Abstract
Students interested in STEM professions are often unsure about exactly will be required of them upon graduation. Academic planning is made even more difficult by the fact that STEM fields evolve quite rapidly. In addition to technical qualifications, such as programming, employers increasingly deman soft-skills, such as communication, self-reflection, conflict management and team work. Underrepresented groups in STEM include women, ethnic minorities and students from non-academic families. Although they may lack role models to help them in academic planning, they can offer employers unique advantages. As minorities learning to integrate into the majority group, they may have learned to switch roles and see alternative perspectives. University job market portals could further aid STEM students from underrepresented groups. Ads placed by potential employers describe the competencies required for entry-level positions. Text mining of job ads enables the extraction of these competencies. A job market portal supported 15 German universities is analyzed using two different machine learning tools: linear regression and neural networks. The analysis of thousands of job ads over 10 years enables the identification of specific competencies desired by potential employers in STEM fields. By tracking the changes in employer demands, trends can be identified showing which skills are becoming more and less desirable over time. This analysis demonstrates how the relative importance of social competencies to technical qualifications changes over time. The probable future importance of individual qualifications can be predicted. This can help students from underrepresented groups in their academic planning.
Author Information
Karin Maurer, Technische Hochschule Nürnberg Georg Simon Ohm, Germany
Heidi Schuhbauer, Technische Hochschule Nürnberg Georg Simon Ohm, Germany
Patricia Brockmann, Technische Hochschule Nürnberg Georg Simon Ohm, Germany
Paper Information
Conference: IICEHawaii2020
Stream: Higher education
This paper is part of the IICEHawaii2020 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window
To cite this article:
Maurer K., Schuhbauer H., & Brockmann P. (2020) Machine Learning to Guide STEM Learning: Relative Importance of Social vs Technical Competencies for STEM Students from Underrepresented Groups ISSN: 2189-1036 – The IAFOR International Conference on Education – Hawaii 2020 Official Conference Proceedings https://doi.org/10.22492/issn.2189-1036.2020.9
To link to this article: https://doi.org/10.22492/issn.2189-1036.2020.9
Comments
Powered by WP LinkPress