Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
Please note that all times are shown in the time zone of the conference. The current conference time is: 25th Nov 2020, 04:04:45pm SAST
Automated Generation of Test Questions and Solutions
Warren du Plessis
University of Pretoria, South Africa
While the automatic generation of test problems has been considered in the past, the importance of also generating solutions is largely unconsidered. Additionally, current approaches to automatically generating test problems are normally limited to multiple-choice questions (MCQs) and often only consider language-based problems. The generation of problems and solutions that are based on graphical processes is explored in an attempt to address these limitations. Rapidly generating questions and their solutions helps educators set better questions by allowing more iterations than manual processes, and allows students to be provided with worked examples. These and other considerations are illustrated by considering a number of examples, and suitable software tools are suggested.
Exploring student perspectives on assessment practices in an engineering context
Teresa Hattingh1, Laura Dison2
1North West University, South Africa; 2University of the Witwatersrand, Johannesburg
Background and motivation
Constraints in higher education can result in compromises in pedagogical alignment that can negatively affect student learning. Increasing student numbers and decreasing teaching capacity can drive assessment practices, in particular, further away from a learning-oriented paradigm. As more students fail, the impact of capacity constraints becomes more significant, and assessment practices can move in a direction that does not enhance learning and leads to a deterioration of student performance over time. If assessment practices and the quality of students’ approaches to learning are to improve sustainably, a thorough understanding of how assessment influences student learning is required.
The purpose of the study is to gain insights, through the voice of the student, into an engineering school’s assessment practices and how these shape student learning. The study adopts a theoretical lens based on assessment for learning (Sambell et al., 2013) and learning-oriented assessment (Carless, 2015) thinking.
This study used focus group interviews with 24 engineering students at a South African university in their second, third and fourth year of study where failure rates are still high even though students have already passed their first year of study. Four focus groups were conducted using maximum variation sampling and clustering students based on their academic performance: a low performing group, a mid-performing group, a high-performing group and a turnaround group who had transitioned from low performance to high performance. Focus groups were chosen to facilitate reflection and exploration of experiences and to allow students to construct ideas in a collective way. The interviews were analysed using thematic content analysis and considered the individual voice, collective voice and interactions within the groups.
Results and Conclusions
The findings are presented using descriptive narratives that provide a rich picture of the current state of assessment and learning in the School. The findings confirm that assessment practices determine the planning and prioritisation of study efforts and student learning strategies. The intention of students when choosing learning strategies is strongly influenced by their perceptions of the assessment environment, with students often exhibiting surface approaches to learning which undermines the development of lifelong learning skills. Although students find themselves adopting studying for passing behaviours in a predominantly assessment of learning environment, there is evidence that they have a strong desire to move towards practices that develop deeper levels of understanding. However, it was found that the authenticity and relevance of the curriculum and a detached social environment in the School influences student engagement and motivation. It emerged that in a South African context a more collectivist approach is needed when considering how students learn and adapt to learning environments. A more collaborative teaching and learning environment is recommended to shift learning in a more positive direction. This study provides valuable insights that can be used to transform assessment practices, providing a better structure for student access, engagement and success in engineering, higher education contexts.
Carless, D. (2015b). Excellence in University Assessment. Oxon: Routledge.
Sambell, K., McDowell, L., & Montgomery, C. (2013). Assessment for learning in higher education. London: Routledge.
"Assessment as learning" as a tool to prepare engineering students to manage ill-defined problems in industry
Bronwyn Claudia Swartz
Cape Peninsula University of Technology, South Africa
Under-preparedness of engineering students to solve ill-defined problems in industry (Pan, Strobel and Cardella 2017) is a significant shortfall in Higher Education (HE). The root cause of under-preparedness is, ‘problems’ that need to be resolved in workplace settings in the real world are notably different to traditional textbook problems presented in class. Therefore, a transformation in pedagogical approach is needed in HE. Industry problems are often more complex and are commonly poorly defined. Moreover, Arifin, Zulkardi, Putri, Hartono and Susanti (2017) argue that a critical factor to the success of organizations in the fast-changing global economy, is the ability of industry-based engineers to demonstrate independence and initiative in problem-solving and develop practical and innovative solutions. Against this backdrop, a redesigned assessment strategy in an Applied Statistics class for final year students at a University of Technology (UoT) presented an opportunity to re-think traditional assessment methods.
“Assessment as learning” is defined by Jones (2019) and Manitoba (2016) as the use of ongoing self- and peer-assessment by students in order to monitor their own learning. This process is characterized by students reflecting on their own learning and making adjustments so that they achieve deeper understanding. Thus, it may be offer a solution when it is used as a tool to assist students during the process of finding solutions to ill-defined problems, thereby make them more valuable and sought after by industry.
In this study an assessment strategy was designed as an intervention which included a series of online self- and peer-assessments (formative online assessments under guidance of the lecturer) which culminated in development of a final industry style report with a potential solution to the problem. Prior to commencement of this study, ethical clearance was obtained through institutional channels. Guided by the Directorate for Learning and Assessment Programmes (2019), three constructs associated with assessment ‘as’ learning were examined in a mixed method study which included a survey (n=31) and a focus group interview (n=6) for a target population of 61 students. The constructs were, 1) student metacognition; 2) teacher role in assessment ‘as’ learning and 3) the role of feedback in assessment ‘as’ learning.
Results indicate that learning took place took place when students worked in groups (67,8%) however less students (43,4%) reported that self-development took place during this project due to factors such as technological challenges. Significantly however, students were more readily able to recognise self-development with group assessments as opposed to individual assessments. Finally, this study also highlighted some of the barriers to assessment as learning in an online context.
Ultimately this study confirmed students were able to meet learning objectives while concurrently undergoing personal growth in their critical thinking, technical competency, communication skills and overall academic mindset. The paper concludes with implications and limitations of the study and recommendations for future research.