Collocations in Corpus‐Based Language Learning Research: Identifying, Comparing, and Interpreting the Evidence

Abstract

This article focuses on the use of collocations in language learning research (LLR). Collocations, as units of formulaic language, are becoming prominent in our understanding of language learning and use; however, while the number of corpus‐based LLR studies of collocations is growing, there is still a need for a deeper understanding of factors that play a role in establishing that two words in a corpus can be considered to be collocates. In this article we critically review both the application of measures used to identify collocability between words and the nature of the relationship between two collocates. Particular attention is paid to the comparison of collocability across different corpora representing different genres, registers, or modalities. Several issues involved in the interpretation of collocational patterns in the production of first language and second language users are also considered. Reflecting on the current practices in the field, further directions for collocation research are proposed.

Assessing Phraseological Development in Word Sequences of Variable Lengths in Second Language Texts Using Directional Association Measures

“…Gablasova et al. () highlighted three important aspects of corpus‐based distributional information for RMCs: exclusivity, dispersion, and directionality. Exclusivity concerns the statistical significance of the lexical associations in RMCs; dispersion concerns the evenness of the RMC distribution in a corpus.…”

“…The most intuitive metric for formulaicity is RMC frequency in a native speaker corpus. However, studies have shown that RMC frequency “may not be the best predictor of the regularity and predictability in language use” (Gablasova et al., , p. 160). Its statistical importance may be subject to the relative frequencies of the individual words in the combination (Evert, ; Gries, ).…”

“…It can be statistically adjusted to more effectively reflect the directional statistical association by taking into account the conditional probability of a word in the absence of the cue, which is referred to as delta P (Ellis, 2006;Gries, 2013). Delta P has been well received as an effective statistical measure for associative learning for linguistic cooccurrences (Ellis & Ogden, 2017;Gablasova et al, 2017;Gries, 2013;Wahl, 2015) and behavioral data (Ellis & Ferreira-Junior, 2009). Delta P is more advantageous than transitional probability in modeling directional associative learning from a cue to an outcome (Ellis & Ferreira-Junior, 2009).…”

“…Phraseology has received central attention in research on language learning and processing (Gablasova, Brezina, & McEnery, ; Wood, ). Psycholinguistic and experimental studies have provided strong evidence for second language (L2) learners’ sensitivity to distributional properties of linguistic structures (Ellis, ), including not only words (Diependaele, Lemhöfer, & Brysbaert, ) but also multiword units, ranging from semantically opaque sequences, such as idioms (Conklin & Schmitt, ; Siyanova‐Chanturia, Conklin, & van Heuven, ) and collocations (Durrant & Schmitt, ; Wolter & Gyllstad, ), to compositional multiword combinations (Hernández, Costa, & Arnon, ).…”

“…Similarly, L2 phraseological acquisition has been analyzed by observing a trend of increasing formulaicity in the RMCs used by learners across proficiency levels (Bestgen & Granger, 2014;Durrant & Schmitt, 2009). Gablasova et al (2017) highlighted three important aspects of corpus-based distributional information for RMCs: exclusivity, dispersion, and directionality. Exclusivity concerns the statistical significance of the lexical associations in RMCs; dispersion concerns the evenness of the RMC distribution in a corpus.…”