What is Giving Compass?
We connect donors to learning resources and ways to support community-led solutions. Learn more about us.
Providing better information on college quality to potential students and their families has been a major focus of bipartisan higher education policy efforts. In late 2015, the Obama administration augmented its College Scorecard to include earnings information for most colleges in the country. Several states have also published their own data on graduates’ labor market outcomes, including average earnings and typical employment rates.
But there is little evidence on how these new outcome-based data affect student decision making and the broader market for higher education.
This report summarizes the results of a three-year effort aimed at assessing the demand for and impact of program-level information on labor market outcomes. We developed an informational website, GradpathVA, that displays academic major–level earnings and other key data points, piloted it at a set of Virginia high schools, and collected data to assess the tool’s effects on high school students’ college-going behavior. Some schools received the treatment version of the website, and others received a control version that looked similar but did not include labor market outcomes.
There did not seem to be significant pent-up demand among high school seniors for information on labor market outcomes, despite its appeal to policymakers and researchers. Only 25 high schools in Virginia agreed to participate in the study, though we aggressively worked to recruit from a pool of more than 300 high schools. Use of the informational tool at the schools that did participate was low, and users at schools randomly assigned to receive the treatment version of the intervention did not visit the website more frequently than those at schools who received the control version.
The rollout of the tool had no detectable impact on students. Students from schools that received the treatment version of the website did not choose majors with higher average wages or institutions that had higher graduation rates or lower net prices than did students who received the control version.
These results indicate that simply publishing and marketing earnings data on a website is unlikely to change the behavior of prospective college students. In fact, publishing the data on a new website may be part of the problem. In focus groups, students said the new web tool was too similar to other online resources they are more familiar with. Efforts to increase the use of labor market data may have more success working through platforms students already use. Future efforts should also consider market-testing different designs, targeting students earlier in their academic careers, and providing resources for students entering education after time in the workforce.