Measuring College ROI

Photograph by Corbis

Calculating the return on investment (ROI) for a college education is a complex endeavor, with the outcome deeply dependent on methodology, assumptions, and other factors. One of the key factors that sets the PayScale methodology apart is that it results in an ROI figure that reflects not only costs and earnings, but also the likelihood that students will graduate and how long it will take.

The PayScale salary data are self-reported by individuals who use its online pay tools. For each school in the ranking, PayScale collected, on average, about 1,000 pay reports from alumni who graduated with a bachelor’s degree over the past 30 years, from 1982 to 2011. Since the data were reported in the last year, these earnings figures are in 2012 dollars. They include base salary or hourly wage, bonuses, profit-sharing, tips, commissions, and other cash earnings; they do not include stock, the cash value of retirement benefits, or other non-cash benefits. Only full-time employees who work in the U.S. were used for this analysis; self-employed, project-based, and contract employees were not. Graduates with advanced degrees were excluded.

Using these pay reports, PayScale calculates the current 30-year median pay for the graduates of each school. PayScale then calculates the amount that a high school graduate would have earned over the same period. Since a high school graduate would have begun earning a salary immediately (rather than after four to six years of college), PayScale first calculates the pay for those who graduated from high school in 1976, 1977, and 1978 (34 to 36 years of earnings). For this calculation, PayScale used the 75th percentile earnings of high school graduates, rather than the median, which is what was used for the 2011 ranking. The 75th percentile more accurately reflects the higher likely earnings of those who succeeded in being admitted to college, as well as the salary impact of having attended college for a few years before dropping out.

To determine how much more alumni from each college earn over and above what a typical high school graduate makes, PayScale simply subtracts the high school graduate earnings from each school’s 30-year median pay.

This earnings differential is what many consider to be the “value” of a college degree, but to calculate a true return on investment, one must factor in the cost of attending college.

To do that, PayScale first calculates the annual cost of attending each school—tuition and fees, room and board, books and supplies—and uses that to determine the total cost of graduating in four, five, and six years. At this stage, financial aid is factored in: Average grant aid is subtracted from the annual cost to arrive at a net cost. All data on costs and grant aid were supplied by the Integrated Postsecondary Data System (IPEDS), a warehouse for federal education statistics. The most recent data on grant aid available through IPEDS are for the 2009-10 academic year. Those data were used for all years in the calculation.

PayScale then creates a weighted average cost based on the percentage of students who graduate from the institution in four, five, and six years. For example, if the net cost to attend a school is $10,000 a year, it would cost $40,000 to graduate in four years, $50,000 in five years, and $60,000 in six years. If the school’s four, five, and six-year graduation rates are 45 percent, 65 percent, and 80 percent, then among those who actually graduate 56 percent do so in four years, another 25 percent do so in five, and another 19 percent do so in six. To find the weighted average, the percentages are multiplied by the dollar values and summed: ($40,000 x 56%) + ($50,000 x 25%) + ($60,000 x 19%) = $46,250.

Once all the calculations are done, determining ROI is fairly simple. By subtracting the cost of attendance from the earnings differential, we learn the value of the degree for a graduate. If you graduate from college, this is the amount you can expect to earn over and above what a typical high school graduate earns over the same period, after deducting the cost of getting the degree.

But not everyone graduates, and for those who don’t, the financial benefits of a college education are far less. So one last calculation is required. By multiplying the ROI for a graduate by the school’s six-year graduation rate, the result is an ROI figure far closer to the truth, one that reflects not only how much you’ll earn, but the likelihood of graduating.

That said, it’s still an approximation. PayScale maintains that the margin of error on 30-year median pay is 5 percent for most schools, 10 percent for those that have large variations in pay among alumni, fewer pay reports from graduates, or large numbers of graduates who go on to get advanced degrees.

While ROI is calculated for each school, there is wide variation within schools among majors, and schools that produce many graduates in high-paying majors such as engineering will, by necessity, have a higher ROI. The PayScale methodology doesn’t capture the value of a college degree for those who go on to get advanced degrees in law, medicine, or other high-paying fields. And since it deducts average grant aid from the total college costs, regardless of how many students actually receive such aid, ROI for those who don’t receive that amount each year, or don’t receive any aid at all, will be less.

To view the complete 2012 ranking, click here.

Before it's here, it's on the Bloomberg Terminal.