When it comes to commercializing their work, many university scientists choose one of two main paths: they turn to their school’s technology transfer office, or start their own company. Despite the fact that the academia-to-startup trend is hardly new, educational and federal institutions have not done the best job of tracking where their research ends up, according to a recent study.

Indiana University Professor of Economics David Audretsch says there are shortcomings in the way such entities track the commercialization of U.S. university research, and a glaring data gap is the number of scientists who have decided to use their work to launch startups.

More than one in four scientists funded by the National Cancer Institute who have obtained patents have also started a new firm, says Mr. Audretsch, who is also director of the Max Planck Institute of Economics in Germany.

Findings like this show that the university is more than a training ground for new scientists. It is also a direct contributor to the U.S. economy, Mr. Audretsch argues—and that’s important when Washington (i.e., American taxpayers) funds 57 percent of the total used for basic research.

With such a heavy outpouring of public financial support, “we, meaning society, expect new cures for disease to emerge from this research,” Mr. Audretsch says. “But we also expect new jobs, and even entire industries to also emerge to justify our massive investment in science and research.”

The study, completed for the Ewing Marion Kauffman Foundation in late March, evaluated the commercialization activity of university scientists by looking at grants given by the National Cancer Institute between 1998 and 2002. Mr. Audretsch and fellow researchers selected the largest 20 percent of those award grantees, or 1,693 scientists, to form the database used in the study.

The findings help to show that the current reliance on publicly available modes for measuring commercialization, like figures from the university Technology Transfer Offices, are outdated, say the study’s researchers. And such methods fail to capture the number of startups in which scientists are engaged.

“Our measurements of industry firms in the economy are relics invented and inherited from the old manufacturing economy that was based on factories, machines, and people working in assembly lines,” Mr. Audretsch says. “We’re only now beginning to uncover new measures that reflect the knowledge-based entrepreneurial economy.”