Language proficiency evaluation Raters Henry Emery PRICESG Linguistic

Language proficiency evaluation: Raters Henry Emery PRICESG Linguistic Sub-Group

Session structure • • • Define terms Operational and linguistic raters The need for adequate training Rater competencies A mechanism for international standardisation of language proficiency assessment: The Association of Language Raters

An examiner: • • delivers a direct language test is trained in a specific test format elicits speech interacts with the candidate/candidates (an interlocutor) • may examine and rate simultaneously

A rater: • • • assesses language proficiency applies the Rating Scale to speech requires familiarity with the test format may examine and rate simultaneously may rate ‘live’ speech may rate a pre-recorded sample

• Who is an operational expert? • What is the role of an operational expert? “to assess operational integrity and technical accuracy” (ICAO doc. 9835) • When should an operational be employed? “test prompts should not be intended to evaluate specific technical knowledge concerning operations” (http: www. icao. int/icao/en/trivia/peltrg. FAQ. htm#27) • In a test of plain English is there a role for an operational rater? • Operational experts: tests which aim to assess both phraseology and plain English proficiency

Operational raters • • • subjective impression ‘pass’ or ‘no pass’ operational integrity technical accuracy overall effectiveness Holistic Descriptors ? Who has the final say, the linguistic or the operational expert?

Linguistic raters • more complex judgement • judge components of performance separately • less subjectivity • justify assessment • identify specific aspects of performance • analytic Rating Scale

The complexities of the ICAO Rating Scale Six key features

Feature 1 • The six operational levels of the rating scale are not of equal intervals • Levels three and four are particularly broad

Level descriptors for Structure 4 Errors in basic grammatical structures rarely interfere with meaning 3 Errors in in basic grammatical structures frequently interfere with meaning 2 Limited control of a few simple memorized grammatical structures and sentence patterns 1 Performs below the elementary level

Feature 2 • A candidate is unlikely to fully ‘fit’ a descriptor at any level • A candidate’s speech may demonstrate features of different levels at the same time

Level descriptors for Interactions 6 Interacts with ease in nearly all situations. Is sensitive to verbal and non-verbal cues, and responds to them appropriately 5 Responses are immediate, appropriate, and informative. Manages the speaker/listener relationship effectively 4 Responses are usually immediate, appropriate, and informative. Initiates and maintains exchanges even when dealing with an unexpected turn of events. Deals adequately with apparent misunderstandings by checking, confirming, or clarifying

Feature 3 • A candidate with a ‘flat’ profile across the six profiles in the Rating Scale will have a ‘jagged’ proficiency

Extracts from level 4 descriptors Vocabulary … sufficient to communicate effectively. . . occasional loss of fluency. . . does not prevent effective communication Errors. . . (in basic structures) … rarely interfere with meaning Pronunciation, stress, rhythm, and intonation … sometimes interfere with ease of understanding

Feature 4 • Pronunciation is notoriously difficult to assess • What does good pronunciation of a second language mean? • Standardise comprehensibility? “Can we be certain that two or more. . . speakers will find the utterance of a foreign speaker equally comprehensible…? ” (Harris 1977: 81)

Feature 5 • Degree of ‘overlap’ between profiles • Linguistic behaviours described in one profile appear in other profiles • Lack of proficiency in one profile will affect performance in another

Feature 6 • Rating scale includes both ‘positive’ and ‘negative’ descriptors Positive descriptor: Can often paraphrase successfully. . . Negative descriptor: . . . may not vary speech flow. . .

Positive descriptors Level 6 Increase in positive descriptions moving from Levels 1 -6 Level 5 Level 4 Level 3 Level 2 Decrease in negative descriptions moving from Levels 1 -6 Level 1 Negative descriptors

Reliable and accurate rating requires thorough training in the application of the ICAO Rating Scale It is not enough to say “ I can apply the Scale”. Raters must demonstrate this ability

Linguistic Rater competencies: • Qualified and experienced teacher of English to speakers of other languages? • Experience in language assessment? • Experience in teaching English for aviation? • Native speaker/certificate of language proficiency? • Background in aviation operations? • Familiar with R/T phraseology?

Standardisation: Making level 4 • Intra-rater reliability • Inter-rater reliability: a. same test format b: different tests c: international standardisation • Standardisation of rating procedure

ICAO Document 9835, Section 6. 7. 4 Raters must • understand the criteria and the context in which the criteria occur • agree upon the standardised procedures for the implementation of the criteria • inspire trust and gain the confidence of the community and the candidates

ICAO Document 9835, Section 6. 7. 4 “A body of well-informed and experienced raters must be formed”

ALR The Association of Language Raters

Speech samples are gathered from organisations around the world Regional ALR representatives (aviation SMEs and linguists) convene to: • rate the speech samples and write rationales for the ratings • organise the rated speech samples into ‘assessment packs’ • produce an association ‘Code of Practice’ • formalise guidance material for language raters

Language Rater Accreditation Service • A prospective language rater applies to ALR for Rater Accreditation • The Association sends an assessment pack to the rater • The rater rates speech samples contained in the assessment pack • If the ratings agree with the ALR standard, an official Certificate of Accreditation is issued

Funding: o Governmental and commercial aviation organisations o Accreditation Service Fees ICAO Executive committee Association of Language Raters (ALR) ALTE ILTA Cyclical presidency Regional Representatives (Language Raters and Operations Personnel) Team Leaders (In-country) Local Rater S T A N D A R D I S A T I O N F E E D B A C K Regional Representatives (Language Raters and Operations Personnel) Team Leaders (In-country) Local Rater

Rater Accreditation • Certificates valid for two years • On expiry, the rater must re-apply for ALR accreditation • ALR-accredited raters for assessments for licensing purposes

Regional ALR representatives will convene every two years to • Rate fresh speech samples for ‘assessment packs’ for the new accreditation period • Review existing policy for rater accreditation • Review feedback gathered by Team Leaders • Produce a published bi-annual report for the international aviation community

Long-term activity • Language Test Accreditation • Inspection and auditing of language testing organisations • Training programs • Regional seminars

ALR The Association of Language Raters Comments and questions: henry@maycoll. co. uk