Centre for Research in Language & Literacy (Vocabulary Studies) Workshop: Extracting, and listing collocations is unpalatable: how it become the compass?

Extracting, and listing collocations is unpalatable: how it become the compass?

Student-led workshop: Rana Hanash

This talk will focus the lens on the methodological aspects of my research. This research aims to both explore and explain the use of verb- noun, adjective- noun collocations in the academic writings of multiple-cases of L2 undergraduates in years one and three. During the exploration phase, both the British Academic Written English corpus (BAWE) and the British National Corpus (BNC) were employed as reference corpora to generate pure academic lists, unique hard lists, and unique soft lists of both verbs and nouns, as well as adjective and noun collocations. MI, MI2, and Delta-P, Log-dice as association measures were used to extract these lists. These lists were examined and compared. Due to its comprehensively, log-dice was used to extract collocation lists from both BAWE and BNC for this study. BAWE was used to generate the initial lists of hard and soft collocations. Overlapped collocations among the lists were defined as non-specialised lists. These non-specialised lists were examined against the extracted v-n and adj-n collocations from the parsed version of the non-academic corpora in the BNC in order to create the pure academic lists which were the unique pairs not present in the BNC. In addition, the hard and soft lists of collocations identified within the BAWE, with the overlapping items were removed to have the unique hard and soft elements only. The extracted lists of collocations found in the learner’s corpora then examined against these three reference lists: The pure academic, unique soft, and unique hard both v-n and adj-v collocations. Thus, given this opportunity, harvesting the insights and results offered by these compass lists and delving into the process of constructing this methodology is the goal of this strand.