This article provides a systematic account of collocational use in Chinese EFL learners’ oral production and explores some of the issues involved, by adopting a corpus-based error analysis approach. The distribution of six types of collocational errors extracted from two sizeable Chinese learner English corpora shows that verb-noun collocations pose the greatest difficulty for Chinese learners. An exploration of the correlation between the learners’ English proficiency and collocational performance finds that the learners’ knowledge of collocation has not developed alongside their knowledge of vocabulary in general. Our further descriptive and diagnostic analyses of verb-noun errors indicate that 1) Chinese EFL learners have the greatest difficulty with the verbs when using verb-noun collocations; 2) the learners’ use of nouns is also not satisfactory; 3) due attention should be paid to the inappropriate use of the non-lexical elements (prepositions and articles); and 4) the main causes of verb-noun collocational errors include L1 transfer, assumed synonyms, overgeneralization and misselection of the target word. It is suggested that university English teaching in China should attach more importance to the examination and diagnosis of collocational errors and also integrate learner-centered, corpus-based methods into vocabulary teaching.

This research describes an attempt to establish a pedagogically useful list of the most frequent formulaic sequences for engineering undergraduates who need to read the textbooks of their fields in English. The Engineering English Formulae/Formulaic Sequences List (EEFL) was derived from a corpus containing 4.57 million tokens of one hundred college textbooks across twenty engineering subjects. In consideration of formulae for widespread use and pedagogical relevance, a series of criteria were applied. Comparable to a list of 1,000 high-frequency individual words, a total of 1,000 two- to six-word sequences were selected into the EEFL and they accounted for 25.73% of the running words in the Engineering Textbook Corpus. The EEFL, not highly technical in nature, contains the most commonly-used multi-word units traversing the subfields of the engineering domain and engineering majors may encounter these word sequences very often. For matriculating engineering students, the present EEFL and the EEWL (Engineering English Word List common to engineering subjects) may be mutually complementary in providing a pathway to the engineering register and may be helpful for ESP teachers without a background of science and technology when preparing engineering English teaching materials for an EST course required by most engineering-related departments.