The effects of contexts on the acquisition of oral lexical ability in English as a foreign language

ABSTRACT

While there is ample evidence that study abroad (SA) enhances oral fluency in a foreign language, the effects of different types of learning context on other aspects of oral skills, such as vocabulary use, have not received much attention in academic research and are less clear. The present study tries to fill this void by investigating lexical development in oral production by advanced learners of English experiencing two different contexts of acquisition: formal instruction (FI) at home, followed by a 3-month stay abroad (SA). Speech samples were elicited from a group of 30 Catalan/Spanish undergraduates before and after each learning context by means of an oral interview and were later analyzed in terms of lexical diversity, sophistication, density and accuracy. Additionally, we examined baseline data from 25 native speakers of English, elicited through the same task. Our results reveal that both contexts enhance significant development in lexical accuracy and that SA proves especially beneficial for growth in adverb density, which moves towards target-like norm and adds fluidity to learners’ speech. FI, in contrast, causes a significant impact on lexical sophistication, as learners seem to acquire more specific vocabulary during classroom instruction.

KEYWORDS:

• Study abroad
• formal instruction
• vocabulary acquisition
• English as a foreign language
• second language acquisition

Acknowledgements

This research received financial support from the Spanish Ministry of Economy and Competitiveness [grant number FFI2013-48640-C2-1-P], the Agencia Universitària de Recerca (AGAUR) in Catalonia [grant number 2014 SGR 1563], and the three-year doctoral scholarship [grant number FI-DGR 2013]. The authors would like to acknowledge all the members of the SALA team and the students who have contributed to this multi-year project.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 TTR is considered to be highly sensitive to text length (i.e. total amount of words). As text length increases, TTR decreases complicating comparisons of samples in which text length varies considerably. Different alternatives have come into use to correct the text length factor, as mentioned in the text.

2 According to Biber, an information rich text contains nouns, attributive adjectives, prepositions, multisyllabic words, and a high TTR.

3 Error detection was performed with the help of a second coder to ensure reliability. An inter-rater reliability analysis was performed on 10% of randomly selected transcriptions using the Intraclass Correlation Coefficient (ICC), with a two-way fixed effects model and the confidence interval set at .95. Results indicated that inter-rater reliability was acceptable, at .809, showing a strong agreement between raters.

Additional information

Funding

This work was supported by Ministerio de Ciencia e Innovación: [Grant Number FFI2010-21483-C02-01/02,HUM2004-05442-c02-01,HUM2007-66053-C02-01/02]; Agència de Gestió d’Ajuts Universitaris i de Recerca: [Grant Number SGR2005-01086/2009-140].