Public Sector

Illuminating college outcomes, while protecting privacy

Blue figure

Joining sensitive data sets from the Department of Education and the IRS in a way that protected privacy resulted in College Scorecard - a platform that allows students and families to simultaneously consider the cost and evidenced outcomes of a range of possible degrees.

case study image

Summary

Educational accountability, enabled by data collaboration

The challenge: a privacy-safe solution for joining data on educational degrees and financial outcomes.


The Department of Education sought to connect data about degrees with their financial outcomes, from data managed by the IRS.
The resulting insights would be useful to students and parents, but producing the data product involved significant individual data privacy risk.


A collaboration with Tumult enabled the privacy-protected solution, and College Scorecard now helps students and families be better-informed around degree and school choice.

key outcomes

0
Accredited institutions tracked
0
Fields of study cohorts tracked for financial outcomes
0
%
More data released, than with previous approaches

Solutions

Differential privacy safely unlocks the power of combining data from the Department of Education and the IRS. DP can support powerful data collaborations within and in collaborations outside your enterprise.

figure

goals

Support students and families to make informed decisions about higher education

Linking data on student majors and educational institutions with financial outcomes helps families make informed decisions.

Increase students’ and families’ access to information

Enhance decision-making by providing easy access to financial outcomes data, by major and college.

Maintain the privacy of sensitive information

Ensure the confidentiality of individuals' data while sharing insights on academic choices and their economic impacts.

our process

A collaborative, calibrated process

Our process is designed to maximize insights and utility while maintaining the privacy of highly sensitive data sets.

Design
  • Identify data that needs to be released
  • Identify privacy/accuracy requirements
  • Understand parameters that affect tradeoffs
arrowarrow
Tune
Interactively with stakeholders, engage in privacy negotiation:
  • Visualize privacy loss vs. accuracy tradeoffs
  • Tune parameters
arrow
Stage
Implement and deploy in a secure vulnerability-free privacy platform
blue figure
arrow
Deploy
Customers benefit from access to previously out-of-reach insights
arrowarrow
Manage
  • Track approvals to catalog and manage data releases
  • Monitor and audit
  • Manage data re-use

the results

“We need a system that’s inclusive, that delivers value, and that produces equitable outcomes. We need transparency in data more now than ever before.”
– Secretary Miguel Cardona

What value did the College Scorecard project produce?

The platform is the first to make it simple to search and compare colleges: their fields of study, costs, admissions rates and student outcomes.

Powerful and easy-to-use website for students and families

Increasing the transparency of information that helps users understand the costs and benefits of a higher education degree.

Robust data, actionable insights

The site makes it simple to compare college costs, student debt, graduation rates, admissions test scores and acceptance rates, student body diversity, and  post-college earnings and much more.

Expanding data and utility

New updates include median earnings of graduates four years after graduations, new demographic data, including race/ethnicity data for full-time college staff, and student-to-faculty ratios.

More resources

View All
700 staff and contractors support Wikimedia projects, communities, donors, and readers.

Revealing Wikipedia usage data while protecting privacy

Case Study

Wikipedia’s volunteers want a systematic way to prioritize where to focus their work. Which entries are being read most? By which readers where?
DP was the technology that solved for the twin, and potentially contradictory, goals of privacy preservation and actionable insights.

Read more
right arrow
Green shape on dark background

Using differential privacy to safely release earnings data on post-secondary grads

Research

When the U.S. Census Bureau needed to release data on the median earnings of post-secondary graduates, one Tumult Lab founder helped the Census Bureau design a new differentially private algorithm.

Read more
right arrow
700 staff and contractors support Wikimedia projects, communities, donors, and readers.

Revealing Wikipedia usage data while protecting privacy

Case study

Wikipedia’s volunteers want a systematic way to prioritize where to focus their work. Which entries are being read most? By which readers where?
DP was the technology that solved for the twin, and potentially contradictory, goals of privacy preservation and actionable insights.

Read more
right arrow
Green shape on dark background

Using differential privacy to safely release earnings data on post-secondary grads

Research

When the U.S. Census Bureau needed to release data on the median earnings of post-secondary graduates, one Tumult Lab founder helped the Census Bureau design a new differentially private algorithm.

Read more
right arrow

Unleash the power and value of your data.