Show Less

Measuring Oral Proficiency through Paired-Task Performance


Ildikó Csépes

This book intends to provide a theoretical overview of examining candidates’ oral abilities in different examination formats. In particular, it explores specific partner effects on discourse outcomes and proficiency ratings in the framework of paired-task performance. Two investigations, both set in the context of a proposed Hungarian school leaving examination in English, were carried out in order to contribute to a better understanding of the assumed impact of the chosen variables. Study One investigates discourse differences between examiner-to-examinee interaction and peer-to-peer interaction. Study Two explores the impact of the peer partner’s proficiency. The results show that partner effects related to this variable cannot be predicted as either harmful or beneficial since no statistically significant difference was found between 30 candidates’ scores in different examination conditions.


Show Summary Details
Restricted access

9. Discussion and Conclusion 147


9. Discussion and Conclusion We have explored several issues in relation to the validity of the paired exam format within an experimental school leaving examination model for Hungarian secondary school students. Study One focused on the conversa- tional analysis of the interactional features of the paired mode as opposed to those of the individual mode, while Study Two examined three main issues: • the degree to which candidates' proficiency scores are affected by the partner's proficiency level; • the impact of previous learning on candidates' proficiency scores; • test takers' perceptions of the paired oral exam. We will summarise the most important results from both studies and discuss the implications of the findings from the point of view of language testing and teaching in Hungary. Then we will consider limitations of the two studies as well as directions for further research. Implications for language testing research will also be suggested. 9.1. Summary and implications of the findings The main research question in Study One focused on identifying dis- course differences produced in guided role-play depending on whether the in- terlocutor was an examiner or a peer candidate. The analysis explored whether there was equal distribution of power and discourse roles between the exam- iner and the candidate and how the candidates' proficiency level impacted on the discourse outcomes. Based on conversational analysis of six performance samples (three in the paired and three in the individual mode), the study brought to light that there were indeed differences in the discourse produced in the two...

You are not authenticated to view the full text of this chapter or article.

This site requires a subscription or purchase to access the full text of books or journals.

Do you have any questions? Contact us.

Or login to access all content.