2006The Journal of the Korea Contents AssociationRequires access

Cluster Reduction by Korean EFL Students: Insertion vs. Deletion Strategies

Mi-Hui Cho

Open publisher page 0 citations

Abstract

Motivated by the fact that cluster reduction strategies such as inserting a vowel or deleting a consonant in resolving English complex clusters differ depending on studies, this paper investigates the repair strategies employed by Korean EFL students. A total of 60 college students participated in the present study and the participants' production of English voiceless word-initial and word-final clusters was measured using the materials designed for this study. It has been shown that prosodic positions such as onset and coda and the number of cluster sequences influenced cluster reduction strategies. The error rates of both insertion and deletion were noticeably higher in the coda position than in the onset position and both insertion and deletion error rates were higher in triconsonatal cluster than in biconsonantal cluster sequences. Overall, the insertion rate was higher than the deletion rate. However, the deletion rate was significantly higher than the insertion rate in triconsonantal coda cluster sequences. Because of this, the deletion rate was higher than the insertion rate for triconsonantal cluster sequences across onset and coda. Also, the high deletion rate of triconsonantal coda clusters contributed to the high deletion rate for the coda clusters in general.

About this research paper

What this paper is about

Motivated by the fact that cluster reduction strategies such as inserting a vowel or deleting a consonant in resolving English complex clusters differ depending on studies, this paper investigates the repair strategies employed by Korean EFL students. A total of 60 college students participated in the present study and the participants' production of English voiceless word-initial and word-final clusters was measured using the materials designed for this study. It has been shown that prosodic positions such as onset and coda and the number of cluster sequences influenced cluster reduction strategies. The error rates of both insertion and deletion were noticeably higher in the coda position than in the onset position and both insertion and deletion error rates were higher in triconsonatal cluster than in biconsonantal cluster sequences. Overall, the insertion rate was higher than the deletion rate. However, the deletion rate was significantly higher than the insertion rate in triconsonantal coda cluster sequences. Because of this, the deletion rate was higher than the insertion rate for triconsonantal cluster sequences across onset and coda. Also, the high deletion rate of triconsonantal coda clusters contributed to the high deletion rate for the coda clusters in general.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Motivated by the fact that cluster reduction strategies such as inserting a vowel or deleting a consonant in resolving English complex clusters differ depending on studies, this paper investigates the repair strategies employed by Korean EFL students. A total of 60 college students participated in the present study and the participants' production of English voiceless word-initial and word-final clusters was measured using the materials designed for this study. It has been shown that prosodic positions such as onset and coda and the number of cluster sequences influenced cluster reduction strategies. The error rates of both insertion and deletion were noticeably higher in the coda position than in the onset position and both insertion and deletion error rates were higher in triconsonatal cluster than in biconsonantal cluster sequences. Overall, the insertion rate was higher than the deletion rate. However, the deletion rate was significantly higher than the insertion rate in triconsonantal coda cluster sequences. Because of this, the deletion rate was higher than the insertion rate for triconsonantal cluster sequences across onset and coda. Also, the high deletion rate of triconsonantal coda clusters contributed to the high deletion rate for the coda clusters in general.

Key concepts: Coda, Consonant cluster, Cluster (spacecraft), Reduction (mathematics), Word (group theory), Speech recognition, Word error rate, Vowel

Related papers

Back to paper searchBrowse research topicsOriginal source
Cluster Reduction by Korean EFL Students: Insertion vs. Deletion Strategies — Research Paper | ScholarLens