DIGITAL LIBRARY
IMPROVING TEACHING OF MATHEMATICS THROUGH DATA MINING AND LEARNING ANALYTICS
1 Slovak University of Technology in Bratislava, Faculty of Materials Science and Technology in Trnava (SLOVAKIA)
2 University of Economics in Bratislava, Faculty of Econimic Informatics (SLOVAKIA)
About this paper:
Appears in: INTED2019 Proceedings
Publication year: 2019
Pages: 786-791
ISBN: 978-84-09-08619-1
ISSN: 2340-1079
doi: 10.21125/inted.2019.0276
Conference name: 13th International Technology, Education and Development Conference
Dates: 11-13 March, 2019
Location: Valencia, Spain
Abstract:
In Slovakia today, there is enormous interest in the education and graduation of more students in the Science, Technology, Engineering, and Mathematics (STEM) disciplines. Despite the importance of mathematical skills in quantitative disciplines, high failure rates in first-year university mathematics subjects have been observed in many parts of the world, in Slovakia as well. Therefore the goal of our research was set on how to use learning analytics for possible improvement of university learning outcomes of mathematics. In the described research, we use a combination of action research approaches and learning analytics. We have solved problems that arise from our pedagogical practice as is usual in action research but using quantitative methods that are used in learning analytics. We have used computer science, mathematics, and statistics to extract useful information from large volumes of data. We have gained knowledge that is useful for the further improvement of our educational practice.

The research has had two main objectives. The first was to find out whether the entry test could be used as an indicator of success in the Mathematics I course. The second primary objective was to find out whether it is possible to influence the predicted student success by appropriate a during-term intervention. To verify the stated hypotheses, we used a statistical approach and methods using the SPSS software. The AIS academic information system was used as a data source for research.

The group of respondents consisted of 324 first-year students from all study programs at the faculty. The final test was attended by 283 students, of these, 229 men and 54 women.

The results of the research showed that our entry test could be used as an indicator of future student´s success in Mathematics I (an indicator of success variable (p=0.066) was statistically significant at the significant level 0.1). On the other hand, the results of the research did not confirm the significant impact of the intervention (p=0.800), in the form of supplementary exercises, on the success of the students in the overall assessment of the subject. Although we expected that the outcome of the prediction could be affected by appropriate intervention, gained result is valuable to us as well. We have received information on the need to change the intervention so that we can increase the success of students in the subject of mathematics and consequently also increase the success of study programs at the Technical University.

We are currently preparing a new experiment, including changes of the intervention content and format, that we plan to take in the next academic year.
Keywords:
Mathematics, learning analytics, teaching of mathematics.