Should a Student Need to Cross a Border to Get a Global Education?

The question 0:01

If only about one graduate in ten ever studies abroad, does everyone else miss out on a global education, or can most of it reach them at home?

This is WAC Signal, from the World Assessment Council. Evidence and ideas for the changing world of learning.

Episode eleven. Should a Student Need to Cross a Border to Get a Global Education?

The signal 0:34

For most universities, global education still means a passport stamp. Few students get one. According to the Institute of International Education's Open Doors data, two hundred and ninety-eight thousand one hundred and eighty US students earned academic credit abroad in 2023-24, up six percent on the year before. Measured against the students who finished a degree that year, that is about ten percent of US undergraduates over a whole degree.

In the UK the picture is starker. Universities UK International found that the share of graduating first-degree students who had been abroad fell from a peak of eight point eight percent in 2018-19 to three point five percent in 2021-22, mainly because of the pandemic and new visa rules for Europe after Brexit.

Policy is moving fast. In December 2025 the UK government confirmed it will rejoin Erasmus+ for the 2027-28 academic year, at a cost of about five hundred and seventy million pounds for 2027. The European Commission reports that Erasmus+ supported almost one point five million mobility participants in 2024, and it has proposed a fifty percent budget increase for the 2028 to 2034 program.

Meanwhile, virtual exchange, collaborative online international learning, known as COIL, and short blended programs are no longer pandemic stopgaps. Blended mobility is now written into Erasmus+ rules.

So if most students will never cross a border, how much of a global education can reach them at home?

The evidence 2:34

Start with what physical study abroad delivers. The best-known large study is the Georgetown Consortium Project, published in 2009. It tested one thousand two hundred and ninety-seven US students before and after, most in sixty-one programs abroad and a control group on home campuses. On the Intercultural Development Inventory, students abroad gained an average of two point three three points. The control students barely moved, at minus zero point zero seven.

So abroad beat home, but the gain was small and uneven. Only programs of thirteen to eighteen weeks produced a statistically significant gain. Longer programs plateaued, and students in four to seven week programs showed no significant gain. The most promising lever was not distance but design. Students who had on-site group mentoring often or very often showed the largest gains, though those small groups fell just short of statistical significance.

Where being abroad clearly won was spoken language. In the same study, students abroad made significantly greater oral proficiency gains than students learning the same languages at home.

Crossing a border creates an opportunity for intercultural learning. It does not guarantee that the learning happens.

Now careers. Here self-selection matters most. The European Commission's 2014 Erasmus Impact Study reported that graduates with international experience had an unemployment rate twenty-three percent lower five years after graduation. But the same study found that Erasmus students already scored higher on employer-valued personality traits before their exchange began. The economist Giorgio Di Pietro put it plainly, quote. It would be important to have more evidence that the perceived relationship is causal.

Studies that try to deal with selection give a mixed picture. Using exposure to exchange schemes as an instrument, Di Pietro estimated that Italian graduates who studied abroad were about twenty-three percentage points more likely to be employed three years after graduation, driven mainly by graduates from disadvantaged backgrounds. But a 2018 working paper by Schnepf and d'Hombres, matching Italian and UK graduates, found no significant short-term employment effect in either country. Three years out, the advantage for UK graduates was one point four percentage points.

The clearest career effect is on working abroad later. Studies cited by Di Pietro put it at fifteen percentage points for German students and eighteen to twenty-four points for Italian students.

And the raw UK gap is small. Universities UK International found that mobile graduates earned one point six percent more than graduates who stayed at home, and it states that its analysis cannot account for self-selection.

Who goes is skewed too. In the US, Open Doors reports that White students made up sixty-five point two percent of study-abroad participants in 2023-24, against about fifty-one percent of undergraduate enrollment in fall 2021, by our calculation from federal data. And most go briefly. Sixty-two point two percent, by our calculation, went for a summer or for eight weeks or less. Only two point four percent stayed a full year.

In the UK, the five-year average mobility rate was seven point two percent. For less advantaged students it was five point three percent. For Black students, four point seven percent. For mature students, three percent.

There is also a cost universities rarely count. Robin Shields estimated that the greenhouse gas emissions linked to degree-seeking international students were between fourteen point zero one and thirty-eight point five four megatonnes in 2014, double the 1999 level. Almost no university includes them in its carbon footprint.

Now the alternatives. Virtual exchange connects classes in different countries for sustained, facilitated online collaboration. The US-funded Stevens Initiative reached ten thousand eight hundred and ninety-eight young people in the United States and the Middle East and North Africa between summer 2021 and spring 2022. Of its US participants who answered in summer and fall 2021, seventy-eight percent had never taken part in an international exchange of any kind.

The stronger studies are smaller. At SUNY Brockport and The Hague University of Applied Sciences, one hundred and eight students were in classes randomly assigned either to COIL or to the same group task with classmates. Over a ten-week course, the US COIL group gained more in cultural intelligence than its control, with an effect size of zero point five eight. The Dutch students showed no difference, possibly because their control class had other international input. And the measures were self-report scales.

The tension 8:30

Almost all the virtual exchange evidence is self-reported, and the way the question is asked changes the answer. In the Stevens Initiative evaluations, run independently by RTI International, the standard before-and-after effect size for cross-cultural collaboration was zero point one three. When the same cohorts rated their past selves after the program, it was zero point five seven. That is more than four times larger, by our calculation, from surveys with no control group.

Direct measures can disappoint. Florida International University began phasing in a requirement in fall 2010 that undergraduates take at least two global learning courses. By December 2015 there were one hundred and sixty such courses across seventy of its seventy-two undergraduate departments. Self-reported global perspective scores rose. But on a direct writing task, seniors scored lower than entering students. The university concluded that seniors had little incentive to perform on the low-stakes test, and dropped it.

And the decisive comparison has not been run at scale. Our white paper found no large study comparing a well-designed virtual exchange with a well-designed semester abroad on the same outcome measures.

The implication 10:04

Crossing a border is not, by itself, a global education. Distance without design delivers small, uneven gains, and the career payoff shrinks once you account for who goes. But a video call is not a global education either. Immersion seems to matter for spoken language and for later international careers.

So the risk runs both ways. Virtual exchange is cheaper, and a finance office could treat it as a replacement for mobility rather than a way to widen access. If wealthier students keep the semester abroad while everyone else gets a video call, the gap in the outcomes that do need immersion could widen.

The WAC view 10:53

In our white paper, Global Without Leaving?, we propose the BORDER Test. Six questions, one per letter, decide goal by goal whether a learning outcome needs a border crossing.

B is for Benefit. Which learning outcome are we buying: knowledge, intercultural skill, language, or an international career? O is for Opportunity. Who is currently left out? R is for Requirement. Does the outcome need sustained immersion? D is for Design. Is there facilitation, mentoring and a real joint task? E is for Expense and emissions. What does each participant cost in money and carbon? And the final R is for Results. Will we measure the outcome directly, with a comparison group?

Applied to common goals, the test gives a default format. For knowledge of other cultures, COIL or virtual exchange embedded for credit. For intercultural development, blended or semester mobility with mentoring. For spoken language, a semester or longer abroad. For an international career, long mobility or a work placement. And for graduate employment at home, do not promise it.

The default for every student should be credit-bearing global learning at home. Mobility then becomes a targeted investment for the goals that need it, funded first for the students least likely to go.

The UK's Turing Scheme shows that targeting can work. As reported by Wonkhe, the share of Turing-funded students from disadvantaged backgrounds rose from thirty-nine percent in 2021-22 to fifty-six percent in 2024-25. But Department for Education figures show that higher education placements in 2024-25 were about thirty percent below plan, by our calculation.

For universities, we recommend three things. Make credit-bearing global learning at home, such as an embedded COIL module, a graduation expectation for every student. Fund short, mentored and blended options first for the groups least likely to go. And count the cost and carbon of each mobility place, and publish both.

What to watch 13:28

One. Whether the features that reached disadvantaged students under Turing, such as short group options, early promotion and extra funding, survive the UK's move to Erasmus+ in 2027-28.

Two. Whether anyone runs the missing test: independent comparative trials of virtual, blended and semester-long formats on the same outcome measures.

Three. Whether universities start measuring global learning with performance tasks and comparison classes, rather than retrospective self-ratings.

The sources for every figure in this episode are listed on the episode page, along with the full white paper, Global Without Leaving?, at worldassessmentcouncil.org.

WAC Signal is produced by the World Assessment Council, with AI-assisted research and AI voices, and every claim is checked against its source before publication.

This has been WAC Signal. Evidence and ideas for the changing world of learning.