Saturday, June 26, 2010

All I want for Census is my two cent's worth...

How you mark time says something about what kind of person you are. If you're a political junkie, election years make a big dent in your consciousness. If you're a sports nut, maybe you remember events as being in the same year as the Olympics. These regular events can be the metronome of your life, ticking by multi-year intervals on a frequency to which only you, and those who share your obsessions, are attuned.

For me, census years are a big deal. As a professional planner and amateur spreadsheet nerd, the collection of detailed statistical data every five years is eagerly awaited. Unfortunately, in the case of the Canadian Census, the processing and error-checking takes awhile, and fine-grained data is only published several years after being collected. I remember the angst of doing my thesis in late 2003, knowing that the statistical data I was using was from the 1996 census, knowing that the 2001 data would become available literally the same week my thesis was due. My research would be based on data at the very end of its shelf life, obsolete the moment it was completed.

In any case, when the census does come around, it produces a fresh(er) data set from the last one. Our picture of the country moves forward five years. It's like time-lapse photography. Look at three or four censuses (censi?) and a story starts to emerge.

Some of these stories need to be told. When I looked at rental costs in Ottawa, digging up this info from as far back as 1951, I was shocked. Rent, even adjusted for inflation, was a small fraction of what it is now. ("Now", at the time, being 1996--even before the massive gentrification and housing bubble of the early 21st century!) Shocking, yes, but not surprising. It confirmed yet another suspicion harboured by most people my age and younger, which is that when it comes to basic necessities, the last generation had things easy.

(I wish I had better pictures than this, but the graphics above--pulled from my Word copy of my 2002 GIS project--is a thematic map of rents in the Lowertown district of Ottawa. The numbers, and thus the colours based on them, are indexed to 1996 dollars. Darker colours represent higher values--a lot higher.)

All census nerds have their favourite sections. For me, it's the travel-to-work data--seeing how many people drive to work, ride as passengers, take transit, walk or bike. This kind of data, especially at the DA level of geography--the smallest unit published, generally a few city blocks--is invaluable in proving a key point in city planning: namely, you don't need to build everything with giant parking lots because in a lot of areas, many people don't drive. (Rents being what they are, who can afford a car?)

Conversely, some really interesting and useful questions are not asked on the census. Now, I know there are limits to how long the questionnaire can be. Every field is going to have its wish list and if everyone got their way, the questionnaire would be a hundred pages long. That said, though, there is an extraordinary amount of space in the published census devoted to questions of... well, questionable relevance or practical value. For instance, of the eight or so CSV files that make up the published census, one is devoted almost entirely to languages spoken in the home. I'm not talking about just English or French, or even the major languages of the world. There is row after row after row telling us how many people speak Aleutian or Hakka.

Now look--I'm all for multiculturalism, and I don't think people should be excluded or overlooked because they don't look or talk like me. But honestly--what possible policy direction could come from knowing how many people in a district speak Hakka, especially since numbers are rounded off to the nearest five and therefore, in the vast majority of DA's, round off to zero?

It seems to me that if there's room on the census for questions like this, they can fit in a few from my wish list. Here it is:

1. How many automobiles are owned/leased by your household? This is a key data point that would let me work some magic. I've lost track of the number of times I've heard planning commissioners or developers refuse to consider development--even right downtown, in walkable areas next to high-level transit service--without suburban levels of parking. Asking the question "how many cars" will throw into stark relief the difference between suburbia and the city--and therefore the amount of space we need to be devoting to our cars.

2. How much does your household spend on transportation per month? Cars, for most people, are their second-biggest household cost, amounting (last time I looked ) to something like $7000-$8000 a year on average for one car. We routinely ask how much people pay in major housing payments or rent. One of the things I always hear about the suburbs is that it's cheaper to live there. That may be true if you're comparing the cost of a single-family house in Orleans versus one in Westboro. But when you compare the transportation options in the burbs versus the city, and count the total [housing + transportation] costs, I suspect the difference evaporates pretty quickly. When you consider that owning a car costs $500-$700 a month, suddenly saving $300 on housing payments by living far out instead of downtown doesn't look like such a good deal. By pretending that transportation is not an inherent cost of living where you live, we disguise a relationship that must be understood if we are to make intelligent decisions about how we live.

Saturday, June 5, 2010

New rule.

Okay, new rule:

Anyone who wants to criticize how Barack Obama is handling the oil spill in the Gulf of Mexico must first write a 1000-word essay on what President John McCain would have done differently.