Intersections as Proxies for Home Address: Alternatives for Community GIS
Danielle Wilfong is a doctoral student in the Community Research and Action program with a specialization in socio-spatial research at Vanderbilt Peabody College. She was also the 2020-2021 RPW Center HASTAC Scholar.
Do you remember road trips in the 1990s? I recall badgering my dad to pull over and ask for directions. When we finally convinced him, he would stop at a gas station and ask for a map — not directions. If you are my dad or like him, congratulations, you managed to retain key survival skills in the new millennium.
But if you are like me, you now outsource much of that mental work to Google Maps or Waze. For many people, gone are the days of using the stars, a compass, or a static, non-talking map in everyday life. The irony is, when planning community GIS projects, someone will inevitably suggest collecting intersections (i.e., cross streets) as a proxy for a participant’s home address. (GIS stands for geographic information system and it is a spatial system that creates, manages, analyzes, and maps all types of data.)
This occurs despite the fact that remembering a cross street is like having to remember a phone number without your phone. Most people will not know or will misremember that information. The result is frustration, confusion, or embarrassment for participants and missing or unreliable data for yourself.
I understand the appeal of asking for an intersection. It allows one to forgo collecting, storing, and protecting people’s home addresses. They are also more precise than zip codes. However, there are plenty of ways to improve the reliability of approximating where one lives.
Here are some suggestions for both best practices and for creating launching points for you and your project that don’t rely on intersections.
Use your GIS to develop a plan for data collection. It may be unreasonable to expect participants to know their nearest intersection, but it sure makes sense to assume that participants know where they go in their daily lives and if those places are close to where they live. The researcher is the responsible party for intersections, census tracts, and road designations (i.e., lane, boulevard, avenue, street) and other cartographical knowledge. I encourage researchers to become familiar with a study area in advance of data collection. That knowledge can help increase the reliability and accuracy of your data.
Visualizing your study area in a GIS allows you to geographically define its perimeter and its distinguishing features and landmarks. One way to do this is by overlaying OpenStreetsMap with the spatial unit of your choice (e.g., census block, census tract, radius). From here, you can plan where to canvas and have enough detail about an area to help participants choose locations that represent their home or neighborhood without disclosing their residence.
Here are a few questions that a researcher can use to approximate a home address:
-
Would you say [insert the name of the location where you are canvassing] is your neighborhood convenience store?
-
If you wanted to get food, but did not want to travel more than a one-minute-drive from your house, where would you go?
-
Are you closest to grocery store A or grocery store B?
Use a GIS to collect data. Field collection tools, like Survey123, can be useful for collecting spatial data alongside more standard survey data (e.g., multiple choice questions, Likert scales). These tools can be used on a variety of devices (e.g., mobile, laptop) and some can be used offline and then uploaded once you have access to the internet.
Tools like Survey123 will allow you to type in an address. There are also point and click instruments that allow you to navigate and select a point on a map. If you have the time to make sure participants can navigate the tool and participants have time and interest to learn, a tool like Survey123 may be a good fit for your project. Additionally, these instruments are also useful if the researcher is the primary user.
On the other hand, field collection tools are not always the best choice. Experience has taught me that it is not reasonable to assume that most participants know how to read a map or navigate a digital platform. Sometimes a pen and paper are more reliable. In this case, it still may be useful for you, the researcher, to employ tools like Google Maps to find and/or verify addresses while using a paper format. Whenever possible, make sure to verify the accuracy of addresses during data collection, not afterwards.
In short, if all you are after is someone’s cross street, cross streets are great. If not, it might be time to explore alternative strategies for approximating home addresses. Happy mapping!
Danielle Wilfong is a doctoral student in the Community Research and Action program with a specialization in socio-spatial research at Vanderbilt Peabody College. Before coming to Vanderbilt, she earned a bachelor degree in Linguistics from Duke University and a masters degree in Prevention Science and Practice from the Harvard Graduate School of Education. She has also been a high school English teacher in Shizuoka, Japan and a program associate for a high school mentoring program.