English-taught courses tank student performance, new study finds
A large-scale experiment at a Swedish university found students learning programming in English scored significantly lower and dropped out more than peers taught in Swedish. The finding challenges a widespread assumption in global higher education that language of instruction doesn't affect learning outcomes—a key issue for institutions expanding English-medium programs to attract international students.
Originaltitel: English-medium instruction and impact on academic performance: a randomized control study
<p>Stakeholders and researchers in higher education have long debatedthe consequences of English-mediuminstruction (EMI); a key assumption of EMI isthat student’s academic learning through English should be at least as good aslearning through their first language (usually the national language). This studyaddressed the following question: “What is the impact from English-medium instructionon students’ academic performance in an online learning environment?”“Academic performance” was measured in two ways: number of correctlyanswered test questions and through-put/drop-out rate. The study adopted anexperimental design involving a large group (n = 2,263) randomized control studyin a programming course. Student participants were randomly allocated to anEnglish-medium version of the course (the intervention group) or a Swedishmediumversion of the course (the control group). The findings were that studentsenrolled on the English-medium version of the course answered statisticallysignificantly fewer test questions correctly; the EMI students also dropped outfromthe course to a statistically significantly higher degree compared to studentsenrolled on the Swedish version of the course. The conclusion of this study is thusthat EMI may, under certain circumstances, have negative consequences for students’academic performance.</p>