Innovations in Health Policy

Insights from NYC Open Data

Ryan Quan | [email protected]


Phillip Guo, 2013
  • Overemphasis of the analysis and reflection phases
  • Underemphasis of the preparation and dissemination phases
  • Problem spaces may have no data to begin with
  • Insights have few outlets to actualize into action

Community-Level Change

  • How can we use Open Data to inform change at the local level?

  • What are some avenues for turning insight into action?


Motivating Example

Motivating Example


Motivating Example



One person can instigate change at the community level and resolve unintended policy side-effects:

  • Open Government Initiatives

    • Encourages public oversight
    • Enables creative solutions from civically minded individuals
  • Basic Exploratory Data Analysis

    • Outliers can reveal an interesting story
    • Descriptive statistics can inform future research
  • Social Media

    • Disseminate findings to larger audiences
    • Generate public discussion

NYC 311 Service Requests

Top 10 Complaints

plot of chunk unnamed-chunk-2

Top Agency Complaints

plot of chunk unnamed-chunk-3

Top Non-HPD Complaints

plot of chunk unnamed-chunk-4

Top Noise Complaints

plot of chunk unnamed-chunk-5

311 Rat Complaints

311 Rat Complaints

Using QGIS, an open-source GIS software, we can run a points-in-polygon analysis and identify clusters of complaints inside our district of interest.

Here we identified a building on 151st between Amsterdam and Broadway with several open rat complaint cases from different callers:



  • Pull, wrangle, and visualize open data using open-source software

  • Time-series analysis to observe monthly trends

  • Qualitatively-driven

    • Ex: Supporting rat anecdotes with 311 complaint data
  • Quantitatively-driven

    • Ex: Supporting pothole data with pictures from on-site visits
  • Submitting evidence to respective government agencies for quick-fixes

Other Examples

Benefits of Open Data

  • Reproducibility of Analysis

  • Motivates "neighborhood-specific" research

  • Encourages collaboration and new developments

  • Removes barriers to civic engagement

Downsides to Open Data

  • Potential for abuse

  • Quality of data sources

  • Difficult to go beyond correlational studies