Methodology

Wordcloud EN cropped

The foundation of CASE is a vast, proprietary dataset that has been continuously updated and improved since 2012. It is this data that, combined with innovative analysis, enables us to put academic performance into the context of each individual program.

CASE knows the grade distributions for almost every higher education program since 2004. Additionally, the CASE Score makes use of a multitude of different characteristics of German students such as IQ estimates and personality traits. This information is being collected in the study series "Fachkraft 2020". Fachkraft was initiated by the two founders of CASE, Dr. Jan N. Bergerhoff and Dr. Philipp K. Seegers, as part of their academic research at the universities of Bonn and Maastricht. The study has so far included more than 200.000 students and is the biggest of its kind in Germany.

In addition to analyzing academic achievement the CASE Score also incorporates psychological characteristics that have been found to be predictive of labor market performance. We include a number of cognitive and non-cognitive characteristics of students in the different programs, which we analyze using methods and techniques inspired by findings of contemporary labor economics research.

Scientific Approach

Distribution

CASE uses modern econometric methods to combine a multitude of data sources. This is the basis for our analysis of each individual candidate’s academic performance. To account for the huge variety of study programs we distinguish not only between different institution, but also between study fields, degree types and the respective graduation year.

One aspect of our process is to compare the applicant’s GPA with the distribution of GPAs earned in the student’s field at his or her school. The candidate’s relative position is then adjusted so as to account for differences in student competitiveness between different programs. We distinguish not only between different institutions, but also between study fields, degree types and the respective graduation year. The weighting of these different factors is primarily based on data that is exclusive to CASE, which includes individual information of 200,000 German students. Using this data the CASE algorithm was optimized using the actual biographical data of the students surveyed by us.