Stephan, an economist at Georgia State University in Atlanta. She led a global email-based survey of scientific mobility—the
largest ever at 17,000 people—published in
Nature Biotechnology in 2012. However, she
cautions that ORCID is an incomplete and
biased sample of the world’s scientists (see
sidebar, right). Until researchers account
for those quirks, or enough scientists fill
out their ORCID profiles, it can’t be relied
on as a definitive picture of scientific migration. Even so, it reveals some surprising
patterns and highlights individual stories,
like Kodzius’s, that otherwise would have
Even the basics about the global scientific workforce are surprisingly hard to nail
down. Many countries do share data about
their scientists, and the United Nations
pulls those streams together into semiannual reports on global science. As of
2015, the global head count comes to
8 million scientists. One in five is in an EU
country, and 17% and 19% are in the United
States and China, respectively.
How many of each country’s research
workforce are immigrants? In the United
States, more than a third of doctoral degrees
in the sciences and engineering are awarded
to foreigners on temporary visas, according
to a report last year by the National Science
Foundation. But after the Ph.D., that data
trail goes cold, Stephan says.
The global picture of migration has been
Genome scientist Rimantas Kodzius
is one of the most migratory
scientists alive. He has worked in
10 countries, most recently
Saudi Arabia (pictured here).
NEWS | FEATURES | HUMAN MIGRATIONS
20% 15% 5%10% 0%
1960 1992 2015
P h . D . s18,000
ORCID wasn’t intended as a massive
longitudinal survey of human migration,
but with 3 million profiles and growing,
it is becoming just that. So far, 25%
of the world’s researchers have
added personal information to their
public ORCID profiles, including the
years, locations, and descriptions of
their education and employment.
This data set has biases. ORCID
users skew young, and certain countries
are over- and underrepresented.
The raw data sets are available at
ORCID users voluntarily have created
a giant and growing database.
People with more recent Ph. D.s are
overrepresented in the data set, reflecting
its recent growth and the fact that younger
researchers are signing up for ORCID more
quickly than older ones.
ORCID is not a random sample of the world’s
scientists. Countries are not equally represented,
as revealed by a comparison with 2013 figures
from the UNESCO Science Report.
even more elusive. Building it requires in-
formation about all scientists around the
world year after year. Yet that’s just what
ORCID could eventually offer.
ORCID’s original purpose was simply
to help scientists with common names—
such as Michael Roberts or Wei Wang—get
credit for all their publications by giving
them unique identity codes, says Laurel
Haak, ORCID’s executive director, who is
based in Bethesda, Maryland. The organization maintains a website through which
users can add information to their profile,
including education and employment history. Since its launch in 2012, the number
of ORCID profiles has grown explosively.
So far, 741,867 of the 3 million ORCID users
have chosen to use their online profile as a
public CV that chronicles their education
and work history, and spans up to several
decades for the most senior researchers.
Tracing ORCID users’ countries of residence over time makes it possible to retrospectively trace each person’s migration
pattern. Supermigrators like Kodzius stand
out in the data. (See the cover of this issue for a map of the paths that Kodzius
and other globe-trotters have taken.) With
roughly 10% of the world’s scientists geographically tracked in ORCID, interesting
patterns emerge that no other data set
For example, the data suggest that about
30% of the scientists who got their Ph.D.
in the United Kingdom now live elsewhere,
whereas the same is true for only about
15% of scientists who received their Ph.D.s
in other EU countries (see graphic, p. 692).
The data also suggest an effect of the
11 September 2001 terrorist attacks: a slump
in the migration of foreign scientists to the
United States. Overall, the ORCID data
show that the number of foreign researchers studying or working in the United States
has grown smoothly since 1990, because
the global pool of ORCID researchers has
grown steadily. But there’s one glaring exception: The number of foreign researchers
immigrating to the United States plateaus
in 2002. Further analysis reveals that the
annual rate of immigration to the United
States dropped by about 15% and did not
recover until 2008 (see graphic, p. 692).
Was that slump due to post-9/11 chaos in
U.S. immigration and tougher visa requirements? If so, that could be one of the attack’s longest-lasting costs: thousands of
the most highly skilled workers avoiding
the United States for years.
Causation will be hard to pin down, however, says Kirk Doran, an economist at the
University of Notre Dame in South Bend,
Indiana. In addition to global economic
forces that cause labor markets to ebb and