IMPLEMENTATION OF STACK-BASED AUTOMATED ASSESSMENT IN NON-MATHEMATICAL DISCIPLINES: ANALYSIS OF INSTITUTIONAL SUPPORT MODEL
TTK University of Applied Sciences (ESTONIA)
About this paper:
Conference name: 18th International Conference on Education and New Learning Technologies
Dates: 29 June-1 July, 2026
Location: Palma, Spain
Abstract:
The paper presents the experience of developing and implementing a system for designing assessments based on STACK questions for non-mathematical disciplines at Tallinn University of Technology (TTK UAS). STACK (System for Teaching and Assessment using a Computer Algebra Kernel) is a specialized question type in the Moodle learning management system that enables automatic assessment of mathematical and numerical answers using computer algebra. Developing STACK questions requires familiarity with at least basic programming. For many subject-matter experts, this presents a real obstacle, yet the system enables the creation of assignments that closely resemble traditional paper-based problem-solving tasks. A key advantage of STACK is its feedback mechanism: detailed, automated, and tailored to typical student errors, helping students spot and address gaps in their understanding.
Initially, the use of STACK was limited to mathematics courses, where lecturers had already developed the necessary competencies for designing such questions. However, for lecturers from other disciplines, independent development of STACK tasks proved challenging due to time constraints and the need for programming knowledge. In response, a centralised STACK group was established at the university. The group consists of mathematics lecturers who provide methodological and technical support to lecturers from different institutes. An organisational model was developed to manage the entire workflow, covering the submission of requests, task design, technical implementation, and Moodle integration. This model has broadened the use of STACK beyond mathematics and enabled automation of practice-oriented student assignments.
This study uses a qualitative approach to analyse the implementation of the support model. An online survey with open-ended questions was conducted with five lecturers from four institutes (construction, geodesy, chemistry, and physics) to evaluate the model, identify organisational barriers, and understand how lecturers experienced the shift from traditional to automated assignments.
The results confirm that a small, centralised group is sufficient for institution-wide scaling of STACK. At the same time, the level of involvement of non-mathematical disciplines was lower than expected, which highlights the need for active and sustained promotion when introducing digital assessment tools at the institutional level.Keywords:
STACK, automated assessment, Moodle, digital assessment, higher education.