Mid Sweden University

This study aimed to investigate Swedish EFL learners’ use of delexical verb+noun collocations and to compare the collocational performance of students’ studying year one and year three in upper secondary education. The study primarily found that the students use delexical VN collocation proficiently, even though they underuse collocation in a general sense. Moreover, the frequency and proficiency of the students’ use of collocations developed only slightly between the year one and year three students, which is explained by the notion of a ‘collocational lag’. Furthermore, the results showed that the students resort to ‘lexical teddy bears’; resulting in overuse with certain delexical verbs and underuse with others. Additionally, the study found that the students incorrectly transfer from their L1 when using delexical VN collocations. This is especially evident in the overuse of TAKE (Swedish TA ) and HAVE / MAKE , and the confusion the students’ display with DO and MAKE (Swedish GÖRA ).

In terms of pedagogical implications, the students chiefly need instruction with particularly troublesome areas, such as the distinction between HAVE / MAKE . Furthermore, the students would benefit from being made aware of their tendency to overuse delexical TAKE, which could contribute to the students using collocations in a more rich and varied manner.

For future research, I would suggest conducting a similar study, with similar definitions, and incorporate an analysis of a comparable TL corpus, such as The Louvain Corpus of Native English Essays (LOCNESS), to directly compare the ULEC data to the TL corpus data. Rather than comparing the result with other researchers’ investigations of TL corpora, this juxtaposition would give a clear picture of how the learners perform. Furthermore, I suggest that additional studies are carried out on other types of collocations, such as adjective + noun collocations, in order to determine how problematic delexical VN collocations are in relation to other types. Further studies like these would help to give us a broader picture and a greater understanding of learners’ collocational competence and could bring us important pedagogical implications.