Bottom-up decoding has a link to L2 proficiency

Listening is a major part of ELT, and in particular ELT materials, with a listening section consisting of comprehension questions and gap-fill questions being a standard part of any coursebook unit, though ostensibly used merely to present language to be used in a speaking stage (Ableeva & Stranks, 2013). Bottom-up listening skills are frequently overlooked in the listening section of coursebook units but may be hidden away in the pronunciation section. In this study, Wilson and colleagues (2011) make the case for teaching sound discrimination by linking it to the second language proficiency of 90-300 undergraduate students at a university in Japan.

Methodology

The sample size of 314 students was tested on their English ability with the TOEIC IP test, and then tested upon nonsense syllables of English phonology either consonant-vowel (CV) or vowel-consonant (VC). To assess vowel discrimination, different English vowels were paired with [p], and to assess consonant discrimination, consonants were paired with the vowel [a].The test was answered by selecting from 4 possible options. The scores were recorded for 314 students but this was narrowed down to 90 students for reasons of uniformity (op. cit, p. 2134) for statistical analysis. However, Wilson et al. (op. cit) refer back to the larger 314-student sample to make further observations and recommendations.

Results

After calculating two-tailed Pearson product-moment correlation coefficients for the results of the smaller sample of 90 students there were:

  • a medium effect sizes for overall sound and TOEIC IP score;
  • a medium effect size for consonant discrimination and TOEIC IP test score;
  • a medium effect size for overall sound discrimination and TOEIC IP reading score; and
  • a large effect size for consonant discrimination and TOEIC IP listening score. (op. cit, p.2135)

In addition, the larger group of 314 were analysed and “it was found that a group effect existed for vowels versus consonants (vowels were more difficult to discriminate than consonants), and that there was an interaction effect for position in the syllable. Thus, the most difficult sound to discriminate was the vowel in a VC syllable.” (op. cit, p.2136)

Discussion

These results could be used to show that there is a link between bottom-up skills and overall listening attainment, which should shock nobody, but it should give pause to material developers to pay more attention to bottom-up development. Should learners develop their receptive phonology they may stand a greater chance to raise their attainment in general based on the correlation between sound discrimination and reading scores; the authors believe “some learners are sounding out words when they read and knowledge of phonemes helps them to do so.” (ibid.)

It may also benefit learners if teachers spend more time on vowel instruction rather than consonant instruction seeing as these are difficult. However, it may also be the case that vowel discrimination may also be more difficult to instruct.

References

Ableeva, R. & Stranks, J. Listening in another language – research and materials, in Tomlinson, B. (ed.) (2013) Applied Linguistics and Materials Development. London: Bloomsbury.

Wilson, I., Kaneko, E., Lyddon, P., Okamoto, K. & Ginsburg, J. (2011) Nonsense-syllable Sound Discrimination Ability Correlates With Second Language (L2) Proficiency. ICPhS XVII Hong Kong: August 17-21, 2011.

marc
Marc Jones holds a MA Applied Linguistics & TESOL from University of Portsmouth and is currently studying a research degree with them too. He is interested in listening, phonology, second language acquisition, and Task-based Language Teaching.