An analysis of dropout students in the German higher education system using modern data mining techniques
Year: 2021
Type: dissertation
Abstract: Due to the rising number of students enrolling at German higher education institutions in recent years and the growing demand on the labor market, which increasingly require a higher educational qualification, the topic of students' success and students' dropout becomes increasingly important. This thesis investigates the issue of dropout with data from the student cohort of the National Education Panel Study (NEPS) using modern methods of statistical data mining. The aim is to identify students at risk of dropping out of tertiary education at an early stage of study. An automated early warning system based on the statistical model can help to identify students at risk at an early stage of study and initiate appropriate measures to counteract the dropout process. In particular, ``soft'' determinants are identified, such as study satisfaction, which play a crucial role in the dropout process. In addition to the (time-dependent) prediction of dropouts, the main focus is on the identification of different types of dropouts and the prediction of the final grade.
Author: Giese, Marco
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W3189547295
