Springe zum Inhalt

Veranstaltungsformat: Online

Der Inhalt des Vortrags soll uns die konkreten ethischen Probleme bewusst und greifbar machen und uns dazu anregen, uns an einem Reflexionsprozess über das richtige Handeln zu beteiligen. Dazu werde ich den Begriff der Datenethik erklären und auf den Zusammenhang zwischen Informatik und Ethik eingehen. Ich werde aufzeigen, welchen Weg diese zwei Disziplinen gemeinsam gegangen sind und wer sich aktuell mit Datenethik in Deutschland beschäftigt. Im Hauptteil der Vortrags setze ich mich konkret mit der Frage auseinander, an welchen Stellen der IT-Entwicklung welche ethischen Probleme auftreten können und durch welche Fragestellungen wir Ihnen begegnen können. Dafür habe ich den IT-Entwicklungsprozess in 5 Schritte aufgeteilt, die – unabhängig von dem gewählten Entwicklungsvorgehen – immer durchlaufen werden. Diese IT-Entwicklungsschritte habe ich mit konkreten und aktuellen Beispielen belegt und anschließend mit den ethischen Fragen ergänzt, die man jeweils stellen muss, um dem ethischen Risiko zu begegnen. Mit der Aussicht auf die Handlungsoptionen, die wir haben, werde ich meinen Vortrag beenden.

We analyse who falls victim of crime, the immediate and long term consequences of victimization for victims and their families and the heterogeneity in these effects using rare administrative data on the population of reported victims in Denmark. Victims are more likely to grow up in disadvantaged environments and to be of lower SES. Moreover, victimization has long-lasting negative effects on the labor market outcomes of victims, and these effects are larger and more persistent for female than for male victims. 

I will talk a bit on how we use machine learning in general in the area of labour market policy in DK, and how we relate this to our core business of producing results on employment and education.
As a specific example of our work, I will illustrate our statistical profiling of newly unemployed, both the technical/methodological side as well as the practical implementation and general experiences in this area, and some thoughts on further development.
Finally I will talk a bit on other more recent areas of developing datadriven solutions in the field of labour market policy, drawing perspectives to new possibilities deriving from machine-learning and modern Technology

Unemployment insurance systems in modern labor markets are riddled with a multitude of rules and regulations governing job seekers' economic situation and their incentives to search for employment. These include, for instance, detailed regulations specifying individuals' benefit level and potential benefit duration, job search requirements, conditions for avoiding benefit sanctions, possibilities for earning extra income or additional benefit entitlements by working in part-time or short-term jobs, etc. The complexity of UI systems makes it challenging for job seekers to understand the prevailing rules, their built-in incentives, and the resulting consequences for their personal economic situation. This is potentially problematic, as a lack of understanding may distort individuals' job search incentives and employment prospects.

In this paper, we report the results from a randomized controlled trial among the universe of registered Danish job seekers that studies how reducing complexity affects individuals' understanding of UI benefit rules and labor market behavior. Our intervention exploits an online information tool that provides individuals with continuously updated, personalized information on their remaining UI benefit period, their accumulated working time that can be used to prolong the potential benefit duration, as well as information on essential rules regarding job seekers' benefit duration and benefit sanctions. We match the data from our experiment with data from an online survey and rich information from administrative records to evaluate the causal effects of our intervention on individuals' understanding of the prevailing labor market rules, their job search behavior, and resulting labor market outcomes.

This talk will summarize two studies, which respectively study the role of caseworkers and public employment services for the labor market outcomes of unemployment benefit recipients. A first study asks whether and how much caseworkers matter for the outcomes of unemployed individuals. It exploits exogenous variation in unplanned absences among Swiss unemployment insurance caseworkers. A second study evaluates a large-scale policy change in which the public employment service of one Swiss canton changed its strategy by removing restrictions on job search and granting increased autonomy to job seekers.