Credibility in Empirical Economics Research (WiSe 2026/27)

Veranstaltungszeitplan

Beschreibung

Course Description:

This seminar takes up a question that has moved to the centre of debate across the empirical social sciences: can we trust published research? Starting from the concern that the published record may be a distorted reflection of the underlying evidence, we examine the statistical and institutional forces that produce this distortion, as well as the reforms developed in response.

The seminar opens with two lectures that introduce these issues conceptually. Topics include the anomalous distribution of p-values across the literature, the replication crisis, multiple testing and statistical power, and the practices that have emerged in response — among them preregistration, replication, and result-blind evaluation.

Building on this foundation, students move into an independent work phase with two components. First, each student writes a summary of one of a small set of core papers on research credibility. Second, all students work on the same dataset and the same research question. Comparing the results that different students arrive at makes visible how the many discretionary choices involved in data processing and analysis — the "forking paths" of empirical work — can shape reported findings. This multiverse exercise turns the abstract concerns of the opening lectures into something students can observe and measure directly. Each student documents their work in an individual report.

The seminar is aimed at Master students interested in empirical methods and in how credible knowledge is produced in economics, and who would like hands-on experience assessing and conducting empirical research.

Dozenten

• Professor Dr. Andreas Stegmann (verantwortlich)