The United States has, for the third decade in a row, attempted and failed to create a method by which census data can be collected by some means other than pencil and paper.
It’s not that this would be so bad except that the Census Bureau estimates that the cost of this handheld system is $1.3 billion, and it won’t do what was intended: Allow canvassing workers to enter and transmit census data digitally. It was to be the first truly high-tech census. Plans for the 1990 census called for quite a bit of minicomputer technology to support data collection, and in 2000 the use of data capture systems; those plans didn’t materialize, either.
This article is really discouraging. It further reinforces my belief that agencies like the Census Bureau are loaded with bumbling political appointees being gifted large salaries and prestigious titles for their support of the party in power. The Bureau’s bumbling appointees were Jay Waite and Louis Kincannon. Their CV’s are impressive, their results are not. When asked in 2006 by Senator Tom Koburn what the Bureau would do if the handheld units did not work, Kincannon stated that the handhelds would work, implying there was no need for a contingency plan. He repeated this statement four more times during the Senate hearing, indicating that no plans were in place should the system of handheld units fail to materialize by 2010. Kincannon was replaced in late 2007 by Steven Murdock. Read the entire story here; it’s fascinating.
Fed Ex seems to do a pretty good job using handheld scanners to track millions of packages daily. My wife uses Nielsen SCANTRAK to enter all kinds of data about her weekly purchases, transmits that data on a regular basis and gets awarded points that she can turn into gifts for all her effort. Turbo Tax turns my digitally-created 1040 forms into 2-dimensional bar codes that can be scanned at the receiving end in a matter of seconds.
The technology to do everything the Census Bureau needs has been around for years – commercially, about 15 years already – yet the Bureau cannot figure out how to turn that already-available commercial technology into something that would work for them.
What a shame. What a statement of utter incompetence.
We will pay $1.3 million for a system that will not provide census workers with the tools they need to efficiently do their jobs. We will, instead, pay those 600,000 workers to arm themselves with pencil and paper to collect data from the expected one-third of all Americans who will not return their census forms, only to have that data re-entered by other census workers sitting behind banks of PCs once the paper forms arrive back at headquarters. Total cost for the census: Around $14 billion and growing.
But it’s not entirely a lost cause. Even though the units won’t be able to collect and transmit data, the 151,000 handheld units that will be purchased (note that apparently, not every census worker gets one of these toys) will be able to use GPS to verify the location of every home in America, just like a Tom Tom. An $8,600 Tom Tom.
Maybe by 2020 we’ll decide to build something high-tech that actually works, or make the potentially wise decision to simply continue doing it with pencil and paper. The alternative – the worst alternative – we seem to choose is to pay some company billions for a half-assed, poorly thought-out design that has no chance of getting off the ground.
How sad for us.