This dataset contains results of two controlled experiments with students in TU Delft on the comprehensibility of tabular and graphical risk model representations. The dataset contains the raw data collected with Qualtics survey platform, the dataset processed with responses validated against a baseline of correct responses. The dataset is related to the following publication: Labunets, K. (2018). No search allowed: what risk modeling notation to choose?. In Proceedings of the 12th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (p. 20). ACM.