King's College London

Research portal

Confidence in PISA : Validating an international assessment of student self-efficacy in mathematics

Student thesis: Doctoral ThesisDoctor of Philosophy

Student participation and attainment in mathematics is an issue in education
policies and practices for many countries. Some academics, teachers and policy
makers claim that student attitudes in mathematics exercise a decisive causal
influence on participation and attainment in mathematics. To ensure that
education policies and practices are well-evidenced, it is important that
assessments of these attitudes are sufficiently validated for such claims. The
influential Organisation for Economic Co-operation and Development (OECD)
Programme for International Student Assessment (PISA) includes assessments of
student self-confidence in mathematics. This involves two sets of questionnaire
items, with one set relating to self-efficacy and the other set relating to selfconcept. While the self-concept items refer to the mathematics domain as a whole, the self-efficacy items refer to individual mathematics tasks. Self-efficacy has been incorporated in models of self-regulated learning and, although one model indicates that self-efficacy may be significant for formative assessment, there is disagreement about relations between self-efficacy, participation and attainment. Although there have been several studies of mathematical self-efficacy, their assessments have generally lacked sufficient validations. The OECD validation of the PISA mathematical self-efficacy assessment itself lacks transparency and requires more evidence. The present study provides an independent validation of the assessment and uses mathematical self-concept as a point of comparison. The validation has a mixed methodology integrating evidence from the PISA 2003 data set and documentation for 41 countries with cognitive interviewing of students in England, Estonia, Hong Kong and the Netherlands.The validation identifies threats to validity that weaken extrapolation from the assessment to real-world situations, particularly in Hong Kong. These threats are traced to the formulation of the mathematics tasks in the self-efficacy items. The findings have implications for the interpretation and development of mathematical self-efficacy assessments as the basis for understanding relations with participation and attainment.
Original languageEnglish
Awarding Institution
    Award date2014

    View graph of relations

    © 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454