Geocoding and Maps

In Part 2, I map WGU instructor alma maters by geocoding each university from the cleaned CSV.
The output is a reproducible pipeline that generates an interactive bubble map showing where instructors earned their degrees.

Interactive Map

--- ## Geocoding the Universities

To locate each institution, I used the Google Maps Geocoding API.

  • Created a Google Cloud project and enabled the Geocoding service.
  • Generated an API key and stored it in apikey.yaml (kept private).
  • Google provides $200 of free credit per month; all ~350 unique lookups for this dataset stayed well within the free tier.
  • Results were cached in uni_geo_mapping.json to avoid duplicate requests.
  • A few ambiguous names were corrected with manual overrides.
  • Validation: spot-checked ~10 entries, fixed typos, and confirmed top universities matched expected locations.
  • Final outcome: all universities resolved successfully after overrides.

Data Flow

(Upstream inputs and caches are omitted here for size/privacy.)

Script

  • make_bubble_map.py — reads the instructor CSV, joins counts with cached coordinates, writes the joined dataset, and produces the interactive bubble map.

Notes and Next

  • ~350 unique universities successfully mapped after a few manual fixes
  • Total cost: $0 (entirely within Google’s free tier)
  • The bubble map highlights instructor clusters across the U.S. with a handful abroad

Next in the Series

In Part 3, we’ll move beyond alma maters to the research output of WGU faculty.
Using the Semantic Scholar API, I’ll:

  • Query by instructor name and institution to identify published works
  • Build a reproducible archive of journal articles, conference papers, and books
  • Link each publication back to the normalized instructor dataset
  • Export both a CSV and a searchable interface for browsing WGU faculty research

Read here: WGU Instructor Atlas 3 — Research Archive