Human-Robot Collaboration (HRC) is a form of direct interaction between humans and robots. The objective of this type of interaction is to perform a task by combining the skills of both humans and robots. HRC is characterized by several aspects, related both to robots and humans. Many works have focused on the study of specific aspects related to HRC, e.g., safety, task organization. However, a major issue is to find a general framework to evaluate the collaboration between humans and robots considering all the aspects of the interaction. The goals of this paper are the following: (i) highlighting the different latent dimensions that characterize the HRC problem and (ii) constructing a conceptual framework to evaluate and compare different HRC configuration profiles. The description of the methodology is supported by some practical examples.