By Cyril J. Weir (auth.)
Tests for the size of language talents needs to be developed in line with a coherent validity framework in keeping with the most recent advancements in concept and perform. This leading edge e-book, by means of a global authority on language checking out, offers with all key facets of language attempt layout and implementation. It offers a highway map to potent trying out in response to the most recent ways to check validation. A publication for all MA scholars in utilized Linguistics or TESOL, and for pro language teachers
Read or Download Language Testing and Validation: An Evidence-Based Approach PDF
Similar language arts books
New communications know-how has been a boon to educating and studying topics of English, from examining and writing to literature akin to Shakespeare. This booklet explores the ways in which info and communications expertise, or ICT, could be hired in instructing English and enriching the talents of scholars.
Best researchers within the box of spoken discourse and language educating supply an empirically educated, issues-based dialogue of the current kingdom of analysis into spoken language. They deal with a number of the advanced and worthwhile possibilities provided via those rising insights for language schooling and, particularly, for TESOL.
This publication presents a image of the sphere of language acquisition before everything of the twenty first Century. It represents the multiplicity of ways that signify the sphere and gives a evaluation of present issues and debates, in addition to addressing a number of the connections among sub-fields and attainable destiny instructions for study.
Extra info for Language Testing and Validation: An Evidence-Based Approach
The difficulties involved do not, however, absolve us from attempting to make our tests as relevant in terms of context as is possible; see Chapter 6 for ways in which we might address this. Bachman and Palmer (1996: 23) see test authenticity as ‘the degree of correspondence of the characteristics of a given language test to the features of a TLU (target language use) task’. The greater the fit, the more confidence we may have in relating test performance to likely real-life behaviour, given that the test task also meets the demands of the other validities referred to in this Part – for example, theory-based validity and appropriate cognitive processing.
There is a further serious question which might be asked of internal consistency estimates and which takes us back to the earlier reliability/validity debate. Would we want a high internal consistency estimate in all cases? Though one might be expected for a multiple-choice test of knowledge of structure or discrete lexical items in the 5,000-word list it might be rather naïve to assume that skills/strategies such as listening and reading are as unitary (see Urquhart and Weir 1998, Grabe and Stoller 2002, who argue for a partially divisible view of reading).
Given its problems, this type of reliability is seldom carried out and parallel forms reliability is the preferred choice. 2 Parallel forms reliability The administration of parallel (alternate) forms of a test in independent sessions provides us with alternate-form coefficients. , they should meet the same test specifications in every respect. 2 (Continued) on the same breadth of items and any input would be the same in length, degree of topic familiarity and difficulty level, etc. The tests would have been constructed to be equivalent but not identical.