Experimental Design
We built on two interventions implemented as field experiments into the Berliner-Studienberechtigten-Panel. In the first intervention, participants were initially randomized at the school level for an information intervention within schools in spring 2013 as part of the Best Up-Project registered under BestUp (AEARCTR-0004586). The second intervention exploited the rich existing information of students within the Berliner-Studienberechtigten-Panel and at the end of the Best Up-Project in 2016, participants were transferred to the PostGrad-Project registered under PostGrad (AEARCTR-0002446), which comprised a second information intervention randomized at the student-level. While these two projects studied the short- and medium-term effects of each intervention on educational decisions, in particular on enrollment in undergraduate studies, on primary choice, and subsequent enrollment in postgraduate studies, we are interested in the labor market outcomes of these initial high school graduates, who graduated in 2014. More precisely, we are mainly interested in participants' earnings, occupational choices, employment spells, outcomes measuring job (mis)match, comparing current earnings profiles to the expectations measured prior to undergraduate enrollment, and firm-level outcomes.
BestUp (AEARCTR-0004586): The information treatment in the Best Up-Project, the initial project, was randomized at the school-level, where in nine schools out of 27 schools an information workshop took place of about 20 minutes in length. At the end of this workshop, a 3-minute video summarized the presented information. Students participating in the workshops received information about the costs and benefits of studying compared to vocational education and training (VET). The material comprised information about costs and benefits of undergraduate studies compared to VET in terms of (i) income, (ii) occupational positions, and (iii) unemployment risks, which were based on scientific results. In addition, students learned about major-specific earnings and gender earnings differences and about financial opportunities to fund an undergraduate degree (student financial aid, stipends, etc.).
PostGrad (AEARCTR-0002446): In the PostGrad-Project based on the data of the Best-Up-Project, the information intervention was randomized at the student-level, drawing from rich pre-PostGrad-treatment information in the data from the Berliner-Studienberechtigten-Panel. Students in the treatment group were provided with information about the costs and benefits of postgraduate studies, such as information about (i) income, (ii) occupational positions, and (iii) unemployment risks, which were based on our calculations using the German Microcensus. This treatment was administered online and was embedded within the baseline questionnaire. This ensured that participants read the provided information. Questions that followed the treatment comprised only statistical facts, such as gender, birth year, and birth month.
To examine the long-term treatment effects of these information provisions, we included a question in the surveys within the Berliner Studienberechtigten Panel that allows us to link the survey data with the registry data of the Federal Employment Agency (IAB).
For those participants who answered yes, i.e., provided their consent to this question, the Research Data Center (RDC) of the IAB links their survey data with information available in the registry data.
For our new project, we will receive access to a project-specific and password-protected working space of the IAB's RDC at the end of June/beginning of July 2025 (after submitting this registration report). We expect a potential sample size of nearly 400 adults who consented to the above question and were traceable in the registry data. We will estimate the long-term treatment effects for both high school graduates who pursued a bachelor’s or master’s degree and for those who pursued vocational education and training directly after high school graduation.
Regarding the IRB, we refer to the IRB approval for projects AEARCTR-0004586 and AEARCTR-0002446, which include the information interventions we built upon.