A Development of hybrid network technology-enhanced language learning model based on social constructivism approach for promoting EFL learner autonomy

Abstract

To describe a three-year process of research and development of Hybrid Network Technology-Enhanced Language Learning (HybridNTELL) model within an English foundation course for first year undergraduate students in Chulalongkorn University, Bangkok, Thailand. The process of HybridNTELL model research and development was based on Vygotsky (1896-1934)-inspired social constructivism approach to foreign language learning. The model design was driven by the common challenges in teaching English as a foreign language (EFL) in higher education: increasing class size and diversity, limited exposure to the target language, and the washback effect of high-stake examinations, in contrast with a growing demand that EFL learners develop a working knowledge of English to participate in all kinds of 21st century global communications. To overcome the challenges, the HybridNTELL model combines online learning activities with English as a foreign language classroom learning environment. Social constructivism is used as the framework for the model design and to provide a perspective through which the model is evaluated. This study investigates how the students with different proficiency levels learned in and benefited from the HybridNTELL environment. Quantitative and qualitative data collection and analyses were conducted in three classes consisting of 143 students from the population of approximately 2,500 EFL students. 90 students from these three classes were selected by stratified random sampling method based on their proficiency scores of 30 in each high, mid and low group to examine their language development. Findings suggest that previous English proficiency levels have no effect on the students' autonomy and language development. The pre- and posttest suggest that the learners improve their language skills (F (2, 87) = 7.04, p < .001) regardless of their previous English proficiency levels, F(2, 87) = .24, p = .078. However, there is an interaction between the effect of English proficiency levels and degrees of learning autonomy demonstrated in the four task types (F(6, 87) = 15.96, p <.001) on the improvement of English proficiency. Furthermore, the experimental students' mean score on the curriculum-based achievement test shows significant difference from the overall population's (t = 6.66, p < .001). Learning variables are (1) the degree of autonomy, (2) patterns of social interaction, (3) discourse patterns in collective scaffolding, and (4) teacher's scaffolding in HybridNTELL environment are incorporated as the strong predictors of students' English proficiency development (R² = .87, F = 63.03, p < .001) and their curriculum-based achievement (R² = .96, F = 81.54, p < .001). Findings suggest that students make good learning progress with HybridNTELL model but the different degrees of achievement are dependent upon the patterns of their learning process in the environment. HybridNTELL project draws attention to issues concerning the process of learning model development and the delivery that should be addressed in research for effective application. The evaluation of this research and development of the HybridNTELL model is based on empirical evidence consisting of students with different proficiency levels learning English in large classes. This research, thus, provides an applicable guideline for institutional practices and basis for future research.