Evidence base and collection periods

This study focuses on student-loan borrowers enrolled in income-driven repayment plans.

Survey data collection spans January 2025 through December 2025. The dataset contains 1,810 survey records from this borrower population, including 1,658 completed surveys. Available-case denominators vary by measure.

Interview data collection with this borrower population began in May 2026 and remains ongoing. As of August 2026, the qualitative sample includes 67 interview participants and 2,933 coded response units.

Sample recruitment and limitations

Participants were recruited throughout 2025 through public calls on social media, outreach shared through email lists with assistance from Protect Borrowers, and posts in online communities focused on borrowers using income-driven repayment (IDR) and Public Service Loan Forgiveness (PSLF), along with outreach supported by other organizations. Participation was voluntary, producing a nonprobability convenience sample of borrowers who encountered the recruitment materials and chose to respond.

We do not know how closely this sample matches the broader populations of borrowers enrolled in IDR plans or pursuing PSLF. The federal government publishes limited information about the demographic and household characteristics of these populations. Readers can review the available Federal Student Aid Public Service Loan Forgiveness Data and the study Another Lesson on Caution in IDR Analysis: Using the 2019 Survey of Consumer Finances to Examine Income-Driven Repayment and Financial Outcomes for examples of the limited evidence available on these borrower populations.

Findings describe the people who participated in this study and should not be interpreted as nationally representative estimates of all IDR or PSLF borrowers—but could be strong guideposts to the wider populations.

Survey sample at a glance

These unweighted summaries describe valid responses in the convenience sample. Denominators vary because not every participant answered every question.

Age: 69% were ages 30–49
Gender: 75% selected Female
Race or ethnicity: 71% White; 14% African American/Black
Individual income: 66% reported $65,000 or more
Education: 42.9% reported a Master’s Degree
Employment sector: 83% worked in government, education, or nonprofit settings
Federal loan balance: 40% reported balances of $100,000 or more
PSLF enrollment: 81% reported being enrolled in PSLF
Current loan situation: 44% SAVE forbearance; 28% repayment; 21% PSLF/TEPSLF forgiveness

Survey estimates

All survey findings are unweighted. Percentages, averages, and distributions use valid responses available for each measure. Results are descriptive and should not be interpreted as causal effects or as automatically statistically significant differences.

Thirty survey outcomes

Financial position and well-being: Homeownership; FICO Score; Monthly Rent Payment; Monthly Mortgage Payment; General Savings Account; Retirement Account; General Investment Account; Credit Card Debt; Medical Debt; Financial Stress Scale; Kessler Psychological Distress Categories; Food Security Category; Self Harm Thoughts; and Suicidal Ideation.

Self-harm, suicidal experiences, and student-loan relationships:

Self Harm - Ever Made Plans or Attempts
Self Harm Thoughts are Related to Student Loan Debt
Suicidal Ideation - Ever Made Plans and Attempts
Suicidal Thoughts are Related to Loans

Education, college value, and repayment information:

Family Education Expectations
Degree for a Well-Paying Job
Degree Importance Over Time
Children: Two-Year Credential
Children: Four-Year Degree
College Experience Importance
College Was Worth It
Servicer Trust
Servicer Information Clarity
Federal Guidance Trust
Federal Information Clarity
Repayment Uncertainty

The data explorer displays the exact survey wording for the added financial, education, repayment-information, and sensitive follow-up outcomes.

The four added self-harm and suicidal follow-up outcomes use available responses to those questions. Their valid denominators are smaller than the full survey because these items were not answered by every respondent; the explorer displays the valid n for each outcome and comparison.

FICO values outside 300–850 were excluded. Valid scores are categorized as Poor (300–579), Fair (580–669), Good (670–739), Very Good (740–799), and Exceptional (800–850), following Experian’s published base FICO ranges. Food security categories are High (score 0), Marginal (1), Low (2–4), and Very Low (5–6). Self Harm Thoughts and Suicidal Ideation are coded “Yes” when a respondent reported the thought on any number of days above zero during the past 12 months, including “7 Days.”

The seven dollar-amount measures are displayed in five fixed, publicly interpretable ranges. The highest category is open-ended and includes all larger reported values. General investments exclude retirement accounts; general investments, credit-card debt, and medical debt retain $0 as a distinct first category. The explorer reports the mean and median overall and beside every comparison-group bar, together with the valid sample size; means may exceed medians when a distribution is strongly right-skewed.

Comparison variables

Age
Federal Loan Balance
Individual Gross Income
Racial or Ethnic Group
Gender
Married
Dependent Children
Highest Earned Degree
Job Sector
Current Federal Loan Situation
Enrolled in Public Service Loan Forgiveness
Kessler Psychological Distress Categories
Food Security Categories
Self Harm Thoughts
Suicidal Ideation
Monthly Rent Payment
Monthly Mortgage Payment
General Savings Account
Retirement Account
General Investment Account
Credit Card Debt
Medical Debt

Age categories are 18–29, 30–39, 40–49, 50–59, and 60+. The seven financial amount variables use the same five fixed ranges when selected as comparison variables.

Small samples

Small categories are not suppressed because the public file contains only summarized results and no individual identifiers. Every estimate based on 15 or fewer valid respondents is marked: “Small sample: This result is based on 15 or fewer respondents and may be unstable. Interpret comparisons with caution.”

Interview selection and evidence

Among 1,192 survey respondents who indicated that they were willing to be interviewed, 370 had been contacted as of the most recent count and 67 interviews had been completed. Potential interviewees were randomized for selection from the pool of willing respondents. Interview data collection with student-loan borrowers enrolled in income-driven repayment plans began in May 2026 and remains ongoing.

Guided AI Thematic Analysis (GAITA), led and reviewed by the researcher, organizes the corpus into 15 analytic themes; each theme opens into a public-facing story with four participant quotations and related evidence. Coded excerpt totals describe coding density, not prevalence in the broader borrower population.

Interviews provide depth and context but are not statistically representative of all survey respondents or all borrowers enrolled in income-driven repayment plans. Quotations remain connected to the relevant outcome or qualitative theme; they are not automatically attributed to demographic subgroups selected in the data explorer.

Interview quotations have been lightly edited to remove verbal fillers, false starts, and repetition, and may be shortened for clarity. These edits do not change the speaker’s meaning. All names are pseudonyms.

Research and funding

Conducted by Daniel A. Collier, Associate Professor at the University of Memphis and Research Fellow at the Princeton University Debt Collection Lab.

This research was funded by the Lumina Foundation. The findings and conclusions presented are those of the research team and do not necessarily reflect the views of the Foundation.