Exploring students’ beliefs about web-based collaborative learning and their practices: a qualitative case study of university English-as-a-foreign-language readers

Web-based learning (WBL) leverages web resources and platforms to enable flexible, time- and place-independent learning, including collaborative learning through internet-supported platforms. Prior work highlights WBL’s affordances (e.g., flexibility, peer interaction, self-efficacy) and its importance when face-to-face learning is disrupted (e.g., pandemics). While students’ beliefs (evaluative stances) are known to shape and be shaped by practice, most WBL research has used quantitative designs that capture correlations at single time points, offering limited insight into how beliefs evolve and relate to practices over time, particularly in EFL contexts. Existing qualitative studies are few, often short in duration or reliant on single data sources. This study addresses these gaps by qualitatively examining, over one semester, how university EFL readers’ beliefs about web-based collaborative learning (beliefs about the platform and about reading activities) develop, what factors moderate these beliefs, and how beliefs relate to practices. The purpose is to inform researchers and educators seeking to support students’ cognitive engagement and optimize WBL adoption in language learning.

Learners’ beliefs are evaluative stances toward learning that can be consciously or unconsciously held and are sensitive to past and ongoing experiences. Beliefs are multi-layered, comprising main beliefs and sub-beliefs, and can be positive, negative, or mixed; their net positivity reflects interactions among sub-beliefs. Positive beliefs generally motivate practice, and negative beliefs constrain it, though the belief–practice link is dynamic and can be moderated by contextual factors (e.g., distractions, knowledge limitations, alignment with context). In WBL research, numerous benefits have been reported (e.g., flexibility, interactive learning, enhanced self-efficacy). However, most studies on students’ beliefs in WBL are quantitative, focusing on correlations at single time points and offering limited understanding of belief dynamics. Limited qualitative work (e.g., interviews only; short durations) hints that beliefs can become more positive with experience and support (e.g., teacher assistance, alignment with context), but does not trace nuanced, longitudinal interactions across belief layers. This study expands the literature by using multiple qualitative data sources over a semester to track the evolution and interaction of EFL students’ beliefs about web-based collaborative learning and their relationship to practice.

Design: Qualitative case study of a semester-long (4 months) EFL reading course at a Chinese university, focusing on one cohort (n = 20) to examine beliefs about WBL over time. Context and WBL activities: An internet-based collaborative platform (China-based, functionally similar to Google Docs) enabled synchronous/asynchronous co-authoring and editing; all activity was auto-saved and visible to peers. Two recurring out-of-class tasks were implemented for each of 10 course texts (linguistics, language acquisition, cross-cultural communication): (1) pre-class critical reading questions (logic within/between paragraphs, stance/evidence, meaning organization) and language-focused questions; (2) peer assistance (answering, commenting). Students could engage anytime/anywhere before instructor-set deadlines; in-class debriefs addressed questions and extended discussion. Participants: Twenty first-semester English-major undergraduates (all female; Chinese; intermediate English readers by departmental placement). All had prior synchronous online lecture experience but no prior WBL. Their pre-university reading instruction was test-driven, emphasizing surface-level meaning, grammar, and vocabulary, contrasting with the course’s emphasis on deep, critical reading. Data collection: Multiple sources over time: (a) Semi-structured interviews—each student was interviewed three times (~48 minutes; in Chinese): mid-semester (eliciting initial/baseline beliefs retrospectively), end-of-semester, and one month after the second interview (to complement prior data). (b) Two reflective essays per student (in Chinese; ~2000 words each): one before the first interview and one at semester end. (c) Researcher field notes throughout, documenting platform activity, peer interactions, and self-initiated questions. Analysis: Ongoing, iterative inductive thematic analysis with open coding across transcripts, reflections, and field notes, complemented by deductive attention to evaluative language resources (appraisal framework; Martin & White, 2005) to identify beliefs (emotions, judgments, appreciations) and their polarity/intensity. Codes were refined with reference to prior WBL and EFL reading literature, grouped into categories, and synthesized into themes. Data sources addressed three research questions: (1) nature of beliefs over time (interviews, reflections), (2) moderating factors (interviews, reflections), and (3) belief–practice relations (interviews, reflections, field notes).

- Beliefs evolved through four phases across the semester: (1) Early phase: Overall slightly negative beliefs about WBL; platform beliefs neutral (operation positive; pedagogical value neutral/negative); reading-activity beliefs negative (could handle: negative; should master: neutral); little interaction between domains; practices limited to required tasks, often near deadlines. (2) Middle phase: Platform beliefs became positive (both operation and pedagogical value), which in turn gently bolstered reading-activity beliefs; reading beliefs were mixed with slight positive dominance; practices showed increased peer assistance and deeper critical engagement. (3) Mitigation phase: Positivity was intermittently dampened by dynamic factors—temporary technical issues (internet/platform glitches), peer pressure from public comparison, slow progress for some, and irregular difficulty spikes due to unfamiliar topics for others; domains partially offset each other (taking turns to restore positivity); mismatch observed between some positive beliefs and suboptimal practices; some messy platform use or relapse into surface reading occurred. (4) Stabilization phase: Stable coexistence of sub-beliefs within both domains with dominant positive sub-beliefs and inert, coexisting negatives; a new meta-belief about embracing learning challenges emerged, helping muffle negatives; practices became regular and timely, with consistent deep questioning and peer responses despite occasional challenges. - Moderating factors: Prior test-driven learning and initial lack of WBL experience; ongoing positive/negative experiences on the platform; technical reliability; peer dynamics (visibility, comparison); text/topic difficulty versus students’ knowledge repertoires; academic context pressures (e.g., GPA); teacher mediation; emergent beliefs about coping with challenges. - Belief–practice relationship: Generally positive beliefs drove engagement, and negative beliefs constrained it, but the relationship was dynamic and sometimes mismatched due to contextual mediators (e.g., external constraints, identity/role awareness, emergent coping beliefs).

The study demonstrates that EFL students’ beliefs about web-based collaborative learning are multi-layered and dynamic. Net positivity shifts over time reflect not only changing experiences but also interactions between two main belief domains (platform use and reading activities) and their sub-beliefs. Early on, positive platform beliefs catalyzed improvements in reading-activity beliefs; mid-course, domains took turns buffering each other during setbacks; later, both domains stabilized with dominant positive sub-beliefs and inert negatives. This nuanced, longitudinal interaction extends largely cross-sectional, quantitative WBL literatures that treat beliefs as static or correlate them at single time points. Multiple emergent factors moderated beliefs and practices, including prior educational experiences, technical reliability, peer comparison, and task/topic difficulty. While positive beliefs typically aligned with stronger engagement, mismatches arose when contextual constraints intervened, underscoring the vulnerability of the belief–practice link to internal and external moderators. Pedagogically, simplifying and supporting platform use, making students’ literacy progress visible, providing ongoing training and individualized support, and allowing time for adjustment can nurture positive beliefs and sustain engagement.

This qualitative case study traced semester-long dynamics in EFL students’ beliefs about web-based collaborative learning and their relation to practice. Beliefs about platform use and reading activities evolved from slightly negative/mixed to stably positive, with dominant positive sub-beliefs coexisting alongside muted negatives. Changes were mediated by prior and ongoing experiences, technical and social factors, and an emergent belief about embracing learning challenges. Overall, a positive correlation between beliefs and practices was observed over time, albeit mediated by contextual factors that sometimes produced mismatches. The study contributes longitudinal, multi-source qualitative evidence about how belief structures interact and stabilize in WBL for EFL reading, informing design and support strategies to foster sustained engagement. Future work could extend the observation period, diversify WBL affordances and reading task designs, and examine broader learner populations to refine instructional tailoring.

- Context specificity: One university EFL reading course and cohort; findings may not generalize to other disciplines, levels, or cultural contexts. - Duration: One semester; longer-term trajectories (e.g., across an academic year) were not observed. - Scope of WBL affordances: Focused on collaborative reading activities; other WBL features and task designs were not systematically varied and warrant future exploration.