A Study Of Syntactic Complexity In The English Writing Of Non-English Majors

Syntactic complexity refers to varied and complex syntactic structures in writing. It is regarded as an important index of the development of second language writing. Researchers have identified clause, T-unit, and sentence as indices to measure L2 learners'progress in syntactic complexity. Clauses include independent clauses and dependent clauses which are subdivided into adverbial clauses, relative clauses, and nominal clauses. T-unit as an index of syntactic complexity first developed by Hunt to test learners'syntactic maturity, is now believed to be an independent clause embedded with adverbial clauses, adjective clauses or nominal clauses. Few researches has been conducted on syntactic complexity at home. It is thus necessary to investigate the development of syntactic complexity in English writing by Chinese college students. This study aiming at systematic exploration of Chinese non-English majors'main linguistic features in writing and their writing ability is conducive to the establishment of a comprehensive system of evaluating college students'writing performance.This study carried out a systematic analysis of the syntactic features in 120 English argumentative compositions. It was mainly aimed at testing whether a L2 learner's writing is becoming more complex in syntax as their L2 proficiency develops. It also attempted to explore whether non-English majors in different terms were significantly different in terms of measuring syntactic complexity in their writing. In addition, the author tried to find out whether L2 learners'syntactic complexity and diversity were closely related to the holistic evaluation.A total of 30 freshmen and 30 sophomores majoring in Applied Mathematics, Chemical Engineering and Technique, Chinese and Chinese Literature, and Preschool Education, from Huainan Normal University, Anhui participated in this study from January to July, 2010. Instruments employed in this study included two writing exams and measure of 20 variables concerning syntactic complexity and diversity. Two raters were involved in evaluating the sample compositions and another instructor was also involved in marking off the whole variables with the researcher. The data collected were analyzed through PASW Statistics 17.0.The study has major findings as follows:1) Frequencies and lengths of producing units were closely related to the variable of term. On the whole, students in higher terms were able to produce more Ss, T-units, DCs, and Cs and longer T-units, Cs, Ss, and EFTs. But none among LT, LC, LS, and LEFT could significantly discriminate between groups.2) The non-English majors were experiencing the transition from using compound sentences to using clauses. C/T was linearly increasing while increase of DC/C took the shape of waves. T/S decreased but was not decreasing linearly. EFT was significantly increasing from the first term to the later three terms. Change in EFT/T took the shape of waves but there was no significant difference between the later three terms. C/S witnessed a whole increase from the first term to the fourth term but failed to report significant difference between terms.3) SSs were used most often, CMSs and CSs less often, and CCSs least often. SSs were linearly decreasing but could not discriminate between groups significantly. Significant difference was found between the students in their use of CMSs and CSs.4) The developmental order, i.e. the transition from coordination to subordination also existed in non-English majors'writing practice. DCs and OCs were increasingly used but no obvious variation was found between terms in terms of the use of DCs. ACs, OCs, and ATCs were frequently used while SCs, PCs, and APCs were seldom used.5) Syntactic complexity could mirror writing quality to some degree. C/T, DC, DC/C, EFT, EFT/T, AC, and SC were significantly relevant to writing scores while no significant correlation was found to exist between any of LC, LT, LCL, LEFT, and LS and writing quality. ATCs were not closely related to writing quality,either. SCs could significnatly discriminate the students in the fourth term and those in other three terms but failed to discriminate between low, intermediate, and advanced groups. OCs could significantly discriminate between low-level writers and the other two groups, the advanced writers and the other two groups.The findings further proved that C/T and DC/C were effective in different language environments and the findings that syntactic complexity significantly and positively correlated with writing quality had some revelations to writing instruction and writing evaluation.