6 Assessing productive and interactive skills Personal refl ection

Task-specifi c versus generic scales
First, scales may be task specifi c (designed to partner a single writing 
or speaking prompt) or generic (designed to work in relation to a 
class of tasks of a certain type). The ‘class of tasks’ could imply the 
range of tasks included on one test, the range of tasks that might be 
undertaken in a language programme, or a very wide range of tasks, 
including tasks undertaken in different languages. This is the case for 
the ACTFL Profi ciency Guidelines (American Council for the Teaching 
of Foreign Languages, 1999) or the scales included in the Common 
European Framework of Reference for Languages (CEFR) (Council 
of Europe, 2001).
An advantage of the task-specifi c approach is that the scales can 
communicate the features of a successful response very clearly and in 
some detail to the rater and to learners, perhaps helping to guide 
learners on how to improve performance. However, task-specifi c 
scales have two serious shortcomings. First, every time a new prompt 
is devised – an essay about music replacing an essay about sports, for 
example – a new scale is also required. This makes task-specifi c scales 
expensive and impractical to develop. Second, it is diffi cult to see 
what such scales can tell us about more general abilities than the 
ability to carry out the one specifi c task to which they relate. In 
profi ciency assessment, they might show an employer that an assessee 
is good at one very clearly delineated function, but may not provide 

---

146