Given the various differences between learners, teachers, and instructional methods in English Medium Instruction (EMI), a common purpose of EMI research involves investigating the potential variation between groups. The analysis of variance (ANOVA) test is a common technique used to address such a research aim, as it tests whether there are significant differences between the means of different groups. This chapter introduces the ANOVA test to readers by highlighting how it has been used in research within the field of EMI. To illustrate how different forms of the ANOVA test can be employed, the chapter then provides two case studies: (1) the use of a one-way between subjects ANOVA to examine the differences between three groups of students with respect to their perceptions of the role of English in their academic and career goals and comprehension level of EMI courses; and (2) the use of a mixed ANOVA in a quasi-experimental study that examined the differences in pre- and posttest writing performance and academic motivation of two groups provided with different types of feedback. Each of the case studies summarizes the assumptions required for the use of ANOVA, discusses potential problems that may face EMI researchers, and introduces alternative procedures.

To develop human capital in the globalized world, governments have implemented policies that require the teaching of some or all school subjects in English. However, the implementation of this policy has faced criticism and challenges in some countries where linguistic and cultural diversity is prevalent. These challenges include school segregation based on the medium of instruction, inadequate English proficiency of teachers and students and less interaction in English Medium of Instruction classrooms. Some researchers have investigated these challenges through international assessments such as the Programme for International Student Assessment and Trends in International Mathematics and Science Study. Others have examined EMI mathematics and science classrooms qualitatively through observations and interviews. These studies showed that teachers and students are more comfortable, and the classroom environment is more interactive when they use their mother tongue. In general, the findings favoured the mother tongue education for both cognitive and noncognitive variables. Researchers recommend either switching to EMI after achieving a certain level of English proficiency or providing language support for students who are already in EMI systems. Finally, a case study from Wales suggested that providing questions in both the mother tongue and English might mitigate unfair linguistic advantages in international assessments.

This chapter explores the design, development, and format of the Likert-type scales and response categories used in an online questionnaire for quantitative data collection for a recent empirical case study exploring attitudes, challenges, and perceptions of first-year undergraduate students at an English Medium Instruction (EMI) university in Hong Kong. Questionnaires are among the most widely used methods for research in the social sciences and can be an important and valuable source of data, which can be converted into measures of the numerous variables being examined. A variety of rating scale formats and designs with differing numbers of response categories and sequences are used in survey research. While researchers are typically confronted with a surplus of design and format choices, there is often little in terms of research, guidelines, or standards directing them toward which styles and formats to choose. Based on the survey design and development for this recent EMI-related study, and drawing from the literature, this chapter reviews how such choices and decisions were made, how the Likert-type scales were designed, and how these decisions may have influenced the overall success of the data collection and analysis. The case studies in Chapters 7, 8, and 11 of this book also adopt Likert-type scales in their questionnaire design, and these could be read together to supplement the understanding of the current chapter.