Abstract:

In German higher education institutions and in higher education politics, the topics of the compliance to the required maximal duration of academic studies and the dropout rates are currently very important. The goal is to reduce the dropout rates and to help students not to stretch their study excessively beyond the regular study duration.

One approach to assist students in their study planning is to provide them with opportunities for a better self-assessment. These can be based on a review of the learning progress in individual subjects [1] or on a visualization of the data of university work of students in a peer group [2].

The solution presented in this paper is to provide a student with anonymized data on the university work of students during the past years. Using this data, he can see which percentage of students has shown similar learning behaviors in the initial semester like him. Additionally, he can seem which percentage completed their studies successfully.

However, the students from the risk groups will probably not be able to draw the correct conclusions about their own learning behavior from a representation of some mass data. Therefore, an interactive approach is described here, which helps the students to filter the comparison data and to analyze the correct data.

The procedure is as follows:
1. The student indicates his own semester.
2. For all past semesters, he enters for how many exams he has registered and how many of them he has passed or failed.
3. He then receives a graphical overview of the university work of students who have completed their first semesters with the same success rate. He can gain the following insights from the surveys:
a. What quota of all students have behaved like him in the past?
b. What quota of them has successfully completed the studies in which time and what quota was not successful (study break)?
c. In addition, he can carry out scenarios for the current semester and assess the chance for success on the basis of the successes of the peer student group.

The paper deals with the following aspects:
1. Detailed description of the data visualization based on concrete scenarios.
2. Experiences from practice: when do the surveys help and when not.
3. Discussion of possible further development.

References:
[1] B. Doersam, Use cases for learning analytics in self-learning courses, Proceedings of the ICERI 2016, Sevilla 2016
[2] A. Rieck, C. Metzger, M. Hinkelmann, J. Luessem, T. Seidl, Softwaregesuetzte Analyse von Studienverlaeufen – neue Grundlagen für Studienberatung, Qualitaets- und Lehrentwicklung, Tagungsband der Jahreskonferenz GMW 2015

Keywords:

Data visualization, university work, dropout rates.