Edited by Eun Hee Jeon and Yo In'nami
[Bilingual Processing and Acquisition 13] 2022
► pp. 109–158
Chapter 5L2 writing and its internal correlates
A meta-analysis
This study examined the overall average correlation between second/foreign language (L2) writing performance and frequently investigated features of writing (i.e., writing-internal correlates). The correlates of L2 writing performance included objective measures of text features (syntactic complexity, lexical complexity, accuracy, fluency, and cohesion) as well as subjective measures (content, argument, organization, cohesion, coherence, grammar, vocabulary, language use, mechanics, and accuracy). A series of moderator analyses were also carried out for each type of objective measure to examine the effects of participants’ age, L2 proficiency, learning context, first language (L1) and L2 distance, task type, writing scoring method, and some characteristics of objective measures. By doing so, the present study aimed to identify key correlates of L2 writing performance and compare their relative importance. To this end, a total of 103 retrieved studies contributed 1,045 effect sizes based on 15,537 independent participants. The results for objective measures demonstrated that fluency had the strongest mean correlation with L2 writing performance (r = .570), followed by accuracy (r = .477), lexical complexity (r = .295), syntactic complexity (r = .271), and cohesion (r = .198). All subjective measure components had strong mean correlations with L2 writing performance (r = .668 to .927), but content and language use features had the strongest effects and cohesion and coherence features showed the least effects. Participants’ age, learning context, L1–L2 distance, writing scoring method, and some measurement characteristics were found to be significant moderators for certain components. The findings of this study have implications for L2 instruction suggesting that fluency and language accuracy of L2 writing should be promoted across the various developmental stages of L2 learners in various conditions, whereas lexical and syntactic competence should be more focused upon when instructing child/adolescent or low/intermediate L2 writers.
Article outline
- 1.Introduction
- 2.Background to the meta-analysis
- 2.1Review of objective measures
- 2.1.1Syntactic complexity
- 2.1.2Lexical complexity
- 2.1.3Accuracy
- 2.1.4Fluency
- 2.1.5Cohesion
- 2.2Review of subjective measures
- 2.3Review of moderator variables
- 2.3.1Age
- 2.3.2L2 proficiency
- 2.3.3Learning context
- 2.3.4L1–L2 distance
- 2.3.5Task type
- 2.3.6Writing scoring method
- 2.3.7Measurement types of complexity and fluency
- 2.1Review of objective measures
- 3.Research questions for the meta-analysis
- 4.Method
- 4.1Literature search and inclusion criteria
- 4.2Acceptable measures of L2 writing performance and internal correlates
- 4.3Coding the primary studies
- 4.4Research synthesis
- 5.Results
- 6.Discussion
- 6.1Objective measures
- 6.1.1Syntactic complexity
- 6.1.2Lexical complexity
- 6.1.3Accuracy
- 6.1.4Fluency
- 6.1.5Cohesion
- 6.2Subjective measures
- 6.1Objective measures
- 7.Conclusion
-
References
For any use beyond this license, please contact the publisher at rights@benjamins.nl.
https://doi.org/10.1075/bpa.13.05koj