Rising above their circumstances: what makes some disadvantaged East and South-East Asian students perform far better in science than their background predicts?

ABSTRACT

The Programme for International Student Assessment, carried out every three years by the Organization for Economic Co-operation and Development across a large number of countries and economies, have shown that socioeconomically disadvantaged students are almost three times more likely than advantaged students not to attain the baseline level of proficiency in science. Some of those disadvantage students beat the odds and perform better than expected according to their low socioeconomic background. They are called resilient students. Using data from 2015’s science-focused assessment and a logistic multilevel model analysis, this study examined the relationships between academic resilience and other non-cognitive skills measured by the assessment across seven East Asian countries and regions. Although there are significant disparities between the countries and regions, the results indicate that enjoyment and interest in science are positively related to science resilience. By contrast, when the student has an instrumental motivation for learning science (he or she is interested in science because it is useful for his or her career plans), the relationship is negative. This provides useful guidance for policymakers, educators, parents, and students on how to foster better Science results for students, and especially for disadvantaged students.

KEYWORDS:

• Resilience
• science attitudes
• PISA 2015
• multilevel logistic models
• East Asian students
• non-cognitive skills

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. TIMSS: Trends in International Mathematics and Science Studies.

2. ICT: Information and Communications Technology.

Additional information

Notes on contributors

Jose G. Clavel

José García Clavel is Profesor Titular de Universidad in the Area de Conocimiento de Economía Aplicada in the University of Murcia (Spain). Doctor since 1997 with a thesis on the application of Correspondence Analysis and Classification and Regression Trees, he has been visiting professor at The Ontario Institute for Studies in Education (Toronto, Canada), The Universitat Pompeu Fabra (Barcelona, Spain) and the Indira Gandhi Institute of Development Research (Mumbai, India). His research is mainly oriented in the application of multivariate techniques to the analysis of multivariate categorical data in fields that range from the International Accounting Standardisation to the Economy of Education.

Francisco Javier García Crespo

Francisco Javier García-Crespo is head of data analysis for the INEE. Graduated in mathematics, Assistant Professor in the Department of Statistics and Operational Research at the Complutense University of Madrid. He is TIMSS National Research Coordinator for Spain, PIRLS Data Manager and National Sampling Manager for Spain and TALIS Data Manager and National Sampling Manager for Spain. His research is mainly oriented in the application of multivariate techniques using databases from International Large-Scales Assessments.

Luis Sanz San Miguel

Luis Sanz San Miguel is Technical Advisor on data analysis at the INEE and honorary collaborator in the Department of Statistics and Operational Research of the Faculty of Mathematical Sciences of the Complutense University of Madrid. Doctor in mathematics since 1996 with a thesis on the relationships between Bayesian a frequentist approaches in testing statistical hypothesis. Member of the Bayesian Methods group and involved in projects with external funding, and also member of the UCM Animal Experimentation Ethics Committee of the Complutense University of Madrid. The main tasks that I develop include data analysis for Spanish international evaluations reports (PISA, PIAAC, TIMSS, PIRLS, TALIS), being the data manager for the Programme for the International Assessment of Adult Competencies. His research is oriented to statistical methods and applications, including Bayesian methods for multiple hypotheses testing and multivariate regression techniques.