Investigation of Native Speaker and Second Language Learner Intuition of Collocation Frequency

Research into frequency intuition has focused primarily on native (L1) and, to a lesser degree, nonnative (L2) speaker intuitions about single word frequency. What remains a largely unexplored area is L1 and L2 intuitions about collocation (i.e., phrasal) frequency. To bridge this gap, the present s...

Full description

Saved in:
Bibliographic Details
Published inLanguage learning Vol. 65; no. 3; pp. 533 - 562
Main Authors Siyanova-Chanturia, Anna, Spina, Stefania
Format Journal Article
LanguageEnglish
Published Hoboken Blackwell Publishing Ltd 01.09.2015
Wiley-Blackwell
Wiley Subscription Services, Inc
Subjects
Accuracy
collocation frequency
Collocations
Comparative Analysis
Computational Linguistics
corpus
Correlation
Foreign language instruction
Form Classes (Languages)
Frequency of Occurrence
Intuition
Italian
Italian as a Second Language
Italian language
Judgment
L2 learners
Language acquisition
Language Research
Native Speakers
Nonnative Speakers
Nouns
Phrase Structure
Second Language Learning
Task Analysis
Word Frequency
Online AccessGet full text
ISSN0023-8333
1467-9922
DOI10.1111/lang.12125

Cover

More Information
Summary:Research into frequency intuition has focused primarily on native (L1) and, to a lesser degree, nonnative (L2) speaker intuitions about single word frequency. What remains a largely unexplored area is L1 and L2 intuitions about collocation (i.e., phrasal) frequency. To bridge this gap, the present study aimed to answer the following question: How do L2 learners and native speakers compare against each other and corpora in their subjective judgments of collocation frequency? Native speakers and learners of Italian were asked to judge 80 noun‐adjective pairings as one of the following: high frequency, medium frequency, low frequency, very low frequency. Both L1 and L2 intuitions of high frequency collocations correlated strongly with corpus frequency. Neither of the two groups of participants exhibited accurate intuitions of medium and low frequency collocations. With regard to very low frequency pairings, L1 but not L2 intuitions were found to correlate with corpora for the majority of the items. Further, mixed‐effects modeling revealed that L2 learners were comparable to native speakers in their judgments of the four frequency bands, although some differences did emerge. Taken together, the study provides new insights into the nature of L1 and L2 intuitions about phrasal frequency.
Bibliography:Appendix S1: Cohen's K Test Statistics Examining the Agreement Between Native Speakers' (NS) and Non-Native Speakers' (NNS) Judgments and Corpus Frequency Values for the Four Collocation Frequency Bands. Appendix S2: Verbatim English Translation of the Instructions Accompanying the Questionnaire.
ark:/67375/WNG-L0HKSBK2-V
istex:339B1713AD7139BB6EB25022F190AA0164D94450
Research Establishment Grant (Victoria University of Wellington)
ArticleID:LANG12125
http://www.iris‐database.org
This research was supported by the Research Establishment Grant (Victoria University of Wellington) to the first author. We wish to thank Phil Durrant and the three anonymous reviewers for their very helpful comments on the earlier draft of this paper, as well as Paul Warren, Harald Baayen, and Stefan Gries for their advice with regard to some of the statistical analyses. Any inaccuracies are, of course, our own. The elicitation instruments used for this study can be accessed by readers in the IRIS digital repository
.
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0023-8333
1467-9922
DOI:10.1111/lang.12125