Assessing the Impact of Hierarchy on Understandability

This experiment has been designed to empirically validate the claims made in our framework assessing the impact of hierarchy on model understandability (see here). In particular, we elaborated hierarchical BPMN models and semantically-equivalent BPMN models without hierarchy. Participating subjects had then to answer questions about particular aspects of the models. To analyse the answers, we use the automated data collection infrastructure of Cheetah.


The experiment makes use of repeated measurements. In order to counterbalance learning effects, two experimental groups were defined. Group 1 (code 3563) received the hierarchical model first, group 2 (code 9174) started with the flat model. To retrieve the experimental material, please download the pre-configured version of Cheetah here. By entering the codes mentioned above, Cheetah will either present you with the material for group 1 or group 2.