This chapter discusses how questionnaire-based research can be implemented in the English Medium Instruction (EMI) contexts. It presents an empirical study which that examined Chinese EMI university students’ attitudes and motivation (i.e., integrative and instrumental orientations) toward learning content subject knowledge in English. An EMI scale adapted from Gardner’s (2004) Attitude/Motivation Test Battery (AMTB), which takes the format of Likert-point scale as explained in Chapter 3 of the this book, was validated and administered to 541 EMI students from three Chinese universities. The validity and reliability of the scale were measured, the correlations of the three dimensions (i.e., attitudes, integrative orientation, and instrumental motivation) were tested, and the role of demographic variables (i.e., gender, level of study, disciplinary background) in EMI attitudes and motivation were explored. The research findings suggest the validity and reliability of the scale, the positive correlations among the three dimensions, and the different degrees of EMI attitudes and motivation between male and female students and between soft science and hard science students. The researchers argue that questionnaire-based research is appropriate for the EMI contexts, but its effectiveness can be enhanced if the mixed methods design is adopted.

This chapter explores the use of one-way between-groups multivariate analysis of variance (MANOVA) to investigate the effects of gender on perceived English language ability challenges in an English Medium Instruction (EMI) university in Hong Kong. Contrasting with ANOVA which evaluates mean differences on one dependent variable, MANOVA evaluates if there are mean differences on two or more dependent variables. The two dependent variables used were perceived writing challenges and perceived speaking challenges. The independent variable was gender. We checked for univariate and multivariate outliers, homogeneity of variance-covariance matrices and normality and found no deviation from the standard parameters. The results show a significant difference between males and females in terms of the combined dependent variables, F (2, 75) = 3.444, p = 0.037, Wilks’ Lambda = 0.913 and partial eta squared = 0.087. When the dependent variable is analysed separately with reference to the Bonferroni adjusted alpha level of 0.025, a significant difference is only found between males and females in terms of the perceived speaking challenges, F (1, 75) = 6.658, p = 0.012, partial eta squared = 0.084. A closer examination of the mean score shows that females (M = 3.042, SD = 0.34) perceived speaking English to be more of a challenge than males did (M = 2.71, SD = 0.55). The findings are discussed in relation to teaching and learning in an EMI university.

Grammatical complexity has been considered as an important research construct closely related to second language (L2) writing development. Although theoretical models were developed to demonstrate what grammatical complexity is, few studies have been conducted to analyze how this construct is represented from an empirical perspective. This chapter presents a data-driven investigation on the representation of grammatical complexity with an exploratory factor analysis (EFA). The investigation is based on (1) a corpus of scientific research reports written by Hong Kong students in an English Medium Instruction (EMI) scientific English course, and (2) an EFA, which is a statistical approach to uncover an underlying structure of a phenomenon, which fits this research purpose well. A corpus has been built with the science writing from EMI undergraduate students in Hong Kong. After corpus cleaning, Second Language Syntactic Complexity Analyzer – a software – was applied to output the values of fourteen effective measures of grammatical complexity for running the EFA in SPSS, and a step-by-step instruction was described in the chapter. The final model includes three latent factors: clausal (subordination) complexity, nominal phrasal complexity, and coordinate phrasal complexity. This EFA model is generally consistent with the argument of investigating grammatical complexity as a multidimensional construct (Biber et al., 2011; Norris & Ortega, 2009). In the end, we highlighted the research and pedagogical implications that readers should pay attention to when the EFA is applied in other EMI contexts in the future.

The use of English Medium Instruction (EMI) for teaching content subjects has gained popularity worldwide. There has been a significant interest in exploring different aspects of research in the EMI context, focusing on language and content learning, teaching disciplinary language, evaluating the effectiveness of EMI implementation at various educational levels, and investigating the perspectives of teachers, students, policymakers, and other stakeholders regarding the use of English for instruction. Within this complex landscape, researchers come from diverse backgrounds and orientations. The use of English as a medium of instruction and the exploration of teaching content subjects through English have garnered attention from researchers in various disciplines such as science, technology, mathematics, engineering, medicine, business, history, and sociology. In their recent publication, McKinley and Rose (2019) presented a comprehensive handbook on research methods in applied linguistics, covering the rapid changes in this field. They identify two main types of researchers in applied linguistics: “researcher-practitioners,” who focus on teaching and learning within language classrooms, and “educational psychologists,” who study the cognitive and psychological processes of language learning (2019, p. 1). These classifications are particularly relevant for EMI research, as they underscore the need for an interdisciplinary approach that can address the multifaceted nature of teaching and learning in EMI contexts. The diversity of research orientations represented by these classifications further emphasizes the imperative for a holistic approach to EMI research.

Covariance-based SEM (CB-SEM) has become one of the most prominent statistical analysis techniques in understanding latent phenomena such as students and teachers’ perceptions, attitudes, or intentions and their influence on learning or teaching outcomes. This chapter introduces an alternative technique for SEM, variance-based partial least squares SEM (PLS-SEM), which has multiple advantages over CB-SEM in several situations commonly encountered in social sciences research. A case study in the English Medium Instruction (EMI) context is also demonstrated as an example to facilitate comprehension of the method. The chapter concludes with a discussion of potential applications for other EMI-related contexts and lines of inquiry.

Despite the enduring popularity of path analysis, there has been limited research in the context of English Medium of Instruction (EMI) to illustrate established theories. Moreover, researchers have yet to incorporate statistical data to refine the theoretical models and better elucidate the causal relationships between various factors that potentially influence students’ academic achievement. To fill this gap, this study aims to develop and analyze a well-fitted model that could account for contingent links between variables that directly and indirectly affect EMI students’ academic achievement in science. Drawing on survey data from eight EMI secondary schools in Hong Kong, the current study identified interplayed roles of students’ English proficiency, language use in science classroom, self-perceived English difficulty in the science classroom, and self-concept on science learning on science achievement by using path analysis – one of the structural equation modeling (SEM) models, which is also illustrated in Chapter 5 of the book.

English Medium Instruction (EMI) research has highlighted the transition from secondary schools to EMI higher education as a critical stage that shapes students’ learning behaviors and perceptions. However, longitudinal studies that draw on a quantitative design to outline students’ patterns of academic development during the transition period are scarce. While investigating students’ language-related academic difficulties, previous research has predominantly treated students as a unitary cohort without exploring the disparities that may arise from individual difference variables. This chapter reports an empirical case study that adopts a longitudinal quantitative design to identify patterns of change in students’ perceptions of lecture listening difficulties during their first semester transitioning into an EMI university in China. The study also explores variations in these patterns that are associated with students’ English listening proficiency upon entry. Key methodological procedures for designing and implementing the study will be introduced, while along with sharing suggestions for handling the typical challenges of attrition and missing data in longitudinal quantitative research. The chapter concludes with methodological implications of the study for EMI research, and offers suggestions for future research based on a critical reflection of the study’s limitations.