Abstract

As part of a Department for Education and Skills initiative, Cambridge Assessment is developing a new suite of language assessments (Asset Languages) to be available at six stages of ability across a number of languages. These assessments are designed to be broadly comparable, in terms of functional level, across languages at each stage. Until recently, functional comparability of language qualifications has not received much attention in the UK, e.g. Coleman (1996, p 7) argues that the label ‘foreign language to degree level is meaningless due to discrepancies across British universities. Researching cross-language comparability is, however, becoming increasingly important, partially in response to Council of Europe initiatives, e.g. the Common European Framework of Reference (CEFR), which argues for levels that have shared meanings (Council of Europe 2001, p 5), and partially in response to concerns over existing UK qualifications, e.g. there was an investigation into the comparability GCSE grades in Modern Foreign Languages, 2005 (Halpin 2005).

This paper introduces a learner-centred mixed-methods comparability study focused on beginner to intermediate (A1 – B1) secondary school readers of German, Japanese or Urdu. Data from self-assessments and think-aloud protocols of students working through assessment tasks are used to compare 1) the difficulty of assessments, and 2) the differing ability profiles (both in functional can do terms and in the strategies/skills needed) of learners of these languages”.