Online Program Home
  My Program

Abstract Details

Activity Number: 356 - Contributed Poster Presentations: Survey Research Methods Section
Type: Contributed
Date/Time: Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
Sponsor: Survey Research Methods Section
Abstract #323537
Title: Challenges in Linking Demographic Data at Different Geographic Levels
Author(s): Edward Mulrow* and Rebecca Curtis and Ned English and Yongheng Lin and Ilana Ventura
Companies: NORC and NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago
Keywords: GIS ; geography ; ACS ; ZIP code ; census tract ; calibration
Abstract:

Researchers can be challenged by data sets published at incongruent levels of aggregation. However, there exists the need to combine such data while maintaining its integrity and geographic relationships. We explore two approaches with trade-offs in accuracy and efficiency, with a focus on ZIP codes and US Census tracts. Our first method uses geographic information systems (GIS) to weight tract-level data from the American Community Survey (ACS) based on spatial overlap. The weights are the percent of area overlap for each tract intersecting each ZIP code. This method avoids the duplication of data caused by allocating all of a tract's data to each ZIP code it intersects and allows for a more nuanced distribution of data over matching the tract centroid to the ZIP code. Secondly, we describe a framework that uses calibration techniques to estimate overlapping regions based on published margin totals. The overlap proportion is used to allocate a portion of each tract to the ZIP code. Exploratory analysis provides insight into the strengths and weaknesses of each approach.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association