional living quarters. They will compare the information to what is on their lists and maps and make the required changes. They will add any addresses that are on the ground but missing from the address list. They will delete any addresses that are on the list but are not found on the ground. They will make changes to existing addresses, such as changing an apartment number, if necessary. The listers will collect a map coordinate for every structure that contains a living quarter. In addition, they will classify all living quarters as “housing units” or “other living quarters.” “Other living quarters” are living quarters that are not conventional housing units, but are such places as college dormitories, prisons, skilled nursing facilities, and the like. The listers will make the appropriate changes to the maps. This will include adding new street features that are not on the maps, deleting street features that are not on the ground, and correcting street features, such as correcting the street name. All map and address updates from Address Canvassing will be used to update the MAF/TIGER database for later Census Bureau operations.

Two separate bullet points say "Collect accurate latitude and longitude using GPS for all structures (within 3 meters) and map features (within 7.6 meters)".

This corresponds with the tiny bit of officalspeak Jan said to me that day. She asked if I was the owner of the house, if I lived there, if it was a multi-family dwelling or a single home, etc.

I remain suspicious. Even though it looks as if the Bureau has been collecting this info for years, I feel considerably wary of these kinds of government databases.

More later if something arises.

UPDATED 4:05pm
Before I forget, I ought to remind you that the Census Bureau is not above making their surveys mandatory. From the American Community Survey website:

The American Community Survey is conducted under the authority of Title 13, United States Code, Sections 141 and 193, and response is mandatory. According to Section 221, persons who do not respond shall be fined not more than $100. Title 18 U.S.C. Section 3571 and Section 3559, in effect amends Title 13 U.S.C. Section 221 by changing the fine for anyone over 18 years old who refuses or willfully neglects to complete the questionnaire or answer questions posed by census takers from a fine of not more than $100 to not more than $5,000.

I have not heard back from the e-mail I sent.

UPDATED 8/16/2005 5:00pm
I'm still worried about this, even after more than two weeks of sitting on the idea and considering it; I haven't heard back from the organizations I e-mailed, either. I'll try again and include a few other politicos as well to see if I can stir something up.

UPDATED 3/29/2006 12:06am
Fuck the Census Bureau and Their 2006 Census Test

UPDATED 8/3/2006 1:07pm
Via Billy Beck, I learn that NPR is talking about this: Census Bureau Adopts GPS to Find American Homes:

Two-and-a-half years from now, in early 2009, the Census Bureau plans to send an army of 100,000 temporary workers down every street and dusty, dirt road in America. They will be armed with handheld GPS devices.

Robert LaMacchia, head of the Census Bureau's geography division, says they'll capture the latitude and longitude of the front door of every house, apartment and improvised shelter they find.

"We will actually knock on doors and look for hidden housing units," he says. "We will find converted garages; from the outside, it may not look like anybody lives there."

But census workers will add each dwelling, legal or not, to the Census Bureau's Master Address File.

UPDATED 9/22/2006 10:02am
The Law Protects Your Privacy!