Accuracy, syntactic complexity and task type at play in examination writing
A corpus-based study
This chapter explores the association between syntactic complexity and syntactic accuracy in essays written by Russian learners of English in reply to two examination task types: a description of graphical material (Task 1) and an opinion essay (Task 2). A Poisson regression model served to predict the number of syntactic errors. Two syntactic complexity parameters were statistically significant in predicting syntactic accuracy in both tasks: the numbers of sentences and adverbial clauses. Three more parameters predicted the accuracy in Task 1 only: maximum depth of syntactic trees, and the numbers of adjective + noun and noun + infinitive constructions. Six parameters were related to syntactic accuracy in Task 2: the numbers of all clauses, of tokens and of T- units; the average length of sentence; and the numbers of coordinated and of participle + noun constructions.
Article outline
- 1.Introduction
- 2.Literature review
- 3.Research questions, material and method
- 3.1Research questions
- 3.2Examination task types
- 3.3Corpus data
- 3.4Data analysis
- 3.4.1Evaluating syntactic accuracy
- 3.4.2Parameters of syntactic complexity
- 3.4.3Statistical analysis
- 4.Results
- 4.1Accuracy (error rates) across the two task types
- 4.2Syntactic accuracy
- 4.3Association between accuracy and complexity for syntactic features
- 4.4An aggregated metric of syntactic complexity
- 5.Discussion
- 6.Limitations and further research
- 7.Conclusions
-
Acknowledgements
-
Notes
-
References
-
Appendix