Il-sun Hyun

DOI: Vol.25(No.4) 9-34, 2022

This study explored the comparative effects of two wiki-mediated tasks on 76 EFL college students and their perceptions of its use regarding reading and writing skills. For 10 weeks, 32 students were involved in a wiki-mediated task collaboratively while 44 were engaged in it wiki-mediated Tasks individually. Upon completing the tasks, the students responded to a questionnaire eliciting the perceived effects of the wiki-mediated tasks on their reading and writing skills. The pre- and post-test scores were collected and analyzed with an independent-samples t-test, and their responses to the questionnaire were analyzed by frequency analysis. No statistically significant differences were found in the post-test scores between the two groups, but the students reported different perceptions of the two types of wiki-mediated tasks. While the students involved in tasks individually felt that their reading and writing skills had improved, those who engaged in groups believed that only their reading skills had improved. Nevertheless, the students of both groups showed positive attitudes toward the wiki-mediated tasks as the tasks, which enhanced their self-regulation, interests, confidence, and peer interaction.

Wiki-mediated tasks, wiki modes, collaborative learning, English reading and writing

Hyunjeong Nam

DOI: Vol.25(No.4) 35-65, 2022

The role of multimedia has gained traction in SLA. However, previous research concerning Korean EFL learners has focused on a particular linguistic feature or language skill in multimedia. Given the significance of pragmatic aspect of language in L2 learning, the study explored the effect of multimedia on L2 learning. Different from previous research which was limited to a particular genre of multimedia and speech act, this study used various genres of multimedia (movies, TV shows, commercials, personal YouTube videos) and speech act (15 situations in 5 speech acts). The aim of the study was twofold: to investigate the effect of multimedia-based learning on L2 learners’ pragmatic competence and pragmatic awareness. For this, a total of 94 university students participated in a pre-test and a post-test of Discourse Completion Task (pragmatic competence) and Pragmatic Attitude Scale (pragmatic awareness). Based on the comparison using a paired t-test between the control group without multimedia and the experimental group with multimedia in learning, the study confirmed the effect of multimedia-based learning on Korean EFL learners’ pragmatic competence and pragmatic awareness. Based on the findings, the study has made some pedagogical suggestions.

multimedia, pragmatic competence, pragmatic awareness, speech act

Eunjou Oh

DOI: Vol.25(No.4) 66-92, 2022

This study explored the use of machine translation (MT) for reading in college English classes. A possible mechanism of MT use at play during L2 reading was explained based on Kintsch’s Construction Integration model. Framed in dynamic assessment (DA), intercorrelations between students’ vocabulary knowledge, independent reading scores (IRS), MT-mediated reading scores (MTRS), and learning potential scores (LPS) were analyzed to explain a possible role of MT in L2 reading. Students’ perceptions of MT use were investigated via surveys after the MT-mediated reading assessment and the MT-mediated think-aloud reading (TAR) task as self-instructional training. One hundred ten students participated in the DA task, and one hundred three, in the TAR task. The findings confirmed the role of MT as an L2 proficiency mediating tool in reading assessment. Students’ perceptions of MT use were positive overall. Clarifying meanings of unfamiliar words and difficult sentences was reported to be the areas where the participants benefited most from MT use during the assessment. The positive effects of the MT-mediated TAR task were found to be improved English proficiency, increased sensitivity to self-monitoring, development of global competence, convenience, enhanced self-directed learning, and increased familiarity with English. Pedagogical implications of the findings are discussed.

machine translation, L2 reading, CI model, dynamic assessment, think-aloud

Hanna Park

DOI: Vol.25(No.4) 93-119, 2022

The purpose of this research is to investigate the effects of virtual reality (VR)-based English learning on English as a Foreign Language (EFL) learners’ speaking abilities. The participants were 30 EFL learners from a college in South Korea. They were randomly assigned to either the human-to-human interaction in VR world group (n = 12), human-to-AI chatbot interaction in VR world group (n = 10), or control group (n = 8). This study found that experimental group B, which engaged in English-speaking learning activities with an artificial intelligence (AI) chatbot in the VR world, showed the highest improvement in English speaking after the experiment, indicating that a VR program that incorporates AI technology is effective in improving students’ English-speaking ability. This is because the process of hiding oneself and interacting with another person in the VR world eased learners’ anxiety and increased their interest and self-confidence in speaking, which led to more active participation in class. The results of this study are also consistent with those of previous studies showing that conversing with another person behind a virtual avatar allows learners to participate and speak more actively. Based on the findings of this research, pedagogical suggestions are provided to help language learners improve their speaking ability in EFL contexts.

VR, virtual reality, VR-based English learning, VR learning, VR language learning, VR-based learning

김경회 Kim Gyeonghoe

DOI: Vol.25(No.4) 120-147, 2022

The influence between the diverse online classroom types affecting English speaking performances should be studied extensively for the overall online non face-to-face class situation. This study investigates the effects of online remote classes and video recording activities by considering English speaking accuracy, vocabulary, comprehension, fluency, and pronunciation. 120 learners of English as a Foreign Language (EFL) performed 3 English speaking tests that differed in classroom conditions (Zoom, recording, and control) with or without video-recording activity. A repeated measures multivariate analysis of variance showed that although both Zoom and recording classes affected the learner’s speaking performances positively, there were not any big differences between Zoom and recording classes. However, Zoom and recording classes showed better performance than control classes. In addition, video-recording classes showed better performance on the English speaking tests, especially English fluency and pronunciation, whereas there were not any superior performances found from no video-recording classes. Therefore, these results suggest lecturers should consider the various online classroom types, which can affect the students’ English learning motivation, satisfaction, and interaction, and develop useful classroom activities to maximize the educational effect of online English speaking classes.

online remote class, Zoom class, recording class, video-recording activities, Korean adult learners

김정현 Kim Junghyun

DOI: Vol.25(No.4) 148-169, 2022

The purpose of this study is to investigate the effect of self-directed language learning (SDLL) and self-determination (SD) on English achievement. SDLL and English achievement had a positive (+) relationship, but SD had a negative (-) relationship with English achievement in the correlation analysis. When regression analysis was performed to examine the effect on English achievement, SDLL had a positive (+) value and SD had a negative (-) value. More than half of the students (67.1%) answered that they developed a self-directed learning habit after taking online classes. After taking the online classes operated in weeks 1-7, students answered that they could review the contents well (N = 39, 23.8%). After taking the offline classes operated in weeks 8-14, 50 students (30.5%) answered they had better concentration when taking offline classes compared with online classes. Students responded positively to both online classes and offline classes, but in the case of online classes, there were more negative opinions. In the with COVID era or post-COVID-19 era, many students wanted a mix of online and offline classes as a preferred English class type, which is considered to be influenced by the online classes they had received before.

self-directed language learning, self-determination, academic achievement, with COVID era

남대현 Nam Daehyeon

DOI: Vol.25(No.4) 170-191, 2022

The current study aims to identify research trends in papers published in Multimedia-Assisted Language Learning (MALL). To this end, the bibliographic information on 567 papers published from 2002 to 2021 is analyzed using network visualization analysis. The results find that MALL has published about 28 papers per year during the period; however, it has published more papers than average to respond to social and educational issues promptly. According to the network visualization analysis based on keyword and term extraction, it is revealed that MALL has focused on publishing papers related to the terms ‘blended learning’, ‘elementary school’, and ‘korean university’ over the past 20 years. In addition, the cluster analysis confirms that the research topics that have been discussed in MALL are as follows: blended learning, elementary school learners with writing ability, computer-based language (education), and English education. The results of the timeline view analysis on clusters, compared to its earlier papers, find that published papers in MALL tend to research multiple topics across clusters. The burst terms analysis reveals that blended learning, elementary level, and Korean college topics were burst terms during certain periods. The suggestions and implications of the findings are also discussed.

MALL, CALL, network analysis, visualization, bibliometrics

문종근 Moon Jonggeun

DOI: Vol.25(No.4) 192-221, 2022

The purpose of this research is to analyze the effects of the corpus-based model of flipped learning on Korean EFL learners’ understanding of English grammar. To achieve this aim, a total of 92 Korean middle school students in their first year participated in the experiment for 7 weeks: one control group was given traditional grammar teaching, another experimental group was a flipped learning, and the last was a corpus-based flipped learning group. All experiments consisted of 6 sessions. Pre- and post-test was administered to discover the difference in the effect of grammar learning on the learners' English grammar comprehension of each group. Then, a one-way ANOVA was conducted not only to test the homogeneity between groups but also to find the impact of a corpus-based flipped classroom on learners’ understanding of English grammar. Also, all students participated in the survey and some significant improvement in sessions were shown. The result shows a significant difference between three groups statistically; the group using a corpus-based model with flipped learning significantly improved their understanding of English grammar rules. This implies that learning English grammar using corpus is more effective based on flipped learning than the other two methods; it helps learners to discover the grammatical rules in the given examples. Three practical implications of these results on corpus-based flipped learning model were discussed.

corpus-based, flipped learning, grammar comprehension, homogeneity

박나연 Park Nayeon , 이상민 Lee Sangmin

DOI: Vol.25(No.4) 222-236, 2022

In recent years, with the rapid development of technologies, utilization and adaptation of diverse technologies into education has drawn attention from many researchers and practitioners. Particularly, artificial intelligence and machine learning have great potential to provide tools such as customized adaptive and personalized learning for individual students, which can aid in providing high-quality learning experiences. With the joint advancement of machine learning and artificial intelligence, Deep Knowledge Tracing (DKT) artificial intelligence models have been utilized in education. The key of DKT is to monitor and estimate students’ evolving knowledge state, accurately predict their future performance on future exercises, and ultimately maximize learning efficiency. The current study developed a DKT model with a new parameter, spacing effect, in order to increase the accuracy of the model. It also considered a sentence-level variable as well as word-level variable in spacing effect to further increase accuracy. The newly developed model, SEDKT, was tested using learner trace big data released by Duolingo. The results showed that the accuracy of SEDKT was 0.720 AUC (Areas Under the ROC Curve) and indicated that the accuracy increased compared to the previous DKT models. Based on the results, this study provided suggestions and directions for the future in this area.

artificial intelligence, machine learning, deep knowledge tracing, English education

표시연 Pyo Siyeon

DOI: Vol.25(No.4) 237-257, 2022

This study aims to explore the pedagogical effectiveness of PBL (project-based learning) utilizing multimedia tools on English courses for communication. The research was carried out with a total of 50 university students whose English-speaking levels were pre-intermediate. The detailed report of the PBL case study, the students’ reflection journals and self-evaluation, the survey on the PBL classes, and the final lecture evaluation were analyzed qualitatively as well as quantitatively. The results showed that PBL provides various opportunities to use authentic English students want to express through structured activities. This makes their English knowledge transfer to procedural from declarative one, connecting English learning to their real life. The participants reported that PBL led them to establish a cooperative learning community that promotes a meaningful learning experience of practical English use in both online and offline learning environments. Moreover, in the process of completing the outcome of each project, their initiative and wholehearted behaviors were observed. Such empirical data suggests that PBL English classes, as an alternative to lecture-style ones which make students familiar with passive learning participation, can be highly effective for learners who have anxiety about English by promoting learning motivation through authentic learner-oriented projects.

PBL, project-based learning, student-centered learning, college English education, case study

황요한 Hwang Yohan , 신동광 Shin Dongkwang , 이혜진 Lee Hyejin

DOI: Vol.25(No.4) 258-284, 2022

New frontiers of world-breakthrough technology, such as Metaverse and AI, have been heralded as a promising future of education, and a revolutionary rethink of how we assess and teach is underway. Against this backdrop, this preliminary research endeavored to explore a new future-oriented model of English-speaking assessment, supported by Metaverse and AI. To this end, the paper is organized into three parts: Part I analyzes the status quo of English-speaking tests hitherto and their main facets; Part II explores a new way forward in redesigning speaking assessment with experiential, multimodal-evoked, and automated potentials; finally, Part III evaluated seven English-speaking assessment items using four previously developed models and our own three new ones. Thirty-four experts judged these items based on three criteria: validity, practicality, and authenticity. The results clearly demonstrate how technology could be used to create virtually authentic speaking contexts for learners. The insights gathered from this from this pioneering research inspire future research attempts in this direction and offer new directions for English-speaking assessment.

English speaking, English interview, assessment, metaverse, AI