This correlation can imply causation in two ways.
The one no one mentions is that stupid people may have fewer friends.
The one no one mentions is that stupid people may have fewer friends.
It's the other way, in my experience.
People tend to be friends with other people of similar intelligence and since intelligence follows the normal distribution, people with most friends are of average intelligence.
Income also (mostly) follows intelligence, meaning less intelligent people have it tougher, meaning it is harder for them to set boundaries at work. This in turn makes them accommodate wildly differing opinions as they can not pick their coworkers and some degree of harmony is a must in the workplace.
Meaning they are trained to get along with college graduated engineers as well as flat earth crackpots. Thus they draw from a much wider pool of potential friends.
It is handy to be friends with an old landlady, despite her believing in the healing power of minerals in your pocket. She still has the cash and the spare room to throw the party.
Especially when she regularly hires your flat earther husband to paint her flats as the college student tenants graduate and move on.
To determine causation we'd have to have a set of isolated people and a set of social people and for each set we deliberately pick half and raise the isolated people to be more social while making social people become more isolated. Then we record IQs. Only this experiment will determine causation.