When research in 2011 showed that workers in the fields of science, technology, engineering, and mathematics earned a premium of 25 percent over other workers and have just a 5.5-percent unemployment rate, it reinforced strong economic incentives to get more people into those STEM fields.
But research like that might soon become more difficult to conduct. That’s because the U.S. Census Bureau wants to stop asking people in a key national survey about their field of study.
Since 2009 the bureau has collected data on people’s undergraduate fields of study as part of its American Community Survey, at the urging of the National Academy of Sciences and the National Science Foundation. The survey, which is based on responses from more than three million households, is used to collect demographic information and keep track of trends between the decennial censuses.
On this blog, we’ve previously written about how underemployment hit recent graduates and that ever-present question: “Is your college degree still worth it?” Without the field-of-study data from the American Community Survey, that research would have been nearly impossible to compile.
Before 2009 the National Science Foundation’s National Center for Science and Engineering Statistics conducted its own surveys. But the center’s director, John R. Gawalt, called that process “inefficient” and said it was difficult to find people in STEM fields and to track them over time.
Adding the field-of-degree question to the Community Survey reduced costs and improved what the NSF could do because the Census statistics allowed researchers to dive deeper into the data, especially for those employed in science and engineering fields.
Elimination of the question would leave researchers with “no data sources large enough to study small but crucial academic specializations,” said Mr. Gawalt, noting that the data provide context for efforts to focus on underrepresented populations in the STEM fields, including women, members of minority groups, and people with disabilities.
Higher-education organizations have also weighed in on the proposed change. A newsletter from Georgetown University’s Center on Education and the Workforce called the Community Survey’s question “the only information that students, parents, and educators can rely on to understand the economic benefits of individual college majors.”
The Georgetown researchers are not alone. Here are some other places where the data have been used:
The Hamilton Project, which describes itself as “an economic-policy initiative at the Brookings Institution,” has found that the typical college graduate will earn $1.19-million in today’s dollars. Without the field-of-degree data, it would not have been able to show that a botany major who is 33 years into his or her career is likely to earn $61,000, compared with an engineering major at the same point, who would make $109,000.
Similarly, the Federal Reserve Bank of New York would have no data to analyze for its report on “recent college graduates finding good jobs.”
Some employers might use a report from the Association of American Colleges & Universities to consider how much they might pay a liberal-arts major, but they wouldn’t have a reliable data set to look at without these data.
And the Census Bureau’s own parent agency, the U.S. Commerce Department’s Economics and Statistics Administration, has used the data to take a more in-depth look at the effects of STEM degrees on diversity in STEM fields.
The division chief of the American Community Survey Office, Jim Treat, says a cost-benefit analysis was used to make the recommendation that the question be dropped. In an email to The Chronicle, Mr. Treat wrote that his office had looked at the perceived burden of including each question and had surveyed federal agencies to see which data were used.
If the Census Bureau were to drop the question, Mr. Gawalt said, it would not be impossible for the NSF to run its own survey again, but doing so wouldn’t be cheap. Before the Census added the field-of-degree question, he said, the NSF spent an estimated $17-million to compile the information, and the data had less detail than the American Community Survey provides. The NSF’s elimination of its survey resulted in savings of $4-million every other year.
The field-of-study question isn’t the only one being considered for elimination. The Census Bureau also may remove several questions about marriage history.
The bureau is inviting public comments on the proposed changes through December 31.
