Effect of problem-based learning with think aloud protocol approach on Chinese students’ translation competency

Abstract

Translation competence is considered to be made up of different components, and domain knowledge is one of the most important components. Because translation market has evolved into unprecedented sophistication and specialization, which may cover almost every sphere of people’s life, ranging from legal translation, medical translation, business and finance translation to scientific and technical translation. All these changes require translators to acquire not only linguistic knowledge, but also the background knowledge in one or two specific domains, so that they can provide quality translation service in the market. The thesis intends to develop an integrated approach to improve Chinse students’ translation competence especially in the area of domain knowledge. The case study took place in School of Foreign Languages and Cultures, Ningxia University, China. The research is composed of four stages including the problem justification, the development of the approach, the prioritization and selection of the approach, and the implementation of the selected approach. The research problem is translation students’ insufficient domain knowledge, which affects their translation accuracy and speed. To address this problem, popular constructivist methods and learning in action theory from knowledge management were reviewed, and possible sets of integrated methods were proposed according to the translation industries’ requirements of translators. To select the most appropriate one from among the alternatives without being subjective, Best and Worst Method was adopted, as it uses relatively fewer data and generates more reliable results due to its consistent comparison. Best-Worst Method analysis was operated twice, with domain knowledge ranked as the most important criterion and Problem-based Learning with Think-aloud Protocol selected as the most desirable approach. The selected approach was then implemented and validated. The approach was modified after three-time implementation. The results indicated that the integrated approach stimulated more active learning, and students reported general satisfaction of the approach and better understanding of domain knowledge. A paired t-test showed that there was a significant level of 0.01 between the pretest and posttest (t=-18.662, p=0.000), and the difference in specific comparison showed that the average value of pretest (75.41) was significantly lower than the average value of posttest (87.18); meanwhile, the total error counts of the posttest dropped from 102 to 78.