Network Canvas: An Innovative and Intuitive Network Data Capture Tool for Prevention Research

Introduction: Since disease transmission and prevention often includes a social dimension, health researchers have a substantial interest in capturing social network data to develop an epidemiological understanding of diseases, and to inform intervention development. However, the capture of network data remains a challenging pursuit for many researchers. Several methodological barriers exist in the capture of complex data – and many of the solutions to these barriers require technical expertise and resources beyond the capabilities of social and behavioral health researchers. Network Canvas is an NIH-funded project that hopes to simplify and streamline social network data collection.

The Network Canvas Software Suite has been made possible by a few recent technological advancements, such as the emergence of cost-effective capacitive touchscreen displays, the increasing power of standardized web technologies, and the development of graph databases. All of these allow our team to significantly improve the usability and decrease the complexity of network data capture for both respondents and researchers.

Methods: The design of Network Canvas draws heavily from the field of human-computer interaction (HCI) for optimal ease, engagement, and comprehension through a tactile and simplified user interface. For example, our design allows redundancy to be minimized (i.e., smart skip logic can be easily embedded). And unlike others, our tool allows for multiple dimensions of complexity (e.g., collection of individual, network, and geospatial data) and a back-end graph database (i.e., neo4j) built to efficiently store and manage complex data.

Results: A preliminary investigation into the performance of Network Canvas showed that its use results in a similar number of sexual network members being named, compared with traditional partner-by-partner elicitation methods, but in a fraction of the time.

Conclusions: Network Canvas is being designed to be both free and completely open source; this will allow all members of the prevention research community to actively utilize and contribute to the Suite. Additionally, as we wish this to be a tool developed for and by the community, large portions of our project are dedicated to conducting needs assessments and obtaining feedback from health and social and behavioral researchers who have strong interest in capturing social network data. This input even at early stages will allow us to design a flexible tool that will meet the needs of a broad array of researchers.