Effective legislation and enforcement in the area of road traffic safety has proved to reduce traffic injuries and fatalities in developed countries, yet such research has been limited in China, using driving-under-the-influence (DUI) as an example. Due to lack of high-quality data and primary research on driver decisions, the deterrent effects of DUI laws in China remain unclear, despite the criminalization of drunk driving since May 2011 and associated enforcement efforts by the government. Based on the classical theory of deterrence and recent advances in differential deterrence, this study as the third one in a series on legal deterrence, replicated a scenario-based conjoint experiment and established Hierarchical Bayes models for quantifying the unique effects of perceived certainty, swiftness, and severity of DUI punishment, as well as detecting variations in risk perceptions among general Chinese drivers. This replication study supported the usefulness of conjoint experiments for examining DUI decisions among drivers from different countries and groups, and helped for expanding the external validity of differential deterrence perspectives. It also provided conceptual and methodological insights into conducting experimental research and replication for the advancements of general public policy research in China.