Great questions. I hate that I’m linking to a wikipedia article but it’s actually the best overview of Christakis and Fowler’s research, with links to the original article:
I may have used casual in two different ways in the article. If so, my apologies. The first causal refers to the idea that since the data was collected over 30 years, they could demonstrate cause and effect (as opposed to just demonstrating correlation). In terms of social norms, I probably should have clarified that its a possible explaination for the effect. Then there are also possible explaination for why it hovers around three degrees.
Much more research to be done. At this point, Christakis and Fowler haven’t really done any new research on it and likely don’t plan to. However, the broader body of research on emotional or behavior contagion and social influence is still a popular subject, so hopefully we see this revisited by other scientists soon.
