The ‘promise’ of three methods of word association analysis to L2 lexical research

Abstract

The present study is an attempt to empirically test and compare the results of three methods of word association (WA) analysis. Two of the methods – namely, associative commonality and nativelikeness, and lexico-syntactic patterns of associative organization – have been traditionally used in both first language (L1) and second language (L2) associative research and the third one – collocational aspect of associative responses – is a more recent perspective on associative connections. The central assumption behind the study is that each method captures a different aspect of language users’ lexical organization and, in that, their findings may paint a different picture of how the L1 and L2 lexicons compare as proficiency increases. At the same time, the sensitivity of each method to reflect differences related to proficiency may also point to their potential as a research and assessment tool. Three equal size groups of participants ( N = 180) at different proficiency levels – native speakers (NSs), second language (L2) advanced and intermediate learners of English – completed a familiarity and a WA test in writing. The tests contained 36 items equated for lexical class (nouns, verbs, and adjectives) and frequency of occurrence (high, mid, and low). The participants’ WAs generated to familiar vocabulary were analysed in three different ways, following the methods of analysis practiced by each of the three traditions under investigation. In the main, the results showed that examining the lexico-syntactic patterns of associative organization as well as the collocational aspect of associative links are two more ‘promising’ ways of looking at WAs than examining them with respect to their nativelike associative commonality.

Multi‐ or Single‐Word Units? The Role of Collocation Use in Comprehensible and Contextually Appropriate Second Language Speech

“…The argument here echoes other cross‐sectional findings that more advanced L2 learners produce more low‐frequency collocations with higher mutual information scores during L2 writing tasks (Durrant & Schmitt, ; Garner et al., ) and word association tasks (Clenton, ; cf. Zareva & Wolter, ). (See Siyanova‐Chanturia & Spina, and Tavakoli & Uchihara, for related studies.…”

Word association research and the L2 lexicon

“…There is no significant correlation between similarity to L1 norms and grammar monitoring test scores (Kruse et al 1987) ! Similarity to L1 norms could distinguish between L1 speakers and learners, but could not detect differences in proficiency level (Zareva et al 2005;Zareva & Wolter 2012)  WEIGHTED and UNWEIGHTED scoring systems o Scoring systems that account for the position of a response on L1 norms lists (weighted scoring) are not significantly more sensitive to learner proficiency than those that don't (unweighted scoring) (Kruse et al 1987;Wolter 2002)  Reliability of WA measures o There is a moderately strong test-retest correlation for number of responses produced and more modest correlations for stereotypy (Kruse et al 1987)  Receptive knowledge of WA (as determined by the word associates format (WAF) approach, whereby participants identify items that are associates of the cue/target) o Rasch analysis indicates that the WAF test is reliable (high person separation reliability) (Read 1998) o…”

“…if they are using their L2 (Norrby & Håkansson 2007;Zareva 2007;Håkansson & Norrby 2010) ! if they are using their L1 (Fitzpatrick 2006;Fitzpatrick & Izura 2011) o if they are learning their L2 as a foreign rather than second language (Norrby & Håkansson 2007) o the less proficient they are (Söderman 1993;Orita 2002;Zareva 2007;Zareva & Wolter 2012;Khazaeenezhad & Alibabaee 2013) o if the cue words are adjectives (Nissen & Henriksen 2006)  The likelihood that a participant will produce a synonym response is increased:…”

Comparing Dictionary-induced Vocabulary Learning and Inferencing in the Context of Reading

“…The term feature is defined in the connectionism literature as a specific connection among units or, in cognitive terminology, a specific pattern of neural activation over a number of neurons (Westhoff 2004, Moonen et al 2006, Moonen et al 2014. Every word can be regarded as a union of various features that determine it in a unique way (Zareva and Wolter 2012). And networks involving numerous and varied types of features are more likely to become activated (Moonen et al 2006).…”