Standard vocabulary tasks such as multiple-choice tasks or cued recall tasks can contribute little to our understanding of the nature of Receptive and Productive vocabulary. The main reasons for this include difficulties with task construct definition and the difficulties in ascertaining which mental processes are needed for task completion. The State Rating Task (SRT), described in detail in this thesis, was developed as a plausible alternative to standard vocabulary tasks for assessing Receptive and Productive vocabulary. The SRT, which is based on Multi-State models of vocabulary acquisition, is designed to assess the interaction of Understanding (listening and reading) vocabulary and Use (speaking and writing) vocabulary in a self-report task. The task requires a subject to rate a word into one of five knowledge States. The SRT was tested in two experimental situations. In the first, the SRT was used to assess the types of tokens produced in a stem completion task, and in the second, the SRT was used to trace the developmental patterns of Understanding and Use vocabulary over time. The results from 7 experiments show that the SRT has a high degree of concurrent validity and reliability, and suggest that the SRT is a plausible alternative to standard vocabulary tests for assessing vocabulary development.

This study aimed to estab.lish the influence, or not, of negotiated scoring and gender in a high stakes General Practice examination assessing communication skills. The dataset comprised 1024 simulated consultations. Final Year medical students undertook VOICEs - a six-station OSCE-style examination. Two stations involved role played consultations. Aspects of communication ~ including attitudes - were scored by nego~iation between a clinician and the role player, using holistic descriptive statements. Scorers' independent perceptions and agreed scores were collected. to analyse influence. Also, participants' gender (examiner, 'patient' and student) was recorded. Other demographic/performance-related variables were included and coded for analysis. Data were initially analysed using established statistical testing, but later analysed using Generalised Linear Modelling, to probe questions outstanding from the original analyses. Results showed the examination process to be internally consistent, suggesting good reliability. The majority of null hypotheses relating to bias were upheld. Student gender was persistently significant in individual tests, but not in GLM. The role players' contribution did not bias the examination, and in some cases was more consistent that the GP scoring. Results relating to question content were not significant, but did raise interesting questions about managing uncertainty and the relationship between communication (skills) and content (knowledge/expertise).

Yes/No tests have become well-established as a method for assessing the vocabulary size of first and second language learners (Anderson & Freebody, 1983; Meara & Buxton, 1987). However, despite a general acceptance of their usefulness, no evaluation studies had ever been conducted of the tests. The results from two experiments showed quite conclusively that the corrected scores from five Yes/No tests were unreliable as measures of vocabulary size. The results also revealed that responses to real words alone were the single best predictor of vocabulary knowledge. Three further experiments were therefore conducted using Yes/No tests constructed from real words which had been selected from different corpora. The results from the Rasch analysis showed that all the test scores were extremely reliable and the application of the Rasch model to the data was valid. The results also indicated there is a close relationship between the frequency of a word and its difficulty, but only a moderate one between Yes/No test scores and scores from tests of vocabulary knowledge and general proficiency. It is concluded that using the revised Yes/No format and Rasch analysis together will result in very accurate estimates of vocabulary size. It is suggested that if Yes/No tests are used in conjunction with other measures of vocabulary knowledge, they may help to define more clearly how the lexicon develops.

This dissertation studies the interface between second language testing (SLT) and second language acquisition (SLA) as it applies to a general problem in language use: noted discrepancies between standardized test scores and local placement of University of Balamand (UOB) entrants. A series of empirically based studies were designed to highlight those factors which account for interlanguage as manifested in important aspects of verb grammar: choice of tense, use of tense sequences, subject-verb agreement, and irregular verb morphology, and to compare students' interlanguage(s) in that particular area of grammar with high versus low scores in target language English as measure of their proficiency. A Case Study was carried out on a low-intermediate proficiency level student as determined by UOB entrance test. Analysis was made of the student's choice of relevant grammatical forms in Lebanese-English interlanguage compared with native language Lebanese-Arabic, both in the domains of narration and description. A similar analysis was made of the student's use of the same grammatical forms on the placement exam task. Analysis of metacognitive strategies was also undertaken. Results showed that use of each grammatical form varied with the task rhetorical mode in the sense of converging with! diverging from what is taught. More importantly, the student's low proficiency was only reflected in the narration task, not in the description or placement exam taskThe Main Study was then undertaken on a sample of 28 students from the four DOB English proficiency levels. Primary data consisted of the students' writing on three tasks of different rhetorical modes: narration, exposition, and comparison-contrast. Secondary data consisted of verbal reports by the students, and comments by linguists and expert reviewers. The same method of the Case Study analysis was used. Quantitative results showed that the narration task had the most discriminating power; the comparison-contrast task had the least. A statistically significant difference was found between second language English and native language Arabic accuracy rate of students' use of grammatical structures. Qualitative results showed that both high and low proficiency groups lacked control over tense and tense sequences, but the high proficiency group had more control over 3rd person singular' -s' . In terms of SLA, the empirical study of verb grammar here presents a clearer understanding of multi competence. Consideration is given to the interface between SLT and SLA for the practical problem of finding the right type of test that would be a valid predictor of entrants' proficiency in English as a second language, thus leading to their placement at the right English language level.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.