Census, PSU still apart on Portland population
The U.S. Census Bureau's estimate of the City of Portland's population as of July 1, 2009 has been released. The feds place the population within the city limits at 566,141, a 1.75% increase above the estimate they gave for the year immediately before.
The federal numbers are still noticeably below the figure of 582,130, which the experts at Portland State show as the city's population as of July 1, 2009. The Portland State figure is 2.8% higher (15,989 people) than the federal census number for the same date.
We'll get another Portland State figure this fall, estimating the July 1, 2010 population. The results of the far more extensive federal census effort that's currently wrapping up won't be released until next spring.
In our City of Portland debt clock, we'll continue to use the Portland State figure, and a growth rate of 1.28% a year. But if the feds are smarter than Portland State, the population is actually a bit lower, and the debt per resident is a bit higher, than is shown there.
Comments (12)
What do you expect from a university that is actually a mediocre community college offering a bunch of ethnic and gender study degrees instead of doing real research and offering useful programs.
Posted by John Benton | June 22, 2010 5:00 PM
Now, now. They've got shiny high-rises! And a streetcar! And a sustainability center! Linchpin city.
Posted by Jack Bog | June 22, 2010 5:02 PM
According to the Census, between 2000 and 2009, the city grew from 529,921 to 566,141. That's a compound annual growth rate of only 0.74%.
Posted by Jack Bog | June 22, 2010 5:08 PM
C'mon Jack you know a billion people are moving here and we need more infill, condos, streetcars, MAX lines
Go by hackeysack!
Posted by Mike Hamilton | June 22, 2010 5:16 PM
My guess is that we will see a noticeable re-jiggering of population numbers with the 2010 census. And, we will see that Portland's growth rate will be lower than planning overlords have been touting.
Then again, we may find that medical marijuana cards may be the economic growth engine fueling the rise of the "creative class."
Posted by Garage Wine | June 22, 2010 5:17 PM
The census regularly under counts urban populations. PSU uses a broader array of indicators like housing purchases, employment, etc. in their counts.
The Federal effort might be more extensive, but it's not necessarily more accurate. The Republicans have killed several attempts to use statistical sampling to help make sure it was more accurate.
http://online.wsj.com/article/SB124241977657124963.html
Oh, and John, they offer Engineering and Computer Science graduate programs at PSU along with ethnic and gender studies.
Posted by AMA | June 22, 2010 5:25 PM
Jack Benton,
You forgot to mention it is a "commuter school," but never throw that term around in Admissions and Registration because they want everyone to think otherwise.
I would say maybe 40% to 50% of Portland State's students live in Portland. After 4 to 7 years of getting a Bachelor's, I estimate only a quarter stay in Portland, OR with a good career track job or as part of the silent, majority service class working odd jobs as a barista, waiter, and whatnot.
Portland State University is atrocious at designing course curriculum that matches what real world private sector employers want. This spills over into job placement and internships where they place you with a local nonprofit who has no intention of hiring you and wants you as an intern to do unpaid work that a line worker would normally do.
I know from experience, I graduated with a Master's from PSU before the age of 25. I do have my regrets, PSU is one them, but hindsight like criticism most of the time is an unproductive venture best reserved over a pitcher of beer.
Posted by Ryan Voluntad | June 22, 2010 11:12 PM
Ryan, good read on PSU. I will say from watching a friend in a master's program there that she was in a forced unpaid internship with an agency (and she was not the only one of her classmates required to do that) that her professor was a hired consultant for ... so in fact it appears faculty got paid for delivering unpaid labor in some cases...
Posted by LucsAdvo | June 23, 2010 5:56 AM
The two population figures are never the same. For this one, though, I'd go with PSU's figure. Also, note this bit from your link:
The Census Bureau said the new estimates are not 2010 census population counts but are the last estimates to use 2000 census results as a base.
The Census number is a not based on current counts.
Posted by the other white meat | June 23, 2010 8:48 AM
Neither is PSU's. Not so long ago they showed Portland's population not growing at all, because the city hadn't returned a questionnaire.
They're both estimating and extrapolating, and the question is which one is doing a better job of it. And if I am not mistaken, when the 10-year Census numbers come out, PSU uses them as its base.
Posted by Jack Bog | June 23, 2010 3:12 PM
The census regularly under counts urban populations. PSU uses a broader array of indicators like housing purchases, employment, etc. in their counts.
It seems that a census would miss very poor and transient urban dwellers living in slums. Portland has areas where property values are lower, but in my limited visits to Stumptown, I can't remember seeing areas that would fit into the "slum" category. Thus, Portland would seem to fare better than many big cities when census counts are taken.
The Federal effort might be more extensive, but it's not necessarily more accurate. The Republicans have killed several attempts to use statistical sampling to help make sure it was more accurate.
A "census" and "statistical sampling" are mutually exclusive. A "census" means that you're counting (or polling) the entire population (which is what our decennial effort tries to do). Statistical sampling is a way of inferring what the population is (does, thinks) based on polling of a representative sample. If you're doing a census, there's no point in statistical sampling and vice versa.
For example, if I want to know how people in a city with a population of 1,000 will vote in its mayoral race, I can take a census and ask all 1,000 of them. Or, I can do statistical sampling by taking a true random sample (which is harder to do than it seems) of a small part (say, 20-50) based on statistical formulae. (In fact, from a sample size of about 400, you can draw statistically valid inferences about the behavior of very large populations.)
Posted by rural resident | June 23, 2010 7:41 PM
Technically, *no* count is based on "current counts". It's not really possible. But PSUs is based on more current (and actual) counts.
A "census" means that you're counting (or polling) the entire population (which is what our decennial effort tries to do).
Actually, the US Census does not "count the entire population". It hasn't for quite a while. It uses statistical/probabilistic inference to come up with much of its results. Look it up.
Posted by the other white meat | June 24, 2010 2:42 PM