Michael Striewe received his PhD in 2014 for automatic evaluation of programming and modeling tasks. Since then he has been working on organizational, technical and didactic aspects of Technology-Enhanced Assessment. In the interview he talks about the e-assessment system JACK and the advantages of e-assessment not only in times of pandemics.  

Michael, what can Jack do? 

JACK is an e-assessment system that supports digital teaching by enabling formative and summative assessments. Formative examinations are e. g. practice assignments that I take while learning to check if I am making progress and understanding what I have learned. A summative exam is the classic written exam at the end of a learning phase. Here we test whether the learning process was successful. JACK supports both forms but focuses on formative assessments.  

What is it about formative exams that makes them so fascinating? 

Students should really profit from e-assessment during their learning. On the one hand, the right tasks must be set; on the other hand, it is important that the learner receives the best possible feedback from the system. JACK alone cannot do it. But it offers teachers the opportunity to set subject-specific tasks and write individual feedback for typical wrong answers. 

Who is using JACK? 

In our working group “Specification of Software Systems” we developed JACK 15 years ago for the automatic evaluation of programming tasks – hence the name JACK as an abbreviation for Java Checker. Later on, we also used the system for our mathematics preparatory courses, where first-semester students refresh their math skills before studying. In the meantime, different faculties across the board at the university have been using  JACK  , e.g., business administration, chemistry, biology, linguistics and engineering sciences. The placement tests for the language courses at the Institute for Optional Studies also run on JACK. 

What are the advantages of e-assessment for students? 

First, of course, learning independent of time and place, which is an important argument right now. From my point of view, the biggest advantage for students is that they receive direct feedback on their individual performance. Ideally, JACK is my sparring partner when I am learning, giving me many different tasks until I really get a hang of the material. 

And the benefits for teachers? 

Teachers save time mainly in the supervision of standard problems. In most subjects there are some typical mistakes that almost 50 percent of the people make at the beginning. Once I have prepared this with JACK, that means written good feedback on these classic errors, I can focus on special cases. I can go into the statistics of JACK, look for other mistakes and think about the reasons. This way, I can further develop  the tasks so that the students can better learn from their mistakes. 

Sounds like a lot of work. 

Yes, there is a lot of effort in the beginning to build a task pool. But it pays off in the long run.  

Your working group is constantly developing the system. What else do you want JACK to learn? 

JACK already offers many possibilities to set tasks very flexibly. But there is always something to do with special, subject-specific types of tasks. We have implemented, for example, a technical solution for chemists to design tasks with automatic feedback on specific atomic models. For biologists, we have further developed JACK so that botanical identification exercises can be carried out with it. 

We are currently investigating, among other things, what Artificial Intelligence can do in JACK. As soon as speech recognition works reliably, we could automatically evaluate free written texts. JACK can still learn a lot and the possibilities are far from being exhausted.  

Information about JACK and the research of the Working Group “Specification of Software Systems” in the field of e-learning and e-assessment:  https://s3.paluno.uni-due.de/forschung/e-learning-und-e-assessment