The traditional translation process involves the sender, translator, and receiver, wherein the translator is centered on bridging the gap between the two parties by way of language. The increasing demand for translation in the globalized world has brought in more people in the translation process. In practice, the client submits a translation request to a translation company; the translation company outsources the translation project to a translator, the translator completes the project and sends back to the translation company. The reviser in the translation company proofreads and edits the translation, and sends back to the client. The parties involved in the translation process have increased from a minimum of three to at least five agents.

The development of technology has further revolutionized the translation process and the workflow of the translator. The integration of computer-assisted translation tools to the workbench of the translator allows the translator to reuse previously translated segments and terminologies while translating, which maintains the consistency of the text style and streamlines the translation process. Moreover, online platforms allow translators to collaborate on large projects on a single platform, regardless of location, time, or level of acquaintance. Crowdsourcing and fansubbing thus emerged as a new area to be studied. Translators are no longer the stereotyped working alone and rarely talk to other people type of introverted individual, but are expected to work in a team and join in discussions. The skills required of a contemporary translator is not the same as the skills required of a translator ten years earlier, but has this affect the way translators are trained?

The teaching of translation has long been a teacher-centered approach, where the teacher does most of the work during the lecture. This is still the case in secondary education in some Asian countries, where teachers prepare students for college entrance exams. In tertiary education, when translations become an academic subject instead of an exam question, a one-way approach to teaching students how to translate is no longer sufficient and effective. With an objective to prepare students for the job market, teachers should find ways to encourage students to play a more active role in acquiring the skills and competencies required of a professional translator. This is where flipped translation training came in the way.

A flipped classroom reverses the sequence of lectures, where students watch the content of lectures off-class via pre-recorded videos or slides. It is the responsibility of the students to make sure they have acquired the information required for the class.

During class sessions, students engage in hands-on classroom activities such as discussions, assignments, or collaborative tasks. Through these activities, teachers observe the learning outcome of the students. Any problems encountered by the student could be solved, and misunderstandings could be corrected on the spot. Follow-up activities could be designed after the class to review in-class activity and strengthen the efficacy of learning.

Flipped education has been applied in many disciplines over the past decade (Evseeva & Solozhenko, 2015; Lewis, Chen, & Relan, 2018; Lopes & Soares, 2018; Sun, Xie, & Anderman, 2018; Tan, Yue, & Fu, 2017), including translation (Deng, 2018; Lin, 2019; Tsai & Tsai, 2017; Yu, 2017). The use of flipped translation training enables effective communication among parties involved during the translation process and motivates students to become active learners.

The advancement of technology has given rise to large-scale translation projects that are completed collaboratively by more than one translator. Collaborative translation can be achieved regardless of distance, time zone, or nationality. The problem usually lies in how to collaborate with other translators so that the readers would only see one translator doing the job instead of hundreds of styles. Training on collaborative translation is, therefore important, and should be implemented in the curriculum. Kiraly(2000) highlights the importance of the collaborative learning environment for translation education, and in a flipped translation classroom, the teacher creates an environment where students can exchange ideas and finish tasks cooperatively, providing an excellent opportunity for students to learn to collaborate on a given translation assignment.

Collaborative translation in a translation classroom is often accomplished through peer revisions and group discussions, in which students are paired or grouped to analyze other students’ translations, provide feedback, and learn from their mistakes. Thus, students can revise their draft translations according to the comments received from the peer revision and group discussion process. This form of collaborative translation incorporates other people’s ideas, suggestions, knowledge, and competence. Furthermore, studies have found that students can be experts in peer revisions or group discussions, and they provide valuable insights from their perspective that are beneficial in improving their peers’ translations. However, not all peer revisions or group discussions are helpful in learning, and not all feedback yields positive results. Students are not professionals specialized in subject domains; therefore, certain students may feel reserved in analyzing the comments provided by their peers or even question the accuracy of their feedback.

Group discussions and peer revisions are commonly conducted for providing and receiving feedback for translation activities. Dobao (2012) considered that students learn more effectively through positive interactions from cooperation, sharing ideas, and integrating knowledge. In a translation class, students are often divided into groups to analyze the quality of other students’ translations, thus achieving collaborative learning. Furthermore, students who are highly interested in translation would form a private study group and abide by the same pattern: assign translation exercises and discuss the translations. The translation process involves brainstorming, drafting, and peer revisioning and revising the final version; therefore, students and teachers could comment on translations based on their understanding of the source text to improve the quality of the translations.

Group discussions originate from language learning. For instance, collaborative writing helps stimulate learners to use language and cooperate with peers to solve language-related problems (Swain, 2000; Swain & Lapkin, 2001). By gathering their knowledge and resources to solve these problems, learners perform language-mediated cognitive activities that are considered to help in co-constructing language knowledge and generate a more desirable performance (Donato, 1994; Ohta, 2001; Swain, 2000; Swain & Lapkin, 1998). With language as a semiotic tool, cooperation with experts or learners with more favorable competence helps internalize the constructed knowledge.

Studies have reported that every learner is an individual with different strengths and weaknesses; by cooperating, different resources can be integrated to complement each other, such that everyone is an expert in peer revisions or group activities and has positive effects on peers, yielding more desirable effects compared with an individual endeavor (Antón & Dicamilla, 1999; Donato, 1994; Ohta, 2000, 2001; Storch, 2002; Swain & Lapkin, 1998). However, not all group activities are helpful to learning, and not all feedback yields positive outcomes (Hattie & Timperley, 2007; Kluger & DeNisi, 1996; Narciss, 2008; Shute, 2008). Students are not experts in specialized domains (Hanrahan & Isaacs, 2001); therefore, they tend to be reserved or resistant to feedback from their peers in areas outside their expertise (van Gennip, Segers, & Tillema, 2010; Walker, 2001). Connor and Asenavage (1994) analyzed the references used by students when revising English writing, and they reported that only 5% of the revisions referenced comments from peer revisions, and 35% of the revisions referenced teacher feedback. Paulus (1999) analyzed 12 second language learners and reported that 87% and 51% of writing revisions were conducted in reference to teacher feedback and peer revisions, respectively. Yang, Badger, and Yu (2006) also examined 12 second language learners at a university in China and found that 90% and 76% of the revisions were conducted based on teacher feedback and peer revisions, respectively.

Student emotions during peer revisions considerably affect the feedback content (Shute, 2008). Certain students question the credibility and validity of feedback (Sluijsmans, Dochy, & Moerkerke, 1998), whereas others have indicated that only the teacher should provide feedback (Brown, Irving, Peterson, & Hirschfeld, 2009; Zhang, 1995). Studies on the effects of peer revisions are limited (Cho & MacArthur, 2010; Gielen, Peeters, Dochy, Onghena, & Struyven, 2010; Prins, Sluijsmans, & Kirschner, 2006), and are mostly associated with English writing; studies on peer revisions and group discussions in translation classes are scant.

This study investigates the effect of flipped translation training, with a special focus on collaborative translation, including (1) students’ perceptions of the learning outcome of group presentations and peer revisions and (2) the effects of group presentations and peer revisions on student translations. Understanding students’ perceptions of group presentations and peer revisions and their effects on student translators enables teachers to evaluate the effectiveness of collaborative learning and the flipped classroom approach. Moreover, teachers can propose solutions to constructing an autonomous learning environment, where students learn to collaboratively enhance their competence in determining the quality of the translation, in analyzing the language structure of the translation and in mastering translation skills.

The participants were 22 students enrolled in the general translation course. This is compulsory for third-year students in the undergraduate program of the Department of Foreign Languages and Literatures. Therefore, the students were either in the third year of study or the fourth year of study. The participants had little to no experience in translation and had not received any training in translation before taking the course. We collected 22 questionnaires. The students were allocated class time to complete these activities. Some finished within minutes; others required more time.

The flipped classroom approach was carried throughout the semester. The students were divided into three groups. Since the participants had limited experience and training in translation, the participants were divided randomly. Each group was assigned a subject field, in which they had to search for the text of their choice to be translated by their peers. The translation direction was from L2 (English) to L1 (Chinese Mandarin). The student groups selected one of the following text types: news reports, technological news reports, technical texts, and travel texts. They were then required to find a text piece from that field and assign it to their peers.

In an autonomous flipped classroom environment, the teacher is the facilitator who monitors the flow of classroom activities, guides students throughout the activities, and ensures every process runs smoothly. In this study, the teacher recorded videos on basic translation strategies and provided final feedback to the students on their revised translation. The students were asked to watch videos on basic translation strategies at home and work on translation exercises in the first three class sessions. In subsequent class sessions, the teacher instructed each student group to conduct dynamic searches for the text feature of the assigned text type. The students reported their findings to the class and generated discussions that might be helpful to their translation. Each student was then required to submit the first draft to the course Web site.

The group that assigned the translation analyzed the translations of all of their peers and presented their analysis in-class. The 1-hr-long group presentation covered an analysis of the source text, typology and features of the text in question, suggested translation strategies, difficulties in translation as observed from their peer translations, common errors found in the translations, and preferred translations.

This is followed by peer revisions. Upon completion of the peer revisions, the students were asked to grade the translations by using a Translation Quality Assurance (TQA) form (Mass.Gov, 2017). The TQA form was designed and developed by the Office of Public Health Strategy and Communications to ensure quality for the translated materials produced by the Massachusetts Department of Public Health in the United States. The continued demand for translation in Massachusetts is to increase information accessibility to nearly half of non-native English speakers residing in the region. As this is an ongoing translation process in practice by a government institution, the TQA form provided for reviewers to provide feedback on the translation quality is used in this study. On the TQA form, the reviewer was required to evaluate the translation based on loyalty to the source text, accuracy, register, false cognates, appropriateness for the culture or audience, and grammar and style with the provided statements. The reviewer was also asked to rate the overall translation quality and provide recommendations or comments. The review process was not anonymous; therefore, the students knew who reviewed their translations.

The students revised their first drafts in the class according to the peer revisions, TQA forms, and group presentations. They were allocated at least one hour to revise their translations before submitting their final version to the course Web site. These revisions were then graded by the teacher and checked and compared with peer revisions, TQA forms, and group presentations to analyze the basis of the revisions. Revisions from different sources were highlighted in different colors and calculated according to the number of changes made after referencing to each source.

Mittan (1989) considered that paired revisions generate more opportunities for discussions on writing. Therefore, in this study, the drafts were peer revisioned in class the following week. Student assignments were paired by the teacher after considering absentees on the day of peer revision, and whether the reviewers were in different groups. The students used the “Track Changes” function in Microsoft Word to edit and comment while reading the text. The students were asked to read the translation first before comparing it with the source text. This enabled them to check the language before understanding the meaning. Any insertion, deletion, moving, or formatting was tracked and highlighted.

At the end of the semester, the students were asked to answer a questionnaire analyzing the effectiveness of each review activity. The questionnaire included questions on the usefulness of these activities, the effects of these activities on their confidence in translation ability and language skills, and what they have learned from these activities. The difficulties encountered in these activities were also a subject of the questions. Please see the Appendix for the sample questionnaire. The questionnaire findings enabled us to understand students’ perceptions of these activities and how useful they found them to be.

The students reported the amount of time and effort they dedicated to research, and they evaluated their peers’ translations. They then provided valuable and presentable feedback to their peers. The students were required to consider the accuracy and adequacy of the data they collected, and to discuss them with their group members before presenting the findings to the class. A careful analysis of and targeted research on sources are also required. One question concerned whether they were allocated sufficient time for these activities. Most students responded that the time provided for these activities was sufficient, whereas others responded that excess time was allocated for peer revisions and TQA forms (Figure 3).

The students were also asked to rate the usefulness of these activities. In this study, the ‘usefulness’ of each activity refers to the practical support from peers through the activities in improving the quality of translation. This will, in turn, enable translators to learn how to revise their translations based on the suggestions proposed by others. On a scale from 1 to 5, most of the students rated group presentations as highly useful. Compared with group presentations and peer revisions, more students had a reserved view of TQA forms, with few finding TQA forms to not be useful.

This is supported by the answers to the question regarding the most difficult and interesting part of these activities; 50% and 32% of the respondents considered group presentation and peer revisions, respectively, to be the most challenging.

The flipped classroom approach in this study converted the linear teaching mode in the traditional translation classroom to a learner-centered learning environment where students are encouraged to actively participate in the learning process through group presentation, peer revisions, and translation quality assessment forms.

The flipped classroom approach increased communication between teachers and students, and students and students, encouraged effective students participation in class, and improved students’ learning outcomes. The students were able to accumulate translation skills and experience and improve their translation competence from translation practices, translation analysis, and translation evaluations, all of which were not made possible in the traditional teacher-centered approach to translation teaching. The pedagogical efficacy of these activities can be best described by what Bloom, Engelhart, Furst, Hill, and Krathwohl (1956) referred to as “higher order thinking,” as against “lower order activities” that involve remembering, understanding, and applying. In other words, as compared to the traditional classroom, collaborating in the flipped classroom through group presentations, peer revisions, and TQA forms prompted more effective learning.

The findings revealed that feedback from peer revisions and group analysis positively affects student translations. This effect helps students build confidence in their translation competence and enhances their language skills. Among the three

The flipped classroom approach in this study converted the linear teaching mode in the traditional translation classroom to a learner-centered learning environment where students are encouraged to actively participate in the learning process activities conducted in the study, the respondents considered the group analysis of student translations to be most interesting and helpful, although most difficult, with more than half of them referencing the presentation slides during the revision process.

The findings revealed students’ understanding of the source text, competence in evaluating translations, competence in analyzing the language structure of translations, mastering translation strategies, and learning from the translations conducted by their peers. The students were expected to learn from the analytical process, and to create an autonomous learning environment and obtain the attitude and competence required for lifelong learning.

The feedback provided by the peers on mistranslation, coherence, consistency, punctuation, redundant expressions, and readability highlighted the shortcomings of the student translators. Novice translators typically use words that sound correct; however, what sounds correct to one person might not always sound correct to another. This discrepancy can be resolved through collaborative translation and learning. The comments provide valuable suggestions that motivate student translators to rethink the usage of words and expressions, the logic of the wording, and the flow of the text. In addition, student translators become more aware of the competence required of a professional translator, and there is always scope for improvement, regardless of how proficient they believe their language skills are. Most important, the translators learned how to improve their translations.

Reading, evaluating, and commenting on the translation of others is also a critical lesson that was learned. One can always learn from countless references and new concepts to research by reading their peers’ translations. It is often more objective for a third party to identify errors or improvement opportunities; reviewers frequently notice how they have made the same mistakes. Reviewing the translation translated by reviewers is consequently useful for their revision. Students can become more sensitive to texts and learn how the same text has different interpretations. Reviewing and commenting on peers’ translation is challenging; however, after several trials, the student participants reported gaining confidence in providing feedback.

Collaborative learning and translation are an interactive process in which both parties benefit from each other. In a translation classroom, collaborative learning and translation can be achieved by collaborating to enhance the translation quality through group analysis, discussions, and peer revisions. In the industry, collaborative translation is a standardized process in which a piece of text is translated, edited, and approved before submission to the client. Integrating the collaborative learning and translation process in the classroom prepares the students for the roles of a translator, editor, quality checker, and project manager.