How data from 3 major sources can reveal the ROI on college degrees

The economic payoff from a college education continues to be a hot topic, mainly because the number one reason students give for going to college is that a degree will improve their chances of landing a desirable job and earning a good salary. A college degree is still seen as the best ticket to economic security.

Measuring the financial benefits of college is also of interest to policy makers and university administrators as they consider where to invest their limited funds. And it’s a frequent topic for researchers studying economic inequalities and socio-economic mobility.

These days, a popular measure of the tangible value of college is to compare the earnings of individuals who have earned a college degree to those who haven’t. Another measure— increasingly invoked in college rating schemes such as Forbes or by policy organizations like Brookings and The Equality of Opportunity Project — is to compare the social mobility and career earnings of students who attend one college versus another, or who major in different academic fields.

Consumers have several options for evaluating the return on investment (ROI) from attending a given college or declaring a particular major.

Two widely used alumni earnings databases are the Department of Education’s College Scorecard and PayScale’s College Salary Report. A third, more recently developed measure is the Price-To-Earnings Premium, or PEP, developed by Third Way, a national think tank.

Degreechoices’ rankings of national universities use College Scorecard data, with a nod to Third Way’s methodology, to evaluate degree programs by value for money in an accessible, user-friendly way.

How comparable are these 3 databases? Would a prospective student reach the same conclusion about the payoff from attending Stanford if she consulted the Scorecard or PayScale? How different would a university’s ROI look to a university trustee, depending on whether he examined the Scorecard or Third Way’s PEP?

It turns out the methodologies and calculations behind the 3 databases differ in several ways so it should come as no surprise that the estimated economic payoff may vary, depending on which source is used.

The Department of Education’s College Scorecard

College Scorecard is an initiative by the U.S. Department of Education. Its aim is to make data on economic outcomes of college degree programs more accessible to the public. Among other indicators, College Scorecard publishes data on average net cost (after financial aid) and average post-graduation earnings for students who received at least $1 in federal financial aid (which covers 80%+ of all college students.

PayScale’s College Salary Report

This regularly updated report provides consumers with information about the early and mid-career median salaries of graduates from both public and private not-for-profit colleges at the associates and bachelor levels. Based on a salary survey self-reported and submitted by respondents, the salary data is broken down by major and college.  Note, it does not provide information about specific majors at different colleges.

Third Way’s Price-to Earnings Premium (PEP)

Third way calculates a PEP for a specific college. It also provides the amount of time it takes students to recoup their costs for a specific college degree. Relying on data gleaned from the U.S. Department of Education and the Census Bureau, Third Way investigators:

1. Calculate the amount a student pays out of pocket to attend a given institution (i.e., the net amount after scholarships and grants are factored in). They assume 4 years of cost for a bachelor’s degree, 2 years for an associate degree, and 12months for a certificate.
2. Using College Scorecard data, they calculate the median salaries of college graduates for specific majors within an institution (currently limited to 2 years after graduation) along with the median salaries of those with only a high school diploma within the same state where the college is located.
3. Divide the amount in #1 by the difference between the median salaries of college graduates and the high school graduates, gathered in #2.
4. The quotient, or PEP, tells you the number of years it would take for students to recoup the net costs of their education.

PEP data at the individual program level are available in a spreadsheet that can be found here.

How do the tools differ?

1. Data Sources

• The College Scorecard relies on the Integrated Postsecondary Education Data System (IPEDS)and National Student Loan Data System (NSLDS) for much of its data. It gathers students’ earnings data from de-identified tax records and then aggregates them at both the institutional and program level.
• PayScale is a voluntary system where individuals complete an online survey and self-report their earnings and other data. In earlier years of its Salary Report, PayScale’s sample was criticized for being too small and unrepresentative. PayScale claims that its latest survey has grown to a sample of 3.5 million respondents, representing more than 4,000 degree-granting colleges.
• The PEP is calculated using salary data from the College Scorecard based on graduates who are not attending another school for additional education. After programs were excluded for a variety of reasons (e.g., too few graduates), more than 38,000 programs enrolling more than 2 million students were included.

2. Students included

College Scorecard reports data only for students who receive federal financial aid, a group likely to be less economically secure than students who aren’t eligible for such support. It includes actual salaries, gathered from tax records, for students who have either attended a certain college or completed their degree programs and are not still enrolled in additional education.

PayScale, on the other hand, includes students who have completed their bachelor (or associate) degree, but doesn’t limit its sample to recipients of federal aid. It sorts its data according to 2 different groups of students attending 4-year colleges:

• Bachelor’s only – just those students who earned their BA/BS from a given school and did not go on for an additional degree
• All alumni – includes the students above plus those who earned additional graduate or professional degrees

The PEP, like College Scorecard, includes data only for students who received federal financial aid. In its initial iteration, it was restricted to institutional-level rather than program-level data.  It presented outcomes for students 10 years after attending a given school, regardless of whether they graduated. In a subsequent version that includes information about program-level ROI, it uses salary information for graduates only.

3. Time period covered

College Scorecard 10 years after enrollment (or 5 years after graduation for program-level data).

PayScale reports salaries for 2 time periods:

• Early-career salary – median salaries for alums 0-5 years after completion
• Mid-career salary – salaries at 10+ years from graduation

The PEP’s institutional-level data are based on salaries 10 years after students attended a given school. Its program-level ROI is based on graduate earnings, 2 years after degree completion.

How is the data different?

As one illustration of how such methodological differences matter, let’s consider a student who earned a B.S. in Chemical Engineering from the University of Illinois Urbana-Champaign. What would we learn about that student’s likely earnings from each of the 3 sources?

• According to College Scorecard, is $93,416 just 4 years after graduation.
• With Payscale, the data needs to be examined in several ways to approximate a comparison with the Scorecard’s figure. Payscale shows the early and mid-career earnings for a University of Illinois graduate with a bachelor’s degree to be $68,300 and $123,600 respectively. Including graduates who earned additional degrees, the comparable salaries are $70,200 and $131,200. The median early career earning for individuals with a bachelor’s degree in chemical engineering, regardless of college attended, is $76,900 while the median for mid-career is $135,900.
• Based on Third Way’s PEP tool, the median salary is $93,416 which is identical to the Scorecard. In addition, it shows that, on average, students would need 1.27 years after graduation to recoup their out-of-pocket educational investment.

What about a student fortunate enough to be accepted by both Harvard and Stanford?

• According to the College Scorecard, depending on the field of study, a Stanford graduate could expect annual earnings of $124,080, 10 years after enrolling. A Harvard graduate would be looking at earnings of $101,817.
• PayScale pegs the median early-career annual salary for Stanford graduates at $87,100 and their mid-career median at $156,500. The comparable Harvard figures are $80,900 and $156,200, respectively.

Based on PEP scores, graduates of both Harvard and Stanford could expect to recoup their out-of-pocket educational costs in less than a year.

Data limitations

Although each of the 3 tools provides useful information, it’s important to keep in mind their limitations.

• All of them are based on averages or medians so they are, at best, estimates of what a specific student can expect to earn from completing a particular program.
• None of the tools measure lifetime earnings, which are likely, depending on the field of study, to have very different trajectories. In some fields, annual salaries may increase substantially over the course of a career; in others, entry compensation may not be very different from the amount paid at the end of a career.
• Cost of living is not factored into the figures so 2 individuals completing the same program may be paid very different salaries depending on where they live.
• Academic programs with few students are often excluded either because of an insufficient number of individuals to calculate a meaningful average or— in the case of the Scorecard and PEP— because of concerns about the privacy of individual students.
• Other factors besides a college degree influence career earnings. For a good review of those factors, a new report from the Georgetown University Center on Education and the Workforce (CEW) reveals the impact that age, field of study, occupation, gender, race and ethnicity, and location can have on lifetime income.

Final thoughts

Of course, annual compensation or lifetime earnings is just one consideration in choosing where to attend college. Many students find their education to be worth the cost for several other reasons such as learning more about something they love, meeting new people, preparing for graduate school, and enriching their understanding of the world.

But increasingly, economic payoff is an important contributing factor for many students’ college decisions. In such instances, more information is better than less. Although the Scorecard is the most comprehensive database and is updated regularly, consumers are best advised to consult various sources for estimates of future earnings potential and to remember the limitations of each one.

If you’re looking for a university ranking system based on the concept of ROI, check out our lists of the Best National Universities, the Best Public Universities, and the Best Colleges for Computer Science, among dozens of other college rankings.