Comment of the day: The trouble with projections
Posted 7/28/2008 by Joseph AskinsLincoln Square/Ravenswood and nearby Andersonville share many characteristics(see above) that make them great places to live. Two of my favorite neighborhoods. Pick Andersonville up and move it east closer to H2O and it would be difficult for me to picture a better hood. Of course the rest of Edgewater nearer the lake is improving and has some of the better characteristics of its Swedish named neighbor.
As for your reference to the 2005 mid decade census data I suggest reading that with a big clog full of salt. The track record for mid decade census data is only slightly better than George W. Bush's track record as a peacemaker or my record as a husband. Two wars and counting for both of us now that I think about it.
In 1995, and up until the 2000 census data actually came out, those who relied on it predicted a significant population loss for Chicago as a whole and also largely missed some significant other changes citywide.
From my BOG perspective, boots on the ground, Andersonville is NOT becoming more hispanic. Actually the exact opposite is happening. From my meager study of the overall data the authors seem to predict rising hispanic population in most city neighborhoods to include the neighborhoods north of downtown and near or just west of the lake. It seems to me they appear to be using some city wide number/factor to predict rising hispanic population in most hoods.
It's almost as if they say "let's add x percent" everywhere plus or minus. While it would not surprise me if the overall city hispanic population continues to grow., likely at a much slower pace than in the recent past, the growth will come in areas on the NW and SW sides. Like other groups that came before hispanics are now making their way to the burbs or never living in the city in the first place.
The north lakefront and north side gentrifying neighborhoods are seeing a decreasing hispanic population. Those folks are largely either moving west or leaving the city entirely and not being replaced by others of a similar background.
- IrishPirate, on the pitfalls of relying on five- and 10-year demographics projections, especially when talking about neighborhoods like Andersonville.
If you haven't watched our Andersonville drive-around, give it a look-see. And if you have any insight about the neighborhood, feel free to chime in with your thoughts.
Comments
7/28/08
IrishPirate said:
I don't even believe that the mid decade census data is even accurate as a "snapshot" of the demographics of the neighborhoods for 2005. Their projections are even more questionable.
I could go into why I don't trust the numbers. Simply put, when I see data that claims that the number of hispanics and blacks increased in Uptown between 2000 and 2005 while the number of melanin challenged white folks decreased I realized that somewhere their methodology is screwed up. They made similar projections in 1995 and were way freakin' off. The Bush administration must have hired these folks to "project" weapons of mass destruction in Iraq prior to 2003.
The number of black folks has probably declined slightly, but there is no doubt that the number of hispanics has dropped dramatically. The changes can be seen in the local schools, shops, and streets. It wouldn't surprise me if the actual 2010 census shows the percentage of hispanics in Uptown at around 5 percent or less. The Buena Park section of Uptown used to be plurality hispanic and now outside the corner at Broadway and Cuyler there is no significant hispanic presence in the southern part of the neighborhood. Something similar, but less dramatic has happened throughout the neighborhood.
Anyway before I get accused of being the large headed son of the late demographer Pierre DeVise I will move on.
Interesting article from the New Republic.
The article uses Chicago as the main example of change.
Carter said:
Take it from someone who worked for the census (1990) - there's a lot of good reasons not to trust the data. I was both a test-giver for new/potential employees, and then later a "renumerator," meaning I double checked random samples of human-collected data.
More problems than I could shake a stick at - I'm talking non-existent addresses (and businesses, and waterhouses, and…) with fully completed data - obviously completely fraudulent info entered by some slacker, likely in the comfort of his/her own home. I heard first hand from supervisors how woeful the workers were.
Then there's the info that came from the Census - the buildings with horribly incorrect entries (ex: listed with apts 1 - 50, then again with letters for the units, sometimes combos of both), relying on said temp workers to catch the mistakes.
I can't even imagine how many people I might have gotten canned (or at least had their data removed) if there was any follow-up, but that's the problem - given the timing, there didn't appear any such real corrective action. this is no Attorney General audit here, if you get my drift.
pirate might enjoy the fact that I and my fellow rabble rousers in the second stage (renumerators) got our training at his favorite L stop at Wilson & Broadway at Roosevelt.
I'd imagine, to be fair, that on the whole the data must have a relatively (not sure what might mean, 80% accurate, perhaps) good accuracy rate. Relative to the tales told by the permanent "riders" on the CTA buses, anyway.
7/29/08
UptownR said:
Well, if the actual Census is 80% correct, then the projections are even worse.
By the way, you meant "Truman" not "Roosevelt", right Carter?
Carter said:
yep, my fault. (memory lapse is somewhat understandable, given I haven't been back there since).
and exactly, the "legit" Census every decade, at least in Chicago, in my experience had holes you can drive a truck through, so I can't even fathom how they would conduct a more half-assed version of it.
IrishPirate said:
UptownR,
please check your BPN email. I sent you a question regarding "da hood" that you may have more insight into than me.
Thanks.