This work-in-progress survey tool will make it easy to run, evaluate, and course-correct resident research surveys of internet use. Code for America has developed this process as parts of its Digital Front Door initiative, and included the results in our Oakland Phase 1 Report:
-
Resident survey results showing channels of internet access.
We believe that mobile-primary users are significantly under-represented in our sample, at least if the most reliable available national statistics are correct.
-
Geography process describing how we correlated self-reported neighborhoods to Census geography.
We asked each participant “What Oakland neighborhood do you live in?,” and allowed free-text responses so respondents could accurately describe their location. Values ranged all widely, from close matches to CEDA-defined neighborhood names, to council districts, zip codes, and large areas like “East Oakland.”
We are using a combination of data from Census Reporter’s API for ACS 5-year data, Google Docs for survey data storage, TypeForm for survey response gathering, and probably Turf.js and Leaflet for map display.
There will probably be a back-end written in Python and Flask.
Oakland analysis from October 2014 can be found in analysis-oakland-2014-10-27.
We’ll be formalizing a few help wanted issues in the coming weeks:
- Survey setup, connection between TypeForm and Google Docs, likely via Zapier.
- Ways for residents to identify their location on a map without pinpointing it. Neighborhoods or tract groups would be best; zip codes are not sufficiently granular.
- Visual representation of in-progress results with race/ethnicity/income correlations.
Percentage of Oakland population Hispanic or Latino, 2010 Census:
