포스트에디팅 교육 실행연구 : 석사과정 영한 포스트에디팅 수업 최적화 방안에 대한 고찰

이준호 韓國外國語大學校 大學院 2021 국내박사

This study is an action research aimed to discover the most efficient and effective way to provide training in post-editing of machine translation for students in the master's courses for interpretation and translation. Post-editing differs from traditional human translation in that humans modify the machine-processed outcome. As a result, the knowledge and skills required to perform post-editing tasks can differ from those required for human translation. Therefore, it is arguable that a post-editing module should take a different approach from conventional translation training. In addition, Korean language differs much from Western languages in their morphology, syntax, and pragmatics. This means the quality of English to Korean machine translation outcome may not be so good as that of Western language pairs, potentially making the post-editing task even more challenging. Therefore, it is necessary to deeply investigate the characteristics of English-to-Korean post-editing. However, there has been insufficient discussion on the distinctiveness of English-to-Korean post-editing, the necessity of post-editing training, what should be taught in the classroom, and how the training should be planned and delivered. Therefore, this study would like to explore the uncharted research areas further to get closer to optimized post-editing training. In this journey, this study presumes it is safer to plan post-editing training within the established framework of translation training. It is well known that better translation training can be achieved with 1) needs analysis, 2) identification of objectives, 3) development of teaching methodologies, and 4) assessment of outcomes. Hence, this study poses a set of questions as follows: 1. Needs Analysis: What are the typical errors or behaviors students show when they perform a post-editing task for the first time? 2. Objective: What kind of knowledge and skills are needed to improve post-editing performance? 3. Teaching method: What kind of teaching and learning activities are required to secure the knowledge and skills? 4. Assessment: What is the best way to assess learning outcomes and how the teacher can utilize the assessment outcome to further improve the module and encourage students to learn more actively? In order to seek answers to the questions, this research adopted an action research approach. Action research is known to be useful in narrowing the gap between knowledge and practice and in resolving the pending issues in any practices. The effectiveness of action research has been proven in many disciplines and translation studies is no exception. Based on the circular and repetitive nature of the approach, this study will provide a thick description how the planning, execution, and assessment of the post-editing module has evolved. as its scope and depth evolve. In order to provide thorough descriptions and in-depth contemplation, this study, unlike in previous research, observed more than 80 students in total for more than three semesters and analyzed multiple data sources repetitively. Therefore, it is expected that this study can fill the research vacancy in English to Korean post-editing training and to build up the arguments of previous research. In addition, faithful to the pedagogical objective of action research, the observations, analysis, and reflections made in this study will help the researcher to better understand post-editing and learn how to teach more effectively based on the interaction with students. In addition, students will be able to learn about the declarative and procedural knowledge about post-editing and improve their hands-on skills to perform post-editing better. As there have been only a few studies about English-to-Korean post-editing training, the details about ‘what to teach’ and ‘how to teach’ will be a good source of information for those interested in post-editing research or planning to create a module. However, considering the evolutionary nature of action research, this study will continue even after the finalization of this thesis. That is why this research presents a set of future research items to further explore unresolved questions. The insights gathered in this study and future research will lay a solid foundation for better understanding about post-editing and truly optimized post-editing training.

순차통역 역량 강화를 위한 학부 한영 통역 수업의 시역 훈련 효과 연구

박예진 한국외국어대학교 대학원 2025 국내박사

This study investigates the impact of sight interpreting (SI) training on the development of consecutive interpreting (CI) competence in undergraduate Korean-to-English interpreting classes. Sight interpreting, defined as the oral rendition of written texts, is recognized as a distinct mode of interpreting that shares certain similarities with consecutive interpreting while also exhibiting notable differences. Frequently incorporated into interpreting classes, SI has seen limited investigation in two areas: 1) its pedagogical effectiveness in developing and assessing sub-skills, such as keyword identification, meaning unit segmentation, and information prioritization, and 2) the impact of SI training on CI competence, analyzed through fidelity, target language proficiency, and delivery. To address this research gap, this study employs a Linear Mixed-Effects Model (LMM) to analyze the effects of SI training on CI performance. To complement and contextualize the statistical analysis, qualitative data—including the researcher’s class notes, students’ reflective journals, and weekly brief surveys—were utilized. These qualitative insights highlighted how learners perceived and experienced the impact of SI training, uncovering details that might not be apparent in the statistical results, reinforcing the findings revealed by the statistical analysis. Two groups of students in Korean-English CI courses participated: the experimental group received SI training as part of the course, while the control group followed traditional CI training without SI. Their performances in mid-term and final exams were assessed by six evaluators, and statistical analysis was conducted using an LMM in RStudio. Results indicated that SI training had a significant positive effect on CI competence. The experimental group outperformed the control group in the final exam, with a mean score difference of 1.556 points (p = 0.003), suggesting a cumulative training effect. While mid-term results showed marginal significance (p= 0.059) and demonstrated a positive trend, the final exam confirmed the long-term benefits of SI training for CI enhancement. The findings suggest that repeated SI practice helps students better process source text meaning, structure output in the target language, and enhance delivery, ultimately improving overall interpreting performance. Significant correlations between SI sub-skills and CI performance were observed across the pre-test, mid-term, and final exams, underscoring their consistent interrelation. Further analysis revealed that identifying key words showed the strongest correlation, leading to a 3.27-point increase in CI performance for every 1-point gain in the key word identification score. Additionally, fidelity—evaluating the completeness and accuracy of message delivery—had the highest correlation (+0.772), demonstrating that SI training significantly enhances source text comprehension and accurate information delivery. The experimental group showed significant longitudinal improvement in sight interpreting sub-skills, including key word identification, meaning unit segmentation, and information prioritization, indicating that the benefits of SI training became more pronounced over time. This finding aligns with previous research emphasizing the importance of sustained practice to achieve meaningful gains in interpreting competence. These results have practical implications for interpreter education, suggesting that SI can serve as a complementary training tool in CI courses, especially for undergraduate students. The training helps develop foundational skills, such as text comprehension, analysis, and information processing, while reducing cognitive load. By engaging students in progressive and systematic practice, SI training enables learners to gradually build these fundamental skills, providing a structured pathway for skill acquisition. Moreover, SI training benefits students across all proficiency levels, helping to close gaps among learners and fostering equitable and inclusive learning outcomes in interpreter education. The study acknowledges limitations, including the relatively small sample size, the focus on a specific language pair, and the challenges in controlling external variables. Future research should involve more diverse learner groups, additional language pairs, and extended training periods to validate the findings’ generalizability and to further investigate the long-term effects of SI training on interpreting proficiency.

학부 1학년생들의 영한번역 과정 중 ST 이해단계에서 나타난 번역문제 유형분석 : 학생들이 지각한 번역문제와 학생들이 비지각한 번역문제 비교를 중심으로

안수경 한국외국어대학교 대학원 2017 국내석사

Analyzing translation problems of first-year undergraduate students in their ST comprehension stage for English-to-Korean translation The purpose of this study is to identify what translation problems first-year undergraduate students face in their source text (ST) comprehension stage for English-into-Korean translation. In other words, this research examines which parts of English texts students felt were difficult and the reasons why students struggled to understand them. The students who participated in this study completed three translation tasks for each text; ⅰ) gist translation, ⅱ) full-text translation, and ⅲ) translation revision. After performing the gist translation task, students completed questionnaires through which they identified what problems they had with understanding the given ST, and their answers were classified as Students’ Perceived Translation Problems (PTP). After the second translation task, that is, the full-text translation, students received indirect feedback from the researcher/instructor and were involved in an in-class discussion concerning their translation. In this step, students’ translation errors―both binary errors and non-binary errors due to insufficient understanding―were checked and classified as Students’ Unperceived Translation Problems (UTP). In other words, UTP signifies the types of errors that students committed in translating parts of the ST that they did not mention as their PTP. This study focused on finding the answers to the following research questions: 1. What are the characteristics of Students’ Perceived Translation Problems (PTP)? 2. If Students’ Perceived Translation Problems (PTP) are solved, does that mean that the students are now fully ready to understand the given English text without any problems? 3. What are the characteristics of Students’ Unperceived Translation Problems (UTP)? 4. What are the differences between Students’ Perceived Translation Problems (PTP) and Students’ Unperceived Translation Problems (UTP)? Once students’ translation problems were identified, they were categorized based on PACTE’s classification of translation problems: linguistic problems, extralinguistic problems, problems of intentionality, and so on. The ensuing analysis indicates that most of the Students’ Perceived Translation Problems (PTP) were located in the linguistic problems category, while many of Students’ Unperceived Translation Problems (UTP) were classified as intentionality problems. The students were inclined to believe that they could understand given English texts once they gained a full understanding of the texts’ vocabulary and syntactic structures. However, the Students’ Unperceived Translation Problems (UTP) showed that students had difficulty grasping the various meanings conveyed by the same words and/or phrases in depending on the situation, that is, the intentionality of the source text. The very idea to conduct this study occurred to the researcher/instructor when running a practical English class for first-year undergraduate students. During the semester, the researcher noticed that Students’ Perceived Translation Problems (PTP) were not the major problems; they were other hidden problems, that hindered the students’ understanding. To gather the empirical data for Students’ Unperceived Translation Problems (UTP), the researcher carried out an exploratory study during the summer break with five students. The participants were given ten English texts to translate into Korean that were three paragraph-long. Although the small number of students prohibits generalization, the study is still meaningful as an exploratory study. The findings indicate that Students’ Perceived Translation Problems (PTP) are different from Students’ Unperceived Translation Problems (UTP), and that teachers should more focus on Students’ Unperceived Translation Problems (UTP) in order to assist their students gaining more comprehensive understanding of English texts. It is hoped that this study can serve as a starting point for future research endeavors on finding ways to improve undergraduate students’ ST comprehension.

학부 통번역전공생의 한영번역 도구로서의 온라인 연어사전과 COCA 사용 경험 반응 연구 : 스크린 레코딩과 인터뷰를 중심으로

조슬기 한국외국어대학교 대학원 2017 국내석사

A study on the attitudes of T&I major undergraduates on the use of an online Collocation Dictionary & COCA in A-B translation The present study draws on the assumption that students’ A-B translation can be produced in better quality which enables more effective communication with target readers by enhancing students’ use of English collocations. And this has been proven through a number of related studies in English Education and Translator education. Based on this, this study ultimately aims to suggest and promote the use of two translation tools focusing on English collocations, namely an Online Collocation Dictionary and COCA, in future A-B translation courses. To achieve this ultimate goal, the present study was carried out as an exploratory research and a needs analysis. And it tried to answer the following questions: i) how T&I students use these translation tools ii) how do they think about the experiences of using these tools, iii) what are the difficulties they face when using the tools, and iv) what are the pedagogical implications of findings of the present study for undergraduate A-B translation education. To answer these research questions, the researcher first conducted a pilot study by analyzing two A-B translations conducted by two T&I undergraduates, to see if the three types of collocations suggested by Lee (2014) are observable in students’ translations and verify the use of the tools in this study. Then, five T&I undergraduates were recruited and participated in the present study for 4 weeks. Using these two tools, the participants translated 6 Korean news articles into English, 1 or 2 texts each week, and were asked to screen-record their translation process. After that, they were pre- and post-interviewed every week by the researcher. The researcher chose participant-oriented and process-oriented approaches as the main research approach. First, the researcher observed students’ translation process using a screen recording program to see how they use the tools; more specifically, their search terms, search orders, common traits or differences among the participants, and others were observed. And the researcher conducted semi-structured interviews which include both structured and unstructured questions. The questions are about their responses regarding the use of the tools and their translation process. The researcher, then, analyzed the data collected through observations and interviews according to each research question. Research questions 1 to 3 were first answered based on the data from both methods of screen-recording and interviews, and the last research question was deduced by analyzing the answers of the other previous three research questions. The findings of the study are significant as this study could provide insight on T&I undergraduates regarding their knowledge on English collocations and their use of related tools in A-B translation. For example, it hints at what students need to study when first using the tools, what difficulties they undergo, which is an effective way of using the tools, and their positive and negative attitudes toward using the tools. And these findings could lead to some pedagogical implications which can be actively applied in current or future translation courses such as aligned course designs, teaching methods, and search strategies for students. Yet, this study is not without its limitations: first of all, it was carried out with five participants and, thus, the size of the data produced is too small for generalization. However, this study is still meaningful as it tried to provide information about the students who are the primary beneficiaries of these tools and the very design and execution of what we call education, which has not been sufficiently studied under this topic so far. Hopefully, this study can serve as a significant start to lead further related studies like those about undergraduate A-B translation education, English collocations in translation and translation tools for A-B translation.

(A) study on the translation of wordplay : application of delabastita's translation methods of puns for analyzing the translation of wordplay in huevos : little rooster's egg-cellent adventure

Delmy Yesenia Gomez Rios Hankuk University of Foreign Studies. Graduate Sc 2018 국내석사

본 연구는 1) Huevos: Little Rooster's Egg-cellent Adventure 애니메이션의 서•영 번역에서 언어유희를 표현할 때 어떤 번역 전략을 이용했는지, 2) 원천텍스트 (ST)에서 나타난 언어유희가 목표텍스트 (TT)에서도 동일한 유머효과를 내는지를 알아보았다. 이 연구는 다음의 주요한 네 단계로 실행되었다. 우선, 분석할 영화를 선택하였고, 분석 방법을 선정한 다음, 애니메이션에 나타난 언어유희를 분석하였다. 마지막으로 평론가와 관객의 비평을 분석하였다. 분석 대상으로 선택한 영화는 멕시코의 애니메이션 영화 『Huevos: Little Rooster's Egg-cellent Adventure』였다. 연구 목적 달성을 위하여 연구자는, 영화를 선택한 다음, 델라바스티타의 언어유희 번역 방식을 적용하여 영화의ST와TT에 나타난 언어유희 번역을 분석하였다. 분석결과는 델라바스티타의 번역틀 총 8개 중에서 7개의 전략을 『Huevos: Little Rooster's Egg-cellent Adventure』분석에 적용할 수 있었다. 이 결과를 토대로, 델라바스티타의 언어유희 번역방식 또는 그에 따라 제시된 번역틀이 언어유희 번역의 분석에 있어 효과적이라는 것을 알 수 있었다. TT에 나타난 유머효과를 살펴보면, 중남미의 관객들은 영화 본연의 유모효과를 충분히 즐겼다. 이들은 영화평가에서도 좋은 점수를 주었고, 영화를 좋아한다는 리뷰를 작성하였다. 본 연구의 한계점으로는, 하나의 영화를 선택하였고, 연구의 범위를 특정 언어 즉, 서반아어에서 영어로의 번역에만 한정하였다는 것이다. 따라서 델라바스티타의 언어유희 번역방식 또는 그 번역 틀의 적용 가능성을 높이기 위하여 다른 언어와 영화들을 대상으로 하는 더욱 많은 언어 유희 연구가 요구된다. 본 연구가 유머번역의 연구, 특히, 언어유희 번역의 연구에 미력하나마 기여할 수 있기를 희망한다.

학부 한영번역교육에서의 번역보조도구 활용 관찰 연구 : 온라인 사전, 검색엔진, COCA의 활용 관찰을 위한 스크린 레코딩과 인터뷰를 중심으로

김인성 한국외국어대학교 대학원 2016 국내석사

Observing Undergraduate Students Using Translation Tools for Korean-into-English Translation The purpose of the present study is: i) to observe and analyze how students use translation tools while they translate from Korean into English and ii) to draw implications and lay the foundations for a follow-up study that explores translation pedagogy for translation tools. In other words, this study was carried out as a needs analysis for future endeavors to better teach translation tools to undergraduate students. To achieve the first objective of the study, the researcher recruited five students as participants and asked them to translate using translation tools. They were given three Korean texts: they were each two paragraphs long, and were about the risks of TV loan advertisements, the call for better honoring war veterans, and the need for higher awareness of computer security. In order to observe how students use translation tools, screen recording was used so that participants could translate in a natural setting without any kind of control imposed. For this, the participants had to be unhindered so that they can feel as if they are doing their homework assignments in a comfortable environment, and not as if they’re participating in a certain study. For the second objective, the researcher chose a process-oriented approach in order to gain a meaningful insight into translation education; therefore, triangulation was used, which consisted of classifying the search queries of the participants made available through screen recording and interviewing the participants. In detail, all the search queries observed were classified according to the four stages of translation process of the interpretive model. And then, a series of semi-structured interviews were conducted, which was mainly designed to keep the researcher from making a subjective classification as well as to gather additional information. The design of the research led to some significant findings. For example, data on what types of translation tools and how frequent participants use them, how they use each translation tool for each stage of the translation process and educational implications from the analysis were gained as a result of the study. The findings of the study are deemed significant as they give further insight into what should be considered when designing a course on using translation tools. Teaching students should be done based on an in-depth understanding of what translation tools they use, how they use them in each translation process, and what could go wrong when they use them. Although the small sample size of the study made it inadequate for generalization, this study is still meaningful in that it tried to observe and gain a better understanding of how students use translation tools, giving a glimpse of what could be generalized in future studies. Hopefully this study can be a meaningful starting point for future endeavors to realize a student-centered approach for teaching translation tools.

학부 전공생들의 인간번역과 기계번역 포스트에디팅 수행 과정 및 결과 분석 : 학년별 비교 및 교육에의 함의

김인성 한국외국어대학교 대학원 2025 국내박사

This study examines how undergraduate training—measured by year level and accumulated translation coursework—affects students’ performance in human translation (HT) and machine translation post-editing (MTPE). A total of 49 students majoring in English Interpretation and Translation at Hankuk University of Foreign Studies participated in both HT and MTPE tasks. Their product quality and process data were analyzed through statistical methods and screen-recording-based qualitative analysis. The results show a statistically significant positive correlation between HT quality scores and students’ completed translation-related coursework (Pearson r = 0.458, p = 0.0018), suggesting that cumulative academic experience contributes meaningfully to HT competence. ANOVA and post-hoc tests further revealed clear differences between lower-year (freshmen and sophomores) and upper-year (juniors and seniors) students, indicating skill development over time. However, greater score variability among seniors may reflect a transitional stage in which increased autonomy in decision-making temporarily reduces task efficiency—an observation aligned with the intermediate stage described by Quinci (2014). In contrast, MTPE performance showed no significant differences across year levels (ANOVA F = 0.83, p = 0.49), nor a significant correlation with coursework (Pearson r = -0.204, p = 0.2057). Instead, task completion rate showed a limited but meaningful association with MTPE quality, suggesting that real-time processing efficiency plays a more central role in MTPE performance than formal instruction. Screen-recording data support this contrast. High-performing HT students engaged in contextual interpretation, selective information processing, and careful revision—signs of accumulated practice and internalized translation strategies. In contrast, high-performing MTPE students intervened only when necessary, maximizing both accuracy and efficiency. Low performers in MTPE, however, often exhibited excessive distrust of MT output, leading to unnecessary rewriting and reduced product quality. These findings indicate that while HT and MTPE are both translation tasks, they involve distinct cognitive demands and skill sets. Although they may appear similar on the surface, they require differentiated pedagogical approaches. Based on this, the study proposes MTPE instruction that prioritizes selective intervention skills and the development of consistent error evaluation criteria.