Predictive validity in the IELTS test: A study of the relationship between IELTS scores and students' subsequent academic performance


Researchers:

Mary Kerstjens

Caryn Nery

Date Published:

14th April 2000

This study investigated the relationship between the IELTS test and academic outcomes. Specifically, it sought to determine the extent to which the IELTS test predicts the subsequent academic performance, as well as the language difficulties, of international students enrolled in an Australian tertiary institution. It also aimed to investigate whether any of the individual tests of Listening, Reading, Writing and Speaking was critical to academic success. This was researched in three ways: through statistical correlations between students' IELTS scores and grade point average, student questionnaires and staff interviews.

The IELTS scores of 113 first-year international students from the TAFE and Higher Education sectors of the Faculty of Business of an Australian university were correlated with their first-semester grade point average (GPA). In the total sample, significant correlations were found between the Reading and Writing tests and GPA (.262, .204 respectively). When Higher Education and TAPE scores were looked at separately, only the Reading score remained significant for the Higher Education group. While none of the correlations was significant in the TAPE group, the magnitude of the correlation between the Writing test and GPA (.194) was very similar to that for the total sample, which was statistically significant.

Regression analysis found a small-to-medium predictive effect of academic performance from the IELTS scores for the total sample and the Higher Education group, accounting for 8.4% and 9.1% respectively, of the variation in academic performance. The Reading test was found to be the only significant predictor of academic performance in the total sample and Higher Education group. IELTS was not found to be a significant predictor of academic performance for the TAFE group.

The qualitative data on students' and staff perceptions of the relationship between English language proficiency and academic performance corroborate to some extent the statistical findings, particularly in relation to the Reading, and to a lesser extent, the Writing tests and the skills they represent. However, while the Listening test was not significantly correlated to academic performance, students and staff from both TAFE and Higher Education highlighted the importance of listening skills in first-semester study. Both staff and students were generally positive about students' ability to cope with the language demands of their first semester of study. Aside from language, staff also saw sociocultural and psychological factors such as learning and educational styles, social and cultural adjustments, motivation and maturity, financial and family pressures to have an influence on the academic outcomes of international students in their first semester of study.