The design of building elements is usually done conservatively by considering safety factors. However, more efficient designs are gaining interest for economic and sustainability reasons. Hence, an adequate prediction tool can improve the design of building elements. Probabilistic modeling, for example, Monte Carlo simulations, represents a remedy to this by examining uncertainties in a structure through uncertain input parameters. In this work, a Monte Carlo simulation is performed to quantify the uncertainty in the modal properties of a hybrid steel–timber building element. The material properties of the timber material and the stiffness of the structural joints are considered uncertain inputs. The probabilistic properties of the timber material are evaluated utilizing Bayesian inference instead of the usually applied empirical methods. Using these inferred timber material properties leads to a good match of simulated and measured natural frequencies of the timber components. These parameters are utilized together with the joints’ uncertain inputs in the Monte Carlo simulation of the hybrid steel–timber building element. The results show a significant span for the identified eigenfrequencies, which proves the relevance of probabilistic analyses for the vibration characteristics of building elements.