I wrote a paper about this 20 years ago; I'll be happy to send it to anyone who asks. In general, I think tract boundaries should be something - usually streets but sometimes water - that local people recognize as being a divider. Freeways are the highest priority, since they're difficult to cross. Then major arterial streets/roads. I also believe, and have said many times, that *continuity is a shibboleth*. That is, it's something everything thinks is of highest importance, but it really isn't. Good tracts for now and the future are more important. I almost completely retracted Detroit before the 1980 census, because when the tracts had originally been drawn -- after the 1930 census -- no one understood their purpose, plus the city had changed enormously with new freeways and urban renewal. I also use housing units rather than population counts as a size criterion. The ideal tract has about 1500 HUs. My reason for using HU counts is twofold: (1) the data are readily available, while pop counts are not, and (2) the sampling errors for ACS data ride on cases, which are households, not people. Hope this is helpful. Patty Becker <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free. www.avast.com <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> On Mon, Mar 4, 2019 at 12:14 PM Reardon, Tim <TReardon(a)mapc.org> wrote: -- Patricia C. (Patty) Becker APB Associates/Southeast Michigan Census Council (SEMCC) *Note: new address as of 6/5/18: *283*32* Franklin Rd, Southfield, MI 48034 office: 248-354-6520 home:248-355-2428 pbecker(a)umich.edu <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free. www.avast.com <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Hello Tim, The goal of PSAP is to produce meaningful statistical geographies for data users while maintaining consistent statistical geography nationwide. The Census Bureau classifies two types of geographies: legal and statistical geography. Statistical geographies are areas defined solely for data collection, tabulation, dissemination, and analysis. That said, we recognize that statistical geographies—census tracts, especially—are used in a variety of federal programs. Though it is tempting to want to change boundaries to meet the needs of a particular program, please be aware that a change that may have a positive effect for one program may have a negative impact on another. It is better to maintain the statistical objectivity and comparability of areas, updating areas to meet the general analytical needs of as many data users as possible. Statistical Geography boundaries generally follow permanent, visible features that are identifiable in the field. Boundaries should: * Follow visible features whenever possible. * Use significant/major features on landscape. * Align to legal geography where a relationship has been established and is unlikely to change over the decade due to changes in the legal entity’s boundary. * Follow legal conjoint boundaries, where applicable. For example, given the stability of city and town boundaries in the Northeast, census tract boundaries may align with the boundaries of a city or town. With census designated places, it is preferable to align with a city boundary than to leave a small gap between the two places. * Reflect current land use patterns where possible. Boundaries should not: * Cut through housing and other buildings. * Include “zig-zags” and sharp angles for no purpose. * Group disparate land use types for no reason. * Make use of less prominent features when a more significant feature could be used. The following criteria apply to census tracts in the United States, Puerto Rico, and the Island Areas: * Census tracts must not cross county or state boundaries. * Census tracts must cover the entire land and water area of a county. * Census tracts must comprise a reasonably compact and contiguous land area, with a few exceptions. * Exceptions to the compactness requirement might include areas with unique population groups, land uses, or settlement patterns, such as coastal communities, a community ringing a lakeshore, valley settlements, and commercial and industrial development along a linear transportation route defined as a special use census tract. * Census tract boundaries should follow visible and identifiable features. * Census tracts must have a basic numeric identifier composed of no more than four digits and may have a two-digit decimal suffix. The Census Bureau may modify and, if necessary, reject any proposals for census tracts that do not meet the published criteria. In addition, the Census Bureau reserves the right to modify the boundaries and attributes of census tracts as needed to meet the published criteria and/or maintain geographic relationships before or after the final tabulation geography is set for the 2020 Census. Census tracts must meet specific population and housing unit thresholds outlined on the next slide Census Tract Thresholds. The following criteria apply to block groups in the United States, including federally recognized American Indian reservations (AIRs) and off-reservation trust lands (ORTLs), Puerto Rico, and the Island Areas: * Must not cross census tract boundaries. * Must cover the entire land and water area of each census tract. * Must comprise a reasonably compact and contiguous land area, with a few exceptions. * As with census tracts, exceptions might include areas with unique population groups, land uses, or settlement patterns, such as coastal communities, a community ringing a lakeshore, valley settlements, and commercial and industrial development along a linear transportation route defined as a special use block group. * Must meet specific population and housing unit thresholds and suggested area and employment thresholds. * Boundaries should follow visible and identifiable features. Thank you, Muhammad Participant Statistical Areas Program (PSAP) U.S. Census Bureau Toll-Free: (844)788-4921 Email: GEO.PSAP(a)census.gov PSAP Home Page<https://www.census.gov/programs-surveys/decennial-census/about/psap… ________________________________ From: Reardon, Tim <TReardon(a)mapc.org> Sent: Monday, March 4, 2019 12:14 PM To: sdc_mlist(a)lists.berkeley.edu; ctpp-news(a)chrispy.net Cc: Joshua Wixom (CENSUS/FLD FED); GEO PSAP (CENSUS/GEO); todd.graham(a)metc.state.mn.us (CENSUS/ OTHER) Subject: PSAP -- Other criteria or principles for delineation? Hi all, I imagine many of us are busy diving into GUPS and getting started with the delineation work. Obviously there are specific Census-defined thresholds for each type of statistical areas, but I’m curious to know if there are other criteria or principles that folks are using as they consider possible adjustments. Historical continuity aside, what makes for a good block group or tract? Should we seek to minimize heterogeneity by drawing boundaries that separate very different neighborhoods/sub-neighborhoods? Or is it better to have block groups/tracts that include a more diverse set of households and residents? I would imagine that the approach taken to these questions will affect not only the confidence intervals of the resulting ACS tables, but also the results of whatever research uses those tables. Any opinions or literature references on this topic would be most welcome! Tim Reardon Data Services Director Metropolitan Area Planning Council, Boston From: Graham, Todd [mailto:todd.graham(a)metc.state.mn.us] Sent: Friday, February 22, 2019 1:10 PM To: sdc_mlist(a)lists.berkeley.edu; ctpp-news(a)chrispy.net Cc: joshua.wixom(a)census.gov; GEO PSAP (CENSUS/GEO) <geo.psap(a)census.gov> Subject: [State Data Centers] PSAP participants in every county: https://www2.census.gov/geo/pdfs/partnerships/psap/Primary_PSAP_Participant… Hi friends of Census Tracts— If it’s useful to you… Census Bureau has published a contacts list of 2020 Census PSAP participating agencies, here: https://www2.census.gov/geo/pdfs/partnerships/psap/Primary_PSAP_Participant… This list covers every county in the nation. If the listing for your county points to an erroneous agency or a nonworking phone number/email, I suggest you alert Census’s Geog Division: geo.psap(a)census.gov<mailto:geo.psap(a)census.gov> By the way, our office, Metropolitan Council of the Twin Cities, has a few upcoming meetings where we’ll be describing our review, validation and redrawing of Census Tracts and Block Groups. We are well into the preliminary work. Starting in March we will be pointing interested stakeholders to preliminary, proposed re-tracting for our 7-county region. And we will be inviting stakeholders to comment or express their needs for Census stats tabulations for sub-city zones, service areas, etc. We will consider whether those needs can possibly be met with 2020-vintage Tracts and Block Groups. It’s possible we may receive some comments from outside of our region. When that happens, we will be redirecting commenters to the other county governments and regional development commissions in our state… May want to bookmark the participants list mentioned above! Hope that helps. Cheers, Todd Graham [EMAILLOGO.png] Todd Graham Principal Demographer | Metropolitan Council | Research todd.graham(a)metc.state.mn.us<mailto:todd.graham(a)metc.state.mn.us> P. 651.602.1322 | F. 651.602.1674 390 North Robert Street | St. Paul, MN 55101 | metrocouncil.org/data<http://www.metrocouncil.org/data> [ConWUs][FaceBook_32x32]<http://www.facebook.com/MetropolitanCouncil>[… [EmailSU] <https://public.govdelivery.com/accounts/MNORGMETC/subscribers/new?preferences=true> -- SDC Clearinghouse (http://sdcclearinghouse.wordpress.com) --- You received this message because you are subscribed to the Google Groups "sdc_mlist(a)lists.berkeley.edu<mailto:sdc_mlist(a)lists.berkeley.edu>" group. To unsubscribe from this group and stop receiving emails from it, send an email to sdc_mlist+unsubscribe(a)lists.berkeley.edu<mailto:sdc_mlist+unsubscribe(a)lists.berkeley.edu>. Please be advised that the Massachusetts Secretary of State considers e-mail to be a public record, and therefore subject to the Massachusetts Public Records Law, M.G.L. c. 66 § 10.