ABSTRACT
The previous chapter reported the results of correlation analyses on the questionnaire responses from students in early–full and late–partial English medium instruction (EMI) schools. This chapter is concerned about EMI science students’ language choices, challenges, English skills, self-concept, and adopted strategies. More specifically, it focuses on the new components of each of the questionnaire scales, which were added to the dataset as new variables. The variables represented the differential loadings of the students’ responses to each of the questionnaire scales onto each component respectively. These factored variables were computed in order to determine whether there were any significant difference between the early–full and late–partial EMI schools, as well as to determine whether there was a possible relationship between them and among other ‘factored’ variables (e.g., students’ reflections on EMI) in the questionnaire items. The ‘factored’ variables were checked individually for normality of distribution. These variables were found to be normally distributed and as such were deemed appropriate for the use of parametric tests. Table 8.1 summarises the ‘factored’ variables of each of the questionnaire scales and the decision for the chosen statistical tests.
