Mark Zuckerberg has been on a podcast tour promoting Meta’s new AI projects and one of his core claims is that “the average American has fewer than three friends.”
After listening to these podcasts, I kept coming back to that stat because it didn’t sound believable. I don’t know anyone who has fewer than three friends. (But I guess no one will know these people and they will exist like dark matter, only visible in survey data.)
From looking into it, there are two popular studies that (kind of) support the claim that the average American has fewer than three friends. The AEI did a study on friendship in 2021 and Pew released a similar report in 2023 that both asked the same question — “Not counting your relatives, how many close friends do you have?”.
In those studies, half of people said they had 3 close friends or fewer. But there are a couple serious issues with that stat.
First, many people are close to their family and extended family so it doesn’t make sense to totally cut out these relationships when thinking of close friends. You wouldn’t say you are starving because you have no food to eat (excluding the food already in your fridge).
Second, and I think this one may have an even larger effect, there is no clear definition of a “close friend”. I spent time reading the survey materials and neither survey clearly defines what a close friend is.
When I asked people I knew how many close friends they had, I would usually get a number much lower than I expected (often close to 3). But when I asked people more about their response, people explained that they were only counting the innermost circle.
When people answer a survey for “how many close friends do you have?” — they mentally rank their friends in tiers based on how close they are. Then they pick the people in the first tier and count those as their “close friends”. This makes the stat worthless because two people’s friendship tiers won’t be the same. One person could have two incredibly close friends they live with while another could have five close friends they see every other week.
The rest of the survey data seems to support my interpretation. When the surveys asked people if they were happy with the quantity and quality of their friendships, the vast majority of people reported that they were quite happy.
I think it’s valuable to build community and help create friendships. But it’s inaccurate to claim that the average American is starved for friendship.