A Framework to Identify Students at Risk in Blended Business Informatics Courses: A Case Study on Moodle

Abstract

Purpose: Students at risk is a cardinal problem. Students' failure has a negative impact on many areas. Students are more liable to fail in courses that call for specific skills' development. A typical paradigm is the Business Informatics courses which combine digital and business skills. The research objective is to address the problem of students' failure in blended Business Informatics courses by identifying students who are liable to fall through. Design/methodology/approach: Students' data in regard to enterprise and informatics skills were analyzed in terms of a binary logistics regression with a view to developing a model to identify students at risk. A binary variable was modeled to describe students at risk and students not at risk. The students' data constituted the independent variables in our regression analysis whereas the variable describing students at risk was the dependent variable. The students' data was collected by well-designed students' learning activities on Moodle. The data was collected after the first course-run. The regression analysis outcome was a classification table indicating students at risk. Findings: The number of laboratory exercises completed along with the self-test's assessment completed were the main risk factors for this Business Informatics course. Given that the laboratory exercises were implemented through Moodle and were explained during conventional laboratory lectures, it appears that both the e-learning part and the conventional part plays a significant role in students' critical achievement in terms of this specific Business Informatics course. In parallel, factors related to practical skills' development (laboratory exercises completed) 1 have appeared to assume a cardinal role in students' final learning outcome. Originality/value: The originality of this research lies in the fact that the issue of identifying students at risk is not addressed in a fragmentary way by just carrying out a specific analysis and coming up with results, like many similar studies in the literature. Thereby, a concrete methodology was developed on the basis of an established generic risk management framework. Therefore, the identification of students at risk is included in the phases of a potent framework. The added value of this research is centered on the fact that this model could potentially be applied to any Business Informatics blended course in order to come up with the respective risk factors. In parallel, this model could be verified by being applied to a plethora of Business Informatics blended courses with a view to generating a prediction model for students at risk for Business Informatics blended courses that share the same learning design.

Developing a Risk Model to Control Attrition by Analyzing Students’ Academic and Nonacademic Data

“…After the analysis of students' academic data based on students' engagement, we can deduce that the real risk factors, which have significant contribution to the reduction of the risk probability in the context of both courses, is the "Number of Self-Assessment Quizzes completed". This finding is in line with many studies (Georgakopoulos et al, 2018;Zakopoulos et al, 2021;Zakopoulos, 2022).…”

“…The academic data could reflect students' behavioral engagement, setting out in the students' interaction with the learning activities (Bujang et al, 2021;Macfayden and Dawson, 2010;Georgakopoulos et al, 2018;Georgakopoulos and Tsakirtzis, 2021;Zakopoulos et al, 2021;Zakopoulos, 2022;Anagnostopoulos et al, 2020;Alyahyan and Dustegor, 2020). The non-academic data could reflect demographic data such as gender and age, along with students 'emotional engagement (Marks, 2000).…”

“…We also modeled the binomial variable student risk to describe students who were about to fail the course as it is suggested in some studies (Macfayden and Dawson, 2010;Georgakopoulos et al, 2018;Anagnostopoulos et al, 2020;Georgakopoulos et al, 2020;Zakopoulos et al, 2021;Zakopoulos, 2022;Tsakirtzis and Georgakopoulos, 2020). The state '0' was modeled to indicate students not at risk whereas the state '1' was modeled to indicate students at risk.…”

“…In that field, a lot of studies have proved that students' interaction with the learning activities plays a significant role in their final learning outcome. Therefore, students at risk could be identified by a proper analysis of their engagement (Georgakopoulos et al, 2018;Zakopoulos et al, 2020;Anagnostopoulos et al, 2020;Georgakopoulos and Tsakirtzis, 2021;Tsakirtzis and Georgakopoulos, 2020;Macfayden and Dawson, 2010;Zakopoulos, 2022). These studies have also indicated that the risk factors are course-oriented.…”

“…These models could analyze various data sets (including academic and nonacademic data). The academic data could refer to students' interaction with learning activities and the entire learning process, pointing up students' behavioral engagement (Bujang et al, 2021;Macfayden and Dawson, 2010;Georgakopoulos et al, 2018;Georgakopoulos and Tsakirtzis, 2021;Zakopoulos et al, 2021;Zakopoulos, 2022;Anagnostopoulos et al, 2020;Alyahyan and Dustegor, 2020;Davidson et al, 2018;Dias et al, 2020;Filippidi et al, 2010). The non-academic data could include data related to students' emotional engagement along with financial and social data (Ismail et al, 2018).…”