Lexical assessment and lexical accommodation in oral examinations are new research dimensions, which have both theoretical and empirical values, however they are still much neglected. The present research aims to investigate: first, whether or not and how (if so) the measures of lexical richness can differentiate between candidates of three different grades of Graded Examinations in Spoken English of Other Languages (GESE) and whether or not those measures can differentiate good performers from poor performers at the same grade of GESE; second, whether or not and to what extent (if so) Chinese examiners accommodate to the candidates at the lexical level. 180 samples from Grade 2, 5 and 7 GESE were collected. All the data were transcribed into Codes for Human Analysis of Transcripts (CHAT) format for the Child Language Data Exchange System (CHILDES) (MacWhinney 2000) for analysis. First, the lexical measures of Token, Type, Guiraud, Guiraud Advanced (AG) and D of both candidates and examiners were obtained and analyses were conducted to investigate the relationship among them. Secondly, qualitative data were collected from interviews with GESE examiners to interpret the quantitative results. The quantitative results indicate that: 1) all the lexical measures can differentiate candidates of Grade 2 from Grade 5 and can differentiate candidates of Grade 2 from Grade 7 as well. However, there is no significant difference between Grade 5 and Grade 7 candidates' lexical variables. 2) In Grade 2 and Grade 5, all the candidates' lexical variables can distinguish between the qualified and poor performers of the same grade. Only Type, D and AG can differentiate between the qualified and poor candidates in Grade 7. 3) All the GESE score variables are correlated with each other, which shows a halo effect; the only GESE score variables that correlate with all candidate lexical variables in the pooled data is Focus. 4) The examiner variables cannot differentiate between qualified performers and poor performers in the same grade. 5) The only lexical variable that reflects the examiner’s lexical accommodation to the candidate is AG. The qualitative analyses indicate that the GESE examiners employ special characteristics in vocabulary assessment and the data also explain some of the quantitative results. It was found that the Chinese local examiners of GESE might apply meaningful and relevant input and the general communicative ability of the candidate as reliable overall rating strategies, and factors that affected the performance of the Grade 7 candidates are also discussed. The findings may not only shed light on a better understanding of the constructs of vocabulary knowledge and lexical richness, the accommodation the Chinese examiners conducted on candidates, but also provide insight into the design and improvement of examination procedures and training of Chinese oral examiners.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:632931 |
| Date | January 2014 |
| Creators | Zhang, J. |
| Publisher | University of the West of England, Bristol |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://eprints.uwe.ac.uk/24511/ |
Page generated in 0.0029 seconds