Assessing the penalties of studying in a foreign language

Many students undertake at least part of their education in a language that is not their mother tongue. This is particularly true for university students.

Linguistically diverse countries – think about India – may use a language that is not the native one for most of the population as a medium of instruction for tertiary education. Moreover, international mobility is high. In the academic year 2019-20, the Erasmus+ programme of the European Commission, for instance, saw 237,800 university students undertake part of their courses abroad, despite the impact of COVID-19.

Add in the effect of international migration and the growing tendency in many countries to offer courses in English and the relevance of the issue becomes clear.

A difficult measurement

What is the impact on students’ performance of studying in a foreign language? It seems reasonable to assume that learning in a language that is not the mother tongue entails a cognitive cost, making it more difficult.

Having said that, it is not easy to provide a credible answer to the question. Let’s take the example of local students and international students attending the same course in the United Kingdom. Given the personal and financial costs involved in moving abroad to study, it may well be the case that the average international student is more motivated or more skilled than the average local student.

Skills and motivation are notoriously difficult to measure so any comparison may be flawed. Due to these underlying differences, it may well be the case that foreign students do better in exams than local students. This does not at all imply, however, that studying in English is easier for Chinese students than for British ones.

Exam performance

So, how to solve this issue? In an article recently published in the journal Labour Economics (in open access) we exploit some unique features of the Free University of Bozen-Bolzano, a university located in the northern Italian province of South Tyrol, an area where both Italian and German are commonly used. Most students attending this university have either Italian or German as their mother tongue.

Besides locals, many students come from the rest of Italy or from countries like Germany or Austria. Most of the degrees taught at the university are trilingual, meaning that a student can undertake courses (and corresponding exams) in Italian, German and English.

A student enrolled in an economics degree, for instance, may read introduction to management in English, macroeconomics in German and microeconomics in Italian, with each course offered only in one language.

We are therefore able to observe students with German or Italian as their mother tongue who are enrolled on the same degree courses, some of which happen to be in their mother tongue while others are delivered in a different language that, in some cases, is the mother tongue of their classmates, while in other cases it isn’t. We can then see whether the (mis)match between mother tongue and language of instruction affects performance, as measured by exam marks.

A large penalty

What we find is that the penalty for learning in a language other than your mother tongue is rather large, corresponding to a loss in grade points of approximately 9.5%.

In our analysis, in 29% of the cases there was a match between language of instruction and mother tongue. In 18% of the cases, students have a very high proficiency in the non-native language of the course – corresponding to C1 or C2 in the Common European Framework of Reference for Languages. For the remaining 53%, knowledge of the other language was intermediate, at B1 or B2.

What we found was that a high proficiency mitigates but does not eliminate the loss of grade points. We also found that the probability of not passing an exam is higher when there is a linguistic mismatch. So students doing a course that is not in their mother tongue are more likely to fail their exams and, once they succeed in passing, their final mark is considerably lower than it would have been otherwise.

Costs and rewards

Does this imply that it is a bad idea to study in a foreign language? Not at all. Economists like to frame education as an investment, with upfront costs (tuition fees, wages not earned, etc) and delayed returns (more opportunities in the job market, better health outcomes and so forth). Studying in a foreign language should be considered in a similar way.

As we document, it is indeed more challenging to study in a language other than your mother tongue. However, in addition to costs, there are also rewards. The extra effort may well be more than compensated for through the broader personal and professional horizons that mastering foreign languages implies. At the very least, someone willing to study in a language other than their mother tongue shows future employers that he or she is not afraid of challenges.

On the other hand, we should consider ways to better account for the penalty we document. If, for instance, getting a scholarship depends on exam marks or renewing a visa depends on the number of exams passed, then students who are not native speakers may be penalised. Given increased patterns of mobility, these issues may become increasingly relevant.

Juliana Bernhofer is honorary research fellow at Ca' Foscari University of Venice and research fellow at the University of Modena and Reggio Emilia, Italy. Mirco Tonin is professor of economic policy at the Free University of Bozen-Bolzano and director of the Research Institute for the Evaluation of Public Policies (IRVAPP) in Trento, Italy.