Research Request - Transit Agency Peer Groups ACS Data Research #1542
Description
Research Question
Single sentence description: How do we classify CA transit agencies into peer groups based on their characteristics in ACS data covering their service areas?
Detailed description: Transit agencies can be compared to each other based on the populations they serve. What demographic characteristics are relevant to track when grouping or clustering agencies into peer groups? Look at incorporating ACS data or tools we already have like the EQI or Replica (which process ACS data) to add to the bigger peer groups research.
How will this research be used?
subtask of #1415
Stakeholders & End-Users
Juan Matute is asking, and the audience is policymakers who want to know how transit agencies are similar to each other and what can be expected of similar agencies re: performance.
Metrics
correlation matrices
Exploratory principal components, k-means clustering or similar
Data sources
Cal-ITP data sources:
Potentially: Replica, EQI
External data sources: ACS data
Remaining data source questions:
Can we come up with any other derived variables that would be useful in existing analysis products?
What types of derived variables from census data exist in the literature?
Deliverables
Exploratory notebook with statistically determined groupings based on relevant characteristics of the population served by agencies
Clean agency-level table with relevant variables that can be integrated into bigger multivariate analysis
Timeline of deliverables
Estimated completion date TBD