Methodology

How We Calculate Debt Burden

Data Source

Our analysis uses the U.S. Department of Education College Scorecard, specifically the "Field of Study" dataset. This comprehensive data source tracks student outcomes across thousands of institutions nationwide.

College Scorecard Field of Study Data

  • Publisher: U.S. Department of Education
  • Dataset: Most Recent Data by Field of Study
  • Data Year: 2023-24 cohort
  • Source URL: collegescorecard.ed.gov/data

The College Scorecard is updated annually by the Department of Education and provides the most comprehensive publicly available data on student debt and earnings by field of study.

Key Variables

We use the following data fields from the College Scorecard:

Variable Field Name Description
Major Name CIPDESC Classification of Instructional Programs description
CIP Code CIPCODE 4-digit CIP code used for category mapping
Credential Level CREDDESC Filtered to "Bachelor's Degree" only
Median Debt DEBT_ALL_STGP_EVAL_MDN Median cumulative debt at graduation
Median Earnings EARN_MDN_1YR Median earnings 1 year after completion

Calculation Methodology

Step 1: Data Filtering

We filter the raw data to include only:

  • Bachelor's degree programs (excluding Associate's, Master's, etc.)
  • Programs with valid (non-suppressed) debt data
  • Programs with valid (non-suppressed) earnings data

The College Scorecard suppresses data (marks as "PrivacySuppressed") when sample sizes are too small to protect student privacy. We exclude these records.

Step 2: Aggregation by Major

The raw data contains separate records for each institution-major combination. We aggregate by major name across all institutions:

  • Collect all debt values for each major
  • Collect all earnings values for each major
  • Calculate the median of each set

We use the median (not mean) to reduce the impact of outliers from individual institutions with unusually high or low values.

Step 3: Minimum Sample Size

To ensure statistical reliability, we only include majors with at least 10 data points (institution-major combinations with valid data). This results in 188 majors in our final dataset.

Step 4: Debt Burden Ratio Calculation

Debt Burden Ratio = Median Debt / Median First-Year Earnings

For example, if a major has median debt of $25,000 and median first-year earnings of $50,000, the debt burden ratio is 0.5 (or 50% of first-year earnings).

Step 5: Category Assignment

Each major is assigned to a category based on its 2-digit CIP code prefix:

  • STEM: Computer Science (11), Engineering (14, 15), Biology (26), Math (27), Physics (40)
  • Business: Business, Management, Marketing (52)
  • Health: Health Professions (51)
  • Arts & Humanities: English (23), Liberal Arts (24), Philosophy (38), Visual/Performing Arts (50)
  • Social Sciences: Psychology (42), Public Admin (44), Social Sciences (45), History (54)
  • Education: Education (13)
  • Communications: Communication & Journalism (09, 10)
  • Other: All remaining CIP codes

Limitations and Caveats

While this analysis provides valuable insights, there are important limitations to consider:

Data Timing Context

This analysis uses the 2023-24 cohort - the most recent data available from the College Scorecard. Earnings are measured approximately one year after graduation, reflecting early-career outcomes rather than long-term earnings potential.

We update this analysis when the Department of Education releases newer cohort data.

Data Limitations

  • Federal loans only: The debt figures include only federal student loans, not private loans or other borrowing.
  • Completers only: Data is for students who completed their degree, not those who dropped out (who may have debt without the earnings benefit).
  • 1-year earnings: First-year earnings may not reflect long-term career trajectories. This particularly affects "gateway" majors where many graduates pursue advanced degrees:
    • Biology, Chemistry, Human Biology: Many proceed to medical school - first-year earnings reflect pre-med jobs, not physician salaries ($200K+)
    • Communication Disorders: Most pursue Master's degrees before practicing as speech-language pathologists ($70K+)
    • Health/Medical Preparatory: By design, a stepping stone to professional programs
    • Psychology, Sociology: Many pursue graduate degrees for clinical or research careers

    Majors with high debt burden ratios may still offer strong lifetime ROI for those who continue to graduate or professional school.

  • Self-selection: Students who pursue certain majors may differ in ways that affect both debt and earnings beyond the major itself.
  • Sample size variation: Some majors have as few as 10 data points while others have hundreds. Majors with smaller samples may have wider variation from true population values.

Interpretation Caveats

  • Individual variation: These are medians across thousands of graduates. Individual outcomes vary significantly based on institution, location, specific career path, and personal circumstances.
  • Career satisfaction: Financial ROI is one factor among many. Job satisfaction, work-life balance, and personal fulfillment are not captured in these numbers.
  • Changing market: Past earnings don't guarantee future results. Labor markets and technology can shift demand for different skills.

Data Updates

The College Scorecard is updated annually by the Department of Education, typically in the fall. We update our analysis when new data becomes available.

Current data version: 2023-24 cohort

Reproducibility

Our analysis is fully reproducible. We provide a Python script that:

  • Downloads the College Scorecard Field of Study CSV
  • Filters and aggregates data as described above
  • Generates the TypeScript data file used on this site

The script and methodology documentation are available in our GitHub repository .