RE: [CTPP] census errors

The Abilene(Texas)MPO noted similar problems when the redistricting data was first released. The entire population of two state prisons was pasted into one tiny block on the wrong side of a road in the wrong census tract and in the wrong CTPP TAZ. Two separate Assistant Division Chiefs for the Bureau of the Census told me that the error would not be fixed. One said that they were encountering problems like it "all over the country." We are now stuck with an entire special place census tract with a reported population of 2, and 4,550 institutionalized persons reportedly living in a small area which is an open field. I was told at the recent TRB conference that this error may be able to be adusted for the CTPP special tabulation, although not for any mainstream data products. Another problem occurred when the entire institutionalized population of another special place census tract was arbitrarily pasted into one small block, although not one where any of the people are actually housed. Fortunately it is still in the correct CTPP TAZ. It does, however, adversely affect the urbanized area under the proposed criteria by incorrectly showing a substantial populated area as completely unpopulated, with the result that most of it would be left outside the UA and prevent further hops. -----Original Message----- From: owner-ctpp-news(a)chrispy.net [mailto:owner-ctpp-news@chrispy.net]On Behalf Of tmank(a)tompkins-co.org Sent: Tuesday, June 05, 2001 9:42 AM To: ctpp-news(a)chrispy.net Subject: [CTPP] census dorm errors I wanted to pass this on to the listserv: I work for the Tompkins County Planning Department in Upsate, NY. When I first looked at the Block level populations this year, I noticed numerous discrepancies, which turned out to be dorm population counts which were incorrectly georeferenced. We have 2 universities here in Ithaca, NY, Cornell University and Ithaca College. In our case, the entire dorm population for Cornell (5,700 students) was put in ONE BLOCK - not in the 14 blcoks where the dorms are actually located. For example, a block that had a 1990 population of 1100 now has a 2000 pop of 2 and a block that had a 1990 pop of 110 now has a 2000 pop of 5882. For Ithaca College, the dorm population was placed all in 1 block, not in the 4 blocks where the dorms are actually located. To make things more complicated, the Cornell Block discrepancies go across municipal boundaries (from the City of Ithaca to the Town of Ithaca) and across four of our current Legislative Districts [see attachement with PDF map]. When I brought this up to the Regional Census Geography dept in Boston, they were not that surprised and said they were seeing this all over the country. The New York State Data Center in Albany says they have had this complaint all over New York State. In any case, we have noted the dorm capacities and redistributed the block populations as we think they should be and will submit for a change through the Census Bureau's Question Resolution Program, which starts in July. We also have gone ahead with our 2001 Reapportionment for Tompkins County pending the Census Bureau's approval of the changes. My questions are: 1) How many of the MPOs are experiencing the same problem? 2) I am concerned that we can estimate the Census errors and change the Block populations, but ONLY THE CENSUS Bureau knows the exact number of students who were in those dorms and their ATTRIBUTES. Only the Census can move those attributes - i.e. those 5,700 student incomes and journey-to-work responses, etc. It seems to me that all the long form data that comes out next year will be skewed - unless they deal with this issue. Which will probably only happen if the issue is as wide spread as I have been made to believe. So what are you all seeing in this regard? Tom Mank Planning Analyst Tompkins County Planning Department