Martin, what actually is policy experimentation?

Policy experimentation could be summarized as a systematic process to assess the impact and effectiveness of policies based on solid evidence. The experimentation aims to ground policy decisions and policy deployment in observable results, based on sound methodology and valid evaluation - rather than on using a trial-and-error approach or on trusting someone’s "gut feeling".

What does it look like in TrainDL?

In TrainDL we focus on developing policy recommendations aiming to integrate data literacy and artificial intelligence educational competencies into the training for pre- and in-service teachers. This process relies on evaluating the legitimacy and actionability of policy recommendations through hypotheses generation, stakeholder engagement, target group interventions as well as associated quantitative and qualitative evaluation of the expected impact. In TrainDL we perform three experimentation cycles of intervention, evaluation, and recommendation generation - for teachers of computer science, followed by STEAM teachers and finally primary teachers.

What is the most exciting part of policy experimentation for you?

In the TrainDL project I am part of the Austrian Computer Society’s team responsible for the work packages Policy Building and Policy Recommendations. Since I have a background in the development, implementation, and evaluation of educational concepts in the area of AI, computational thinking, and robotics I find it very exciting to bring in my expertise on a more strategic level – the policy experimentation process. And of course, another exciting aspect is the interdisciplinary as well as the European context of this project!