With the rapid evolution of technology, translation technology—particularly artificial intelligence (AI)-assisted and neural machine translation—has become a transformative force in the translation industry and translator education (Chan and Tang, 2024a; Chan and Tang, 2024b). To meet the dynamic demands of the language service market, it is essential for translators, translation trainers and learners to adopt a diverse range of digital tools and online resources, which can enhance efficiency, effectiveness, and overall quality of translation education and practice (Chan and Tang, 2025). However, in spite of the growing importance of digital literacy in translation technology, research indicates that both pre-service and in-service translators, as well as some trainers, may not sufficiently acquire the necessary skills to keep pace with the rapid advancements in this field (Venkatesan, 2023). For instance, Chan’s (2025c) study revealed that more than half of translation students possessed limited proficiency in using computer-assisted translation (CAT) tools, despite expressing a strong interest in acquiring post-editing skills, gaining hands-on experience with current software, and participating in system demonstrations. This gap is exacerbated by a lack of accessible, open-source resources and insufficient integration of translation technology training in existing translation programs (Bououden et al., 2024).

Some researchers emphasize that integrating translation technology training into higher education curricula faces numerous challenges, including cognitive, procedural, attitudinal, and institutional barriers (Kenny and Doherty, 2014). Moreover, the rise of generative AI tools, such as ChatGPT and DeepSeek, has introduced new complexities and challenges. While learners increasingly rely on these tools for quick translations, it is observed that many university translation students, in-service translators, and self-learners often receive inadequate training in the ethical and effective use of such technologies (Chan, 2024). These findings demonstrate the urgent need for open, accessible, up-to-date learning resources to equip language learners and professionals with the skills required to navigate the rapidly evolving digital landscape of the translation industry.

Massive Open Online Courses (MOOCs) have become a significant platform for open learning, offering free or subscription-based educational resources to a wide audience without prerequisites (Voudoukis and Pagiatakis, 2022). Classified by Pilli and Admiraal (2016) into four categories based on massiveness and openness, MOOCs range from small-scale paid courses to large-scale, fully open extended MOOCs (xMOOCs), which feature structured, teacher-led content. While MOOCs provide accessibility, flexibility, and learner autonomy, they face criticism for high dropout rates and low completion rates due to limited interactivity, insufficient support, and low learner motivation (Alj, 2024; Khalil and Ebner, 2014; Papadakis, 2023).

In terms of translation technology training, few MOOCs on CAT are available, such as outdated courses on SDL Trados Studio 2011 (Yu and Han, n.d.) and more recent offerings like the University of Washington’s “Localization: Adapting Digital Content to Local Markets” (dos Santos, 2025). However, these resources are often paid, lack institutional quality assurance, and are insufficient for self-learners and in-service professionals seeking.

The critique concerning the absence of real-time interaction and support in MOOCs has prompted the creation of alternative pedagogies, including blended MOOCs (bMOOCs) (Tang and Chan, 2024). bMOOCs combine online MOOC lessons with face-to-face interactions, aiming to address the shortcomings of purely online learning environments. This approach offers several advantages, including enhanced adaptability, promotion of interaction, and encouragement of active learning (Tang and Chan, 2024). bMOOCs, such as flipped classroom models, provide teachers with greater teaching flexibility and reduced time and location constraints, benefiting the effectiveness of both teaching and learning. Blended learning strategies are becoming more and more popular; however, there is still a lack of pertinent research on how learners’ background characteristics—such as gender, age, and prior learning experience—correlate with their satisfaction (Kumar and Kala, 2021). Moreover, despite their potential, bMOOCs have not garnered as much attention as traditional MOOCs, especially in the context of training with translation technologies. This is particularly crucial for English/Chinese translation, which serves the largest population and represents one of the most significant language pairs in the market. It is essential to explore how bMOOC approach can enable language learners to continuously update the digital competence needed for their evolving roles as professional language service providers.

Recent studies have highlighted the growing importance of blended learning approaches, particularly in the context of translation technology education. bMOOCs have been shown to enhance learner engagement and provide flexible, scalable solutions for skill development in specialized fields such as translation technology. For instance, Wang et al. (2024) conducted action research on flipped blended learning on competence improvement in translation technology training. The results of the mixed methods revealed that flipped blended learning enhanced interaction with peers, context and content and improved learners’ translation competence. However, there is a lack of studies underscoring the relevance of adopting a bMOOC approach in translation technology education; it is particularly important as it may not only address the diverse needs of learners but also align with the increasing demand for technology-driven pedagogies in higher education. By situating the current research within this under-researched context, the study aims to contribute to the literature on the effectiveness of bMOOCs in translation technology training.

The novelty and distinct contribution of this paper lies in its exploration of the application of a bMOOC approach in translation technology education, an area that remains under-researched despite its growing relevance in higher education. Unlike previous studies that primarily focus on traditional or fully online methods of teaching translation technology, this research emphasizes the unique advantages of blending face-to-face and online learning modalities to address the diverse needs of learners. By integrating technology-driven pedagogies with the flexibility and scalability of MOOCs, the study not only responds to the increasing demand for innovative educational frameworks but also provides empirical evidence on the effectiveness of bMOOCs in enhancing translation technology training. This contribution is particularly significant as it bridges the gap between theoretical advancements in translation technology and practical pedagogical approaches, offering a learner-centered model that aligns with the evolving landscape of translation technology training in higher education. A larger study (Chan, in press) using a mixed method approach investigated the overall impacts of a blended MOOC approach for translation technology learning (bMOOC-TT), including learners’ usage of digital tools and online resources as well as the learning outcomes of translation technology competence measured by self-assessment and performance in objective tests. While the previous analyses (Chan, in press) emphasize the evaluation of skill acquisition and knowledge gains, this paper focuses mainly on the affective perspective, i.e., quantitative analysis of students’ perceptions and satisfaction with bMOOC-TT.

The bMOOC-TT approach adopted in this study was based on Yousef et al. (2015) evaluation approach, an effective bMOOC combines elements of connectivist MOOCs (cMOOCs), xMOOCs, and face-to-face learning models. It was designed by leveraging the strengths of each model: the flexibility and student-centred network learning of cMOOCs, the interactivity and structure, instructor-led teaching of xMOOCs, and the immediate feedback and synchronous human interaction of face-to-face learning. This bMOOC was specifically for translation technology training, aimed at providing learners with both theoretical knowledge and practical skills. To achieve these learning outcomes, the learning activities were crafted to create an effective and integrated model that addresses common constraints of traditional MOOCs, such as ineffective feedback, limited interactivity between learners and instructors, and insufficient face-to-face engagement.

The online learning activities were open-access and primarily designed for experienced translators, translation students, translator trainers, language professionals, and anyone interested in learning about the application of translation technology in the English-Chinese and Chinese-English translation practice. The delivery mode and learning activities included seven in-person face-to-face workshops, four MOOCs, seven live online seminars via Zoom, and six guest talks. The materials were available in both Chinese and English. Each online seminar and onsite workshop lasted approximately 1 h. The MOOCs were hosted on Udemy, a popular online learning platform, providing unrestricted access to 22 custom-made videos consisting of 10 modules. These modules included a variety of learning resources, such as software demonstrations, video lectures, learning notes, reference readings, and hands-on activities. The modules covered the following 10 topics: (i) Introduction to Translation Technology, (ii) MT, (iii) CAT, (iv) Evaluation of MT Output, (v-vi) Translation in Phrase (formerly Memsource), (vii) Introduction to Trados, (viii) Introduction to Wordfast, (ix) Introduction to Localization and Other Tools, and (x) Comparison of Translation Technology Tools and Summary.

In the study, bMOOC-TT integrated online and onsite learning activities lasting four consecutive months. The participants were encouraged first to gain an overall understanding of translation technology concepts, development, and available tools through the four MOOCs, which included online course materials for self-directed learning. Following that, they participated in live online seminars that provided software demonstrations and synchronic communication between experienced trainers and peers in order to reinforce and deepen their understanding of the materials covered in the MOOCs. To enhance accessibility, the recorded online seminars were also made available. In addition to the instructors, various experts and scholars were invited to deliver a total of six synchronous webinars or in-person guest talks to offer more chances for real-time discussion, allowing participants, industrial practitioners, and researchers to share their insights. Following the conceptual components, face-to-face workshops were held in a language lab, allowing learners to access the technological tools discussed and gain practical, hands-on experience with various translation technology platforms. Figure 1 shows the training procedures and activities.

Utilizing an exploratory design, this research examined students’ perceptions of bMOOC-TT, focusing on its overall usefulness, satisfaction, and learning experiences in various learning activities. It also explored potential relationships between student background characteristics and their perceptions.

Regarding data collection, voluntary response sampling was employed to recruit participants, targeting full-time undergraduates, postgraduates, and distance-learning translation students from universities in Hong Kong. Invitations were sent via email. The sample size of 67 students was determined by the number of participants who completed all the required tasks, following email invitations sent to translation and language students in Hong Kong. While the sample may appear modest, it reflects a fully engaged cohort, ensuring that the data collected is both comprehensive and representative of students who actively participated in the bMOOC-TT learning activities.

To evaluate perceptions of bMOOC-TT, a questionnaire was adapted based on relevant studies (e.g., Yousef et al., 2015) and administered to the 67 participants via the online survey platform – Qualtrics, before and after their participation in bMOOC-TT learning activities. The questionnaire consisted of students’ demographic information, including gender, age, prior experience with translation technology, and previous MOOC learning experience. It also included their usage of digital tools for translation practice as well as their self-rated language proficiency and translation competencies. The questionnaire featured both open-ended response items and closed-ended questions.

The questionnaire aimed to gather participants’ opinions on the overall usefulness of bMOOC-TT, to investigate their satisfaction with various learning activities and topics, and to assess any changes in their perceptions and proficiency. The statistical analysis was also used to investigate the relationship between the learner variables and their perceptions. In the statistical analysis, a multiple regression model was conducted to examine potential differences in participants’ perceived learning experiences resulting from their background characteristics, including age, gender, prior experience with translation technology, and prior experience with MOOCs. All statistical analyses were performed by SPSS 27.0.

The perceived usefulness of bMOOC-TT was evaluated using a five-item questionnaire. Participants rated each item on a 5-point Likert scale, where 1 represented the lowest level of usefulness, and 5 represented the highest. The total possible score ranged from 5 to 30. As shown in Table 1, the average score was 4.5, indicating that learners generally found the bMOOC-TT to be useful and expressed satisfaction with it in various aspects, such as learning outcomes and motivation.

Regarding participants’ perceptions of various learning modes and activities, the highest rating was given to the “Face-to-face workshop” item, with a mean score (M) of 106.42 derived from item summation and a standard deviation (SD) of 20.62. Participants generally rated most other learning activities with an average score above 4 on a 5-point Likert scale, indicating positive attitudes towards all aspects. A relatively low rating was given to the guest talks, with a mean cumulative score of 104.14 and a standard deviation of 19.79. A summary of the mean scores for each item is provided in Table 2.

Regarding the perceived usefulness of various topics covered in the learning activities, participants responded to items using a 6-point Likert scale, ranging from 1 (lowest level of usefulness) to 6 (highest level of usefulness). In this study, participants’ scores ranged from 30 to 60, with a mean (M) of 53.66 and a standard deviation (SD) of 9.32. The reliability of the scale, as indicated by Cronbach’s alpha, was 0.98. Participants rated “Module 6: Translation in Phrase: Translation Memory” the highest, with a mean score of 5.45 and a standard deviation of 0.91. Additionally, participants gave an average score above 5 for all other topics, indicating a generally positive perception of the usefulness of the various topics covered. Relatively lower ratings were given to “Module 8: Introduction to Wordfast” (M = 5.30; SD = 1.09), “Module 9: Introduction to Grammarly, Passolo, QuillBot, and Wordtune” (M = 5.30; SD = 1.13), and “Module 10: Comparison of Translation Technology Tools and Summary” (M = 5.30; SD = 1.06). A summary of the mean scores for each item on this scale is provided in Table 3.

In terms of students’ perceived learning experiences, younger age group participants had better learning experiences and perceptions among all activities of this research project. In particular, younger age group participants had better learning experience and perceptions on MOOCs via Udemy (b = 22.79, 95% CI [12.49, 33.10], t (66) = 4.42, p < 0.001), face-to-face workshops (b = 22.98, 95% CI [10.88, 35.09], t (66) = 3.80, p < 0.001), real-time online seminars (b = 27.48, 95% CI [15.54, 39.43], t (66) = 4.60, p < 0.001), and guest talks (b = 21.87, 95% CI [10.27, 33.46], t (66) = 3.77, p < 0.001). Table 4 summarizes the above results.