Many studies recognize the complexity, accuracy, and fluency (CAF) of a learner's language as important factors that measure second language performance and explain the development of proficiency. However, existing studies on Vietnamese learners of Ko...
Many studies recognize the complexity, accuracy, and fluency (CAF) of a learner's language as important factors that measure second language performance and explain the development of proficiency. However, existing studies on Vietnamese learners of Korean language have not measured syntactic proficiency systematically. The study aims to measure the syntactic complexity and accuracy of the Korean learner‘s written assignments, thereby revealing the syntactic development pattern of students of Korean language departments across four Vietnamese universities. These patterns can be used to suggest educational directions for improving learners’ ability to formulate sentences.
In an attempt to measure syntactic complexity and accuracy, we made the following steps. Firstly, written assignments were collected from second to fourth-year students at University of Languages & International Studies, Hanoi University, University of Social Sciences and Humanities of HCMC National University, and Ho Chi Minh University of Technology. Secondly, we examined how complexity and accuracy have been measured in existing ESL studies and Korean education studies to decide the proper complexity and accuracy measurement index. In ESL studies, major methods of exploring language development of language learners were ‘Suppliance of Obligatory Context’ analysis, ‘Target-Like Use’ analysis and Frequency analysis.
In this study, the syntactic complexity index was determined by three aspects, which were: the length of the sentence (the number of morphemes per sentence), sentence expansion (the number of clauses per sentence, the number of embedded clauses per sentence, the number of coordinate clauses per sentence) and the frequency of related grammatical items in charge of the sentence expansion. And syntactic accuracy was measured from two aspects: general measures and specific measures. The number of ‘Errors per 100-word segments’ was selected as the general measure and the percent accuracy of related grammatical items and the embedded sentence errors were used for the specific measure. According to our results of the complexity index analysis, there is a significant difference between the beginner-intermediate level and advanced level. In particular, the number of embedded clauses per sentence increased significantly from level 2 to level 4, but this trend was almost stagnant after level 5. The result from measuring the usage rate by clause type indicates that the rate of noun clauses is extremely lower than that of other complex clause types, although it is not statistically significant when ranked by proficiency level. At the same time, it can be inferred that the ability to use noun clauses is limited among embedded sentences. Meanwhile, there is no statistically significant difference in the number of coordinate clauses between levels, thus the main factors attributing to the more complex sentence are the increment of relative clause and adverbial clause. To measure qualitative complexity, the development of syntactic complexity and the frequency of grammatical items were analyzed based on proficiency, and groups with similar developmental patterns were classified. Among them, Vietnamese learners hardly or avoid in using connective ending that are ‘-ge, -eo/aya, -a/eo/yeo, -(eu)myeo, -ndamyeon, -la, -goja, -(eu)meulo, -(eu)na, -(eu)lyeogo, -deolado, -(eu)na, -deus’ and so on. Even though these grammar items are regularly used in native writing, they are hardly used across objects of study. Therefore, special teaching methods are required for these grammatical items.
In terms of accuracy measurement, the number of errors per 100 word segment was highest at level 2, and up to level 4 accuracy still did not improve, up to level 5 accuracy improved somewhat and did not make any significant difference until level 6. Sentences’ complexity (sentence length and sentence expansion) was improved along with proficiency, but accuracy didn’t follow this trend. The percent accuracy of related grammatical items is measured by the ‘Target-Like Use’ analysis. Grammatical items that share a common development pattern are grouped into 5 groups. Group 1 is the grammatical items with very low frequency from level 2 to level 6, Group 2 is the grammatical items which has an increase in both percent accuracy and frequency, Group 3 is the group that decreases in the frequency of use but increases in accuracy, Group 4 is the group that increases in frequency at level 4 or level 5 but decreases at level 6 (inverted U shape), and the last group is Group 5 that shows U-shaped pattern of development, percent accuracy suddenly but temporarily drops off only to return once again to their usual level. Grammatical items in Group 1 and Group 4 are items that need improvement in terms of education. After analyzing the error patterns of adverb clauses, relative clauses, and noun clauses, we can broadly classify the indicators of accuracy into four types as follows: (1) misuse of endings, (2) overuse of endings, (3) confusion between endings in the same category and (4) mixing errors with endings in other categories. In this study error (3) was found to be the most common error. The main reason is that in the process of combining preceding clauses and following clauses, Vietnamese learners easily make errors by mixing endings with similar meanings or inappropriately combining them, or in the process of constructing clauses, errors such as subject-predicate inconsistency and absence of essential arguments are made by violating the argument structure of the internal structure.
To sum up, this study is significant in that it quantitatively analyzed the development of syntactic complexity and accuracy of Vietnamese learners, which had not been attempted yet, and measured the usage and error rates of grammar items that caused the development of such complexity and synthesized them to derive educational implications. It has shown the development of syntactic complexity and accuracy by proficiency level for Vietnamese learners from level 2 to level 6 and discusses the syntactic characteristics of Vietnamese learners through the contrastive analysis of Korean-Vietnamese sentences. However, a limitation of this study is that data from the same learners were not collected for a long time, thus lacking individual development analysis in writing skills. The findings of this study can be used for future sentence education research targeting Vietnamese learners.