Detroit Residential Parcel Survey - Methodology

Data Driven Detroit developed the field maps and survey data forms for the Detroit Residential Parcel Survey, www.detroitparcelsurvey.org.

The maps were created with the goal of easy reading in the field. Our focus was on creating legible addresses printed on top of parcel objects on regular 8.5"X11" paper. We used ESRI ArcView software as our mapping tool, and added InfoGeographics' MapApp to do layout and mass production of 598 pdf maps that fully cover all the parcels in the City of Detroit. These two products, especially the MapApp software, were unbelievably essential to our success. We experimented with several ways to separate the geography into manageable peices (Census block groups, Census tracts, Assessors' neighborhoods, DPW collection areas). No existing geography was uniform enough in shape to print for easy reading of parcel addresses. Ultimately, we created our own by an arbitrary fishnet or "Atlas" over the City of Detroit's boundary file, see below attachment - DetroitSurveyIndexMap.pdf. This Atlas allowed us to create the grid cell maps needed for complete coverage of the City, examples are below for H-11 and N-17.

The survey data forms were based on previous residential surveys done by Community Legal Resources - Detroit Vacant Property Campaign and further modified for this survey. DVPC conducted the training for field supervisors and surveyors. The data, parcel records, or address level data (~387,000) was supplied by the City of Detroit - Assessors and was the assessment roll land records for 2009. The parcel data was geospatially joined with the atlas grid layer, and
and coded with its associated Atlas grid cell number. We then used Microsoft Access to create the survey instrument, an Access Report populated with address information, see examples below of the DRPS_FORM* files. We grouped the data by grid cell number, street name, and whether the street address number was Odd or Even given that Detroit addresses are separated by which side of the street they are on.    

Survey field grid cell books were created from printouts of the pdf maps and survey data forms. They were compiled and made into "field books"with our partners
from University of Michigan's Ginsberg Center who managed the field collection. The Survey Field Supervisors grouped sets of grid cells, usually 4, but as much as 6 and as little as 2, together into a field book. The grouping of grid cells was by contiguous grids determined by how the street configuration existed in the geography. Areas with North - South streets were grouped by associated atlas grid number (A-3, B-3, C-3, D-3), and areas with East - West streets were grouped by atlas grid letter (C-22, C-23, C-24, C-25). Surveying was done using about 50 Detroit residents and University of Michigan students working in three-person teams. Each team had one driver with a GPS unit, one surveyor on the right, one on the left, each with an Odd or Even field book depending on which side of the street they were covering. If they turned onto another street, all they had to do was switch their odd and even books. Teams did not leave the vehicles and all field data was collected from the street. 
 
 
Ċ
Unknown user,
Feb 28, 2010, 7:33 PM
Ċ
Unknown user,
Feb 28, 2010, 7:58 PM
Ċ
Unknown user,
Feb 28, 2010, 7:58 PM
Ċ
Unknown user,
Feb 28, 2010, 7:32 PM
Ċ
Unknown user,
Feb 28, 2010, 7:32 PM
Ċ
Unknown user,
Feb 28, 2010, 7:32 PM
Ċ
Unknown user,
Feb 28, 2010, 7:32 PM
Ċ
Unknown user,
Feb 28, 2010, 7:30 PM
Comments