A Multiword Unit Analysis : COCA Multiword Unit List 20 and ColloGram

Abstract

Multiword units receive attention as being an essential part of vocabulary knowledge that will expedite the learning of an L2. However, there is lack of a graded multiword units (MWU) list that can offer direct applications for pedagogy, syllabus design or materials development. The current article aims to report on the development and evaluation of COCA (Corpus of Contemporary American English) Multiword Unit 20 (COCA_MWU20) where the notion of MWU family is adopted. The compilation involved selecting and grading the MWU by grammatical well-formedness, range, and frequency. The 10,000 MWU families which made up the list were utilized in the development of ColloGram, a list-based MWU family analysis program. The COCA_MWU20 can be expected to help L2 learners increase their knowledge of lexical items beyond single words and ColloGram is the first of the kind that can analyze multiword items based on COCA. For researchers, the ColloGram can also be a tool for identifying MWU that may appear in a target text.

Processing of English Formulas by L1, L2, and Bilingual English Speakers Based on Formulaicity and Formula Type

“…This finding accrues supportive evidence that high-frequency idioms are more likely to be represented in the L2 mental lexicon as morphologically complex words than their low-frequency counterparts (Siyanova-Chanturia et al, 2011) Our findings also indicate that collocations induced greater processing challenges for L2 speakers than L1 speakers or bilinguals. Currently, knowledge of collocations has been considered as a component of L2 lexical competence (Forster, 2001;Shin & Chon, 2019). One study found that even advanced EFL learners made more errors with collocations in a written translation task than they did with verb+noun collocations (e.g., serve a sentence, Bahns & Eldaw, 1993).…”

The Role of Congruency in Collocation Acquisition: A Case Study of Vietnamese Students Learning English Adjective + Noun Collocations

“…Regarding frequency, Hong, Rahim, Hua, and Salehuddin (2011) specify that the minimum FREQ of 5 tokens in the British National Corpus (BNC) proves sufficient to be the standard threshold for a word combination to be considered a collocation. Meanwhile, Shin and Chon (2019) point out that 20 repetitions or above should be the minimum cut-off point for frequency in the Corpus of Contemporary American English (COCA) due to its relatively large size of more than one billion words, which is ten times as many words as in BNC (p. 611). Since this research used COCA as a reference source to identify collocations, the operational definition of collocation is constructed as shown below:…”

Language Proficiency and Knowledge in Adjective-Noun Collocations: A Case Study of Vietnamese Learners of English

“…According to Wolter and Yamashita (2015), corpus linguists tend to use "a minimum MI-score threshold of 3 as indicative of significant co-occurrence" to examine how closely related two words are (p. 1201). As regards frequency, while Hong et al (2011) set the minimum FREQ of five tokens in BNC (the British National Corpus) as the standard threshold for a group of words to be considered a collocation, Shin and Chon (2019) points out that 20 repetitions prove sufficient as the minimum cut-off point for frequency in COCA due to its relatively considerable size (with more than one billion words of data, as compared to 100 million words in BNC) (p. 611). As COCA was used in the current study as a reference source for identifying collocation, the operational definition for the current study is formed as follows:…”