← All systems

Methodology

Overview

This system maps the flow of first-time freshmen from their home state to destination colleges and universities, using IPEDS enrollment-by-residence data. It answers questions like: where do California's high school graduates go to college? What share of students at the University of Michigan are from out of state? How do university systems like the UC system or SUNY draw from different states?

Data Sources

IPEDS Fall Enrollment by Residence

First-time freshman enrollment counts by state of residence and institution — the core data for tracking where high school graduates go to college

Format: csv

IPEDS Institutional Directory

College/university directory with name, state, sector, control, HBCU status, and geographic coordinates from IPEDS via the Urban Institute Education Data Portal

Format: csv

NCES CCD School Universe (High Schools)

Common Core of Data school-level directory filtered to high schools, providing enrollment, demographics, locale, and district linkage

Format: csv

College Scorecard

Institutional outcomes data — graduation rates, median earnings, admission rates, cost of attendance — from the US Dept of Education

Format: csv

How It Works

  1. Enrollment by Residence: IPEDS collects data from every postsecondary institution on where their first-time freshmen came from (state of residence). This creates a matrix of origin-state × destination-institution enrollment counts.
  2. Institutional Directory: IPEDS institutional characteristics provide college metadata: name, state, sector (public/private, 2yr/4yr), university system membership, and geographic coordinates.
  3. Flow Construction: We join enrollment-by-residence with the institutional directory to create a complete picture of student flows from states to colleges.
  4. State Summaries: For each origin state, we aggregate total freshmen, in-state vs out-of-state rates, 2-year vs 4-year splits, and top destination colleges.
  5. Institution Summaries: For each college, we aggregate total incoming freshmen, number of feeder states, in-state percentage, and top origin states. We enrich with College Scorecard outcomes data (graduation rates, earnings).
  6. System Summaries: For university systems (UC, SUNY, CSU, etc.), we aggregate across campuses to show system-wide enrollment patterns.

Key Definitions

First-time freshmen
Students entering a postsecondary institution for the first time at the undergraduate level. Does not include transfers.
State of residence
The state where the student lived before enrolling, as reported to IPEDS by the institution.
In-state rate
Percentage of first-time freshmen whose state of residence matches the institution's state.
University system
A multi-campus system reported in IPEDS (e.g., University of California, SUNY, Penn State).

Limitations

  • IPEDS data is at the state-of-residence level, not individual high school or district level. School-level feeder analysis requires state-specific data.
  • First-time freshman counts may not perfectly capture all college-going behavior (e.g., gap years, delayed enrollment).
  • Some institutions may underreport or not participate in IPEDS surveys.
  • College Scorecard outcomes data uses different cohort definitions and may not align perfectly with IPEDS enrollment years.
  • University system membership in IPEDS may not be current for all institutions.

Future Enhancements

  • Incorporate state-specific high-school-to-college datasets where available (e.g., California, Texas)
  • Add district-level college-going rates using CCD data combined with state datasets
  • Peer-frontier scoring: compare college-going rates against similar districts
  • Sankey flow diagrams for state-to-college visualization
  • Choropleth maps showing in-state/out-of-state patterns