Formative Assessment Practices to Evaluate Oral Proficiency of Undergraduate Learners with Digital Technology in Mozambique

An exploratory mixed-methods pilot study, based on Canale and Swain’s communicative competence model, evaluated technology-mediated formative practices at Save University (UNISAVE), in Mozambique, to investigate the institutional integration of Automated Grading Systems (AGSs) for asynchronous video-recorded presentations. The pilot adopted a convenience sample of 11 instructors (84.6% response rate) and 9 third-year undergraduates to validate the research instruments. Descriptive and thematic analyses were performed on quantitative survey data and qualitative interview narratives. Preliminary results indicate high levels of ecosystem alignment for low-barrier synchronous platforms while revealing critical systemic vulnerabilities in asynchronous workflows. Notably, 54.5% of faculty utilise idiosyncratic rubrics, creating conceptual ambiguity about criteria for multimodal delivery. Furthermore, manual evaluation of high-volume video files leads to intense cognitive fatigue, causing feedback delays of up to three weeks, thus interrupting the formative learning cycle. Rigorous data cleaning protocols also identified contradictions in participant responses, emphasising the need for active bias control. The study successfully validates instrument parameters for subsequent large-scale research, concluding that integrating rubric-driven AGS tools is a pedagogical imperative to eliminate feedback latency, alleviate administrative workloads, and preserve assessment consistency. A critical finding of this baseline phase is the absence of a standardised rubric for assessing oral proficiency in the distance ELT curriculum. This deficit detaches local practices from established global assessment frameworks, such as the Common European Framework of Reference for Languages (CEFR) and the TOEFL, which use validated proficiency levels (A1–C2) for construct development. To address this gap, the present study focuses on evaluating ‘sustained monologues’ and ‘extensive speaking tasks’ (Brown & Lee, 2015), operationalised here as asynchronous, video-recorded student presentations. While prior institutional research has primarily focused on synchronous, interactional oral production via videoconferencing platforms (e.g., Google Meet, Zoom), this intervention introduces a "hybrid" rubric that synthesises analytical criteria with holistic rating scales derived from the Cambridge ESOL Grid and the Oracy Assessment Toolkit. To align with the Mozambican national grading standard, the proposed instrument utilizes a 20-point scale with a weighted distribution designed to prioritize communicative competence over purely technical delivery: Content and Communication (5 points each), Grammar and Language Use (4 points), and Time Management and Delivery (3 points each). Subsequent phases of this research will subject this rubric to rigorous validation and pilot testing to verify its intra- and inter-rater reliability.