ΔP as a measure of collocation strength

Abstract

This paper explores the proposed benefits of ΔP (delta P) as a measure of collocation strength. Its focus is on contrasting ΔP with other, more commonly used, association measures, particularly transitional probabilities, but also mutual information and Lexical Gravity G. To this end, first the strong correlation between ΔP and transitional probability is illustrated with the help of two exemplary corpora. This is followed by an analysis of hesitation placement in spontaneous spoken English, based on the assumption that hesitations will not be placed within strong collocations. Results show that, despite their strong similarity, in some contexts ΔP is more predictive of hesitation placement than transitional probability. Yet neither ΔP nor any of the other association measures emerges as the universally best predictor. On the basis of these results, it is suggested that studies should always rely on several association measures.

Promoting L2 acquisition of multiword units through textually enhanced audiovisual input: An eye-tracking study

“…Despite these limitations, our study provides further empirical support for the beneficial effect of TE in captioned audiovisual input (Cintrón-Valentín et al, 2019;Révész, 2018, 2020;Montero Perez et al, 2014, 2018, and extends the findings of previous studies by examining how TE can affect learners' visual processing of MWUs in captioned audiovisual input. It seems that the attention-raising effect of TE has the potential to increase the likelihood that MWUs are picked up.…”

“…Durrant and Schmitt, 2010). MI reflects how strongly two words attract by comparing their co-occurrence rate in a corpus to their co-occurrence rate by chance (Schneider, 2018). Semantic decomposability was rated on a seven-point scale by 33 learners from the same study program as the participants.…”

“…A number of studies have demonstrated that enhanced captions can increase the salience of unknown vocabulary in audiovisual input (Cintrón-Valentín et al, 2019;Montero Perez et al, 2014, 2018. For example, Montero Perez et al (2015) used vocabulary tests and eye-tracking to measure the relationship between learners' attention to, and learning of unknown vocabulary for two types of audiovisual input (full captions, keyword captions).…”

Phraseological Use and Development During a Stay Abroad: Exploring Sensitivity to Frequency and Cue Contingency

“…To do so, researchers used reference corpora to determine the strength of association for word combinations used by learners and then examined whether more strongly associated combinations were more probable after a period abroad. These studies relied on a variety of association measures, including the t‐score, MI score, log Dice score, and delta P score, all of which quantify “how strongly two events are linked” (Schneider, 2020, p. 251). The differences between and the relative merits of each of these measures have continued to be debated (e.g., Gablasova et al., 2017; Gries, 2013).…”

Co-occurrence Strength and Transitivity Effects on Spanish Clitic Case Variation With Reverse-Psychological Predicates

“…One important aspect of association measures is the extent to which they can represent cognitive aspects of language use. Researchers have probed into the cognitive validity of association measures by using them to successfully predict human behavior in a variety of linguistic tasks and phenomena ( Wiechmann, 2008 ; Ellis and Ferreira-Junior, 2009 ; Gries, 2013 ; Gablasova et al, 2017 ; Levshina, 2018 ; Schneider, 2020 ; Li et al, 2021 , among others).…”