CatColab for Model Building
Nate Osgood, together with 4 students from the Computational Epidemiology and Public Health Informatics Lab in Saskatoon, Canada, recently ran a community group model building event focusing on the drivers for homelessness in their city. This post describes the event, and the next steps that they are planning to take.
Together with 4 students from our Computational Epidemiology and Public Health Informatics Lab (CEPHIL), I spent my Friday at one of our community group model building event, this one focused on drivers for homelessness in our city (Saskatoon, Canada).
Although our behavioural ethics review board stipulated that the group should not include people who are currently homeless, the participants were people with lived experience of homelessness, with most having personally experienced homelessness within recent years.
Building on facilitated discussion, the focus of the day consisted of a group model building session. To allow for some diversity of thought and exploration, the participants divided into two teams. The causal loop diagrams resulting from about 1.5-2 hours of work on the part of each team were very thoughtful, and the diagrams captured many important insights and perspectives, and lived experiences. It bears emphasis that while some of those from our lab helped facilitate the discussions, the identification of variables and the existence, directionality and polarity of the links between such variables came firmly from the participants with lived experience themselves. Although they are not yet suitable for public distribution, I thought that I would provide a glimpse of the work products.
For the next stages of this work, CatColab will be a key tool—and arguably the single most important tool in our toolbox to secure substantive insights and value from these diagrams. Building on the strong applied category theory experience of 3 of the 4 students involved, we will be using CatColab to find feedback loops in these diagrams individually, and then when combined. This ability to find feedbacks in this fashion will be a tremendous asset for learning from these diagrams. It will also be used to visualize and explore the diagrams, although it will not be the only tool to serve in this capacity. When the ability to compose causal loop diagrams is added to CatColab, we plan to make central use of that feature as well.
Friday’s event is the first in a series focusing on this pressing problem in our community through tapping the deep and grounded knowledge of those with lived experience. It is a great testimonial to the power of CatColab that it will play such a central role in the value delivery from such events. We hope to contribute to the development of CatColab to further its ability to deliver insights and benefits not only to our research team, but also to community members themselves. I wish to express my—indeed, our—gratitude to the core CatColab team for their delivery of such a valuable tool for insight into complex social issues such as homelessness (together with cognate issues such as mental health, domestic violence and substance use, lack of affordable housing), and CEPHIL’s committment to contributing to the development of that tool to further develop its potential in this sphere.