Web 2.0 technologies and translator training: assessing trainees’ use of instant messaging as a collaborative tool in accomplishing translation tasks

The paper examines how Web 2.0 technologies—especially instant messaging (IM)—shape collaborative practices in translator training, noting that IM is ubiquitous, portable, available, locatable, and multimodal. Given modern workplaces’ collaborative nature, IM has become core to translators’ communication with peers, project managers, clients, and subject experts. Calls in the field advocate aligning translator training with industry realities and integrating Web 2.0 tools within project-based learning (PjBL) to foster autonomy, critical thinking, and collaboration. Despite consensus on the importance of information exchange in collaborative translation tasks, there remains a gap regarding detailed accounts of how IM interactions occur, the roles interlocutors play, relationships between IM practices and performance, and strategies to optimize IM use through moderation and guidance. This study addresses these gaps by analyzing real-time IM exchanges, in-class presentations, and questionnaire responses from 68 MTI trainees working in teams on translation, localization, and language service provision tasks, focusing on IM frequency, timing, modality, conversation types, roles, themes, and problem resolution, and exploring correlations with final assessment outcomes.

The paper situates its investigation within PjBL for translator training, where collaboration is central and communication skills are critical. Prior work highlights both benefits and challenges of computer-mediated communication. Reported challenges include distraction during conversations, time-consuming exchanges, lack of teamwork experience, varying readiness of trainees and trainers, difficulties for introverted trainees, observation/access limitations, and handling culture-bound terms. Studies comparing high- vs low-performing teams suggest early starts, consistency, organization, and deep exchanges correlate with higher performance, while inconsistent, shallow exchanges correlate with lower performance. Methodologically, earlier studies relied largely on questionnaires, observations, presentations, reports, and interviews; raw IM exchange data have been underexplored. Within the industry, there are calls for increased collaboration using mobile messaging platforms. The present study addresses this research gap by examining real-time IM transcripts to understand communicative practices in group tasks under PjBL.

Design: Mixed-methods study combining quantitative and qualitative data. Participants: 68 first-year MTI trainees (58 female, 10 male) at a Chinese university, enrolled in a mandatory Language Services and Project Management (LS&PM) course over 16 weeks. Nine self-formed teams (8–9 members each). Tasks: Three collaborative tasks: (1) Translation (approx. 2500 Chinese characters into English) using Smartcat with MT and post-editing; objectives included terminology management, glossary creation, layout accuracy, deadlines, and project management. (2) Localization of a Chinese restaurant menu for a New York context (POEditor), adapting for local tastes (e.g., vegan prevalence, low spice), persuasive descriptions, design improvements, and justification in class. (3) Language Service Provider (LSP) design and launch via a WeChat mini-program; included service design, marketing language, social dissemination, client tracking, and project management. Technology and communication: Class and team WeChat groups created; IM (WeChat) was the primary communication mode alongside occasional face-to-face meetings. Only raw IM exchanges were analyzed here. Procedure:

• Pre-tasks (3 weeks): Lectures on localization, translation, project management, language services; hands-on tool training; simulated short group tasks; quiz; practice using IM in class WeChat group with instructor guidance.
• Task execution: Instructions and timelines via class WeChat; teams collaborated via team WeChat groups and offline as needed; instructor available for guidance.
• Reporting: Weekly in-class progress reports, roles, challenges, resolutions, teamwork atmosphere, and peer sharing; instructor provided advice and recorded role data.
• Assessment: Translation task assessed with modified Hurtado Albir rubrics; localization assessed on translation/adaptation quality and presentation justifications; LSP assessed on design, business language, services, client responsiveness, and project management. Data collection: (a) Questionnaire (8 Likert-type items) on team camaraderie, IM practices, problems, resolutions, attitudes; average completion time 73.23 s; Cronbach’s alpha = 0.89. (b) In-class presentation notes and interviews on organization and roles. (c) Voluntarily provided IM transcripts: 4,544 IM turns totaling 78,138 Chinese characters/English words, submitted as WordPad files. Data preparation and coding: IM turns defined as a single sent line per participant. Synchronous exchanges defined as replies within <60 seconds; asynchronous otherwise. Two assistants verified and color-coded categories: frequency (blue), themes (red), time slots (yellow), IM type (pink), synchronous (orange), asynchronous (green), problem resolution (gray). Multi-category turns received multiple adjacent codes. Tallies produced per team means and SDs. Analysis: Quantitative analyses in SPSS: descriptive stats, Pearson correlations, t-tests; significance thresholds reported. Thematic analyses of interviews and IM content to contextualize practices.

• Communication frequency and performance:
  • Top-performing teams (1, 7, 8, 9) exchanged the most IM turns: 887, 848, 756, and 678 respectively (>50% of all IM turns). Bottom-performing teams (2, 3, 4, 5, 6) exchanged 357, 297, 279, 255, and 187 respectively.
  • Strong positive correlation between teams’ percentage of total IM exchanges and final assessment scores: r(9)=0.97, p<0.001.
  • Task-specific IM volume correlated highly with final scores: localization r(9)=0.98, p<0.001; translation r(9)=0.98, p<0.001; LSP r(9)=0.84, p<0.004.
• Timing patterns:
  • Most IMs were exchanged on Monday (lesson day, 20%) and Tuesday (17%); least on Friday; volume rose again on weekend as class approached.
  • Time-of-day: unpopular 6:00–8:59 and 00:00–5:59; some teams preferred late evening (21:00–23:59). Significant differences among time slots at 9:00–11:59 (p<0.001), 12:00–14:59 (p<0.001), and 18:00–20:59 (p<0.001).
• Synchronous vs asynchronous communication:
  • Overall: 71% synchronous (3,228 turns) vs 29% asynchronous (1,317 turns).
  • Team extremes: Team 1 had 83.2% synchronous; Team 3 had 60% synchronous.
  • Correlations with performance: synchronous r(9)=0.74, p=0.022 (positive; not statistically significant at conventional alpha), asynchronous r(9)=-0.88, p<0.002 (significant negative).
• Roles and organization:
  • High-performing teams identified and consistently filled permanent roles (team lead, terminology manager, timekeeper, secretary; plus data miner for some tasks). Low-performing teams largely used ad hoc role assignments, typically only a consistent team lead.
  • Positive correlations between number of roles and final assessment: total roles r(9)=0.89, p<0.001; localization roles r(9)=0.94, p<0.001; translation roles r(9)=0.74, p=0.022; LSP roles r(9)=0.94, p<0.001.
• Conversational themes:
  • Task-related themes dominated; IM volume by task: LSP (33%), localization (31%), translation (26%).
  • Non-task-related themes included gossip (up to 14% in Team 4), entertainment (13% in Team 2), personal issues, and scheduling (not strictly task content but organizational).
  • Correlations with final score: scheduling r(9)=0.46, p=0.20; personal issues r(9)=0.35, p=0.36; gossip r(9)=0.16, p=0.68; entertainment r(9)=-0.32, p=0.40. Though not significant, scheduling showed the highest positive association; entertainment showed a negative association.
• IM multimodality:
  • Overall composition of 4,544 turns: text 87% (3,954), images 4% (204), emojis 7% (313), documents 2% (74); voicemails not captured.
  • Team variation: e.g., Team 7 had many images (12%) and documents (6%); Team 9 had no pictures but 11% emojis; Team 6 used no emojis, favoring text (76%), images (12%), documents (21%).
  • T-tests showed differences among forms; significant differences observed for text (p=0.001) and emojis (p=0.005).
• Challenges and resolutions:
  • Communicative issues: unwillingness to engage, poor time consciousness, dislike of IM, laziness, interpersonal frictions. 62% reported all team members got along; 1% reported none did.
  • Technical issues: sharing large files, upload/download problems, logo design choices, translating technical terms; resolved via online file sharing (IM/email), online/offline discussions, troubleshooting, Internet searching, and role exchange.
  • Number of problem-resolution strategies positively correlated with performance: r(9)=0.90, p<0.001. Top-performing teams used more (around four) strategies; most low-performing teams used at most two.
• Task performance levels: Average final scores were LSP 89.3%, localization 83.3%, translation 82.4%.

The study demonstrates that IM is an effective collaborative tool in translator training within a PjBL framework. High-performing teams communicated more frequently, adopted synchronous exchanges more often, organized themselves with clear and consistent roles, discussed a broader set of relevant themes, and employed multiple problem-resolution strategies. IM exchanges clustered around instructional periods (lesson day and the following day), suggesting alignment with course rhythms and that timely, focused IM use can enhance task progress. Synchronous communication facilitated brainstorming and topic-focused dialogue, which was associated with better outcomes than asynchronous messaging. Organizational structures—particularly permanent roles such as team lead, terminology manager, timekeeper, and secretary—were associated with higher performance, highlighting the importance of defined responsibilities and workflow management. While non-task-related themes occasionally fostered camaraderie and mitigated tension, excessive entertainment-related exchanges correlated negatively with performance; scheduling-related discussions, while not task content, appeared to support coordination and thus performance. Text remained the dominant modality, with selective use of images, documents, and emojis; modality preferences varied by team, and significant differences across forms were observed. The results support providing IM usage guidelines in PjBL settings—covering schedules, role allocation, topic focus, moderation, and documentation of challenges/solutions—to optimize collaborative efficacy without undermining learner autonomy and critical thinking central to PjBL.

The study investigated how MTI trainees used instant messaging to collaborate on translation, localization, and language service provision tasks within a PjBL framework, analyzing real-time IM exchanges, presentations, interviews, and questionnaire responses. It found that top-performing teams exchanged more IMs, favored synchronous communication, maintained permanent roles, discussed more themes, and used multiple problem-resolution strategies. IM practices correlated with final assessment performance across tasks. Based on these insights, the authors propose integrating IM into translator training through customizable guideline templates provided before tasks, including conversation schedules, role assignments (fixed and ad hoc), prioritized discussion themes, moderation to maintain task focus, language preferences, and logging of challenges and resolutions. Such guidelines can improve efficiency, engagement, and focus while preserving the autonomy and problem-solving ethos fundamental to PjBL. Future research should expand participant diversity and delve deeper into technical issues not fully captured by IM transcripts.

The study involved a single cohort of 68 participants from one MTI program, limiting generalizability. It focused primarily on communicative aspects; technical issues were only partially captured, and voicemails could not be exported from WeChat. Future studies should include more participants across multiple programs and more comprehensive assessments of technical challenges beyond what appears in IM transcripts.