PREDICTORS OF TERTIARY LEVEL PERFORMANCE IN NON-ENGLISH SPEAKING BACKGROUND STUDENTS

Candidate: Richard Hewison, Dr. Elaine Chapman (Principal Supervisor) Prof. Tom O’Donoghue (Co-Supervisor)

Chapman, E., & O’Donoghue, T. Predictors of TERTIARY LEVEL Performance IN Non-English Speaking Background Students.

This research proposal, presented by Richard Hewison of the University of Western Australia, proposes to identify factors that predict academic performance of Non-English Speaking Background (NELB) international students in Australian tertiary education. The grounds for the purpose of the study reside with the myriad universities, which receive important financial contributions from international students, thus they have an obligation to identify the sources of the problems faced by students and to provide those students with appropriate and effective academic support. The research question designed for this topic asks, “to what extent do formalized English language Proficiency (ELP) scores act as predictors of academic performance in students’ first year of study in an Australian education system?”

The type of data necessary to answer the correlational research question consisted of demographic data, specifically relating to personal university profile; previous education; study stream and course units; university offer status; age; nationality; gender; and English Language Proficiency scores. The proposal also inquired for data about personal and psychological traits, and deeply held opinions and attitudes regarding self-efficacy, anxiety, beliefs about knowledge and learning, and personal learning styles and strategies. The former type of data were gathered from private records, with the data collection site being an unnamed, private tertiary education college specializing in business, information and mass communications courses in Western Australia. After extracted and coding the data into Excel spreadsheet files, the data was imported to SPSS for analysis, and a canonical correlation analysis was performed on the data with both the ELP levels and the demographic variables (age, gender, level of education previously attempted) entered as predictors. The latter type of data previously mentioned were proposed to be collected by sending out four modified surveys, the Foreign Language Classroom Anxiety Scale (FCLAS), Schommer scale, Approaches to Study Skills Inventory for Students (ASSIST) survey, and the Self-Efficacy Scale. A three-panel model path analysis and cluster analysis was the proposed method to analyze this data.

This research proposal includes all necessary aspects of a quality research proposal, as well as considering and potentially addressing all possible ethical issues. What impressed me the most was the extensive description of the potential limitations, the significance of the study, and he even including an estimated cost and timeline of the study, which is something I never would have thought to include.