Examining the Link Between Low Self-Control and COVID-19 Measures Adherence: A Specification Curve Analysis Approach

Translating criminological theory into empirical tests requires explicit assumptions, clearly defined estimands, and transparent analytical decisions. This study revisits the association between low self-control and rule adherence by applying a multiverse analytical framework to nationally representative panel data on adherence to COVID-19 regulations in Germany. Using specification curve analysis, we assess whether the relationship between lower self-control and reduced adherence remains consistent across a wide range of theoretically defensible modelling choices. The results indicate a robust association, although effect sizes vary substantially depending on how low self-control is operationalised, while the inclusion of covariates has comparatively little impact. These findings demonstrate that ambiguity in theoretical interpretation, unclear estimands, and researcher degrees of freedom can meaningfully shape empirical conclusions. We argue that multiverse approaches provide a structured way to enhance transparency and improve the credibility and reproducibility of criminological research by making analytical decisions more visible and interpretable.