You probably know 611 people. Here’s how we know.
Asking how many Michaels you know unlocks all sorts of knotty social and demographic puzzles.
How many people do you know? You’ve probably never counted. Well, now you don’t have to. Tyler McCormick has worked it out: around 600.
Or more precisely 611, according to estimates by McCormick, a professor in the statistics and sociology departments at the University of Washington. That’s a national average, but McCormick can actually compute an estimate for you, or anyone. His technique is a fascinating illustration of the power of statistics to illuminate society: not just how many acquaintances the average person has, but the number of homeless and other hard-to-reach populations.
There are different degrees of how well you know people. This particular measure is a broad one.
Asked how many close friends they have, about half of Americans say three or fewer, according to a 2021 survey. The British anthropologist Robin Dunbar, drawing on studies of the brain sizes of humans and other primates, estimates a person can only maintain about 150 relationships. The so-called Dunbar number, he has said, “applies to quality relationships, not to acquaintances.” A Pew Research study found adults on Facebook had an average of 338 friends on the site.
Your acquaintances might help more than your friends
The number of people you know, without considering them friends, is probably much larger. McCormick’s definition: “that you know them and they know you by sight or by name, that you could contact them, that they live within the United States, and that there has been some contact” in the past two years.
This broader circle of acquaintances, as opposed to friends, matters quite a bit. The sociologist Mark Granovetter posited in a 1973 paper, “The Strength of Weak Ties,” that casual connections and acquaintances are more helpful when it comes to job hunting than close friendships.
LinkedIn demonstrated this with a five-year experiment in which its People You May Know algorithm randomly suggested that some users connect to others with strong ties—they had overlapping careers or contacts—and suggested that others connect to users with only peripheral connections. Sure enough, the latter group was more likely to find a new job.
Weak ties are tricky to measure. Who sits around and enumerates how many people they know?
Count your Michaels
McCormick as well as Matthew Salganik of Princeton and Tian Zheng of Columbia, his co-authors of the paper that introduced the estimate of how many people we know, devised a clever workaround. They asked respondents how many people they know named Michael, Stephanie, James or nine other common names.
There are more than three million Americans named Michael, around 1% of the population, according to the Social Security Administration, so Michaels should also make up 1% of your acquaintances. If you know eight Michaels, you probably know about 800 people.
Repeating the exercise with a dozen names yields a series of estimates that can be used to refine the answer. This approach was pioneered by Christopher McCarty at the University of Florida and co-authors in 2001. In addition to names, respondents were asked if they knew anyone who is Native American, a postal worker, on kidney dialysis, widowed, diabetic and so on. McCormick focused only on names, since you’re less likely to know your acquaintance’s job, ethnicity or if they have diabetes.
The real power of this fun little estimate is how it represents a cutting-edge statistical technique to shine a new light on our society.
Hard-to-reach populations
Asking about all those Michaels was originally part of an effort to measure hard-to-reach populations, such as the homeless. Standard techniques, such as a phone survey, are useless. But what if you ask a large sample of people how many homeless they know? If you can estimate the overall size of respondents’ social networks, then you can use their responses to estimate the share of the population that is homeless.
In recent years, researchers have used the technique to estimates things that are otherwise difficult or impossible to measure such as the number of sex workers, the frequency of the use of performance-enhancing drugs or drugs users at risk of HIV, or even religious identification.
The technique has shortcomings. As with most opinion polls, there is no way to verify whether the answers are correct, in the way population samples can be benchmarked to a comprehensive census, because there is no formal census of one’s acquaintances.
The relevance of a particular name or characteristic might vary widely among demographic cohorts. People named Michael tend to be younger than people named Robert or James. Traditional western European names might not be useful for people who don’t live in primarily European communities. Many of the behaviors that make someone hard to reach in the first place might also be unknown to their acquaintances. You wouldn’t necessarily know if a distant acquaintance sometimes used drugs. So researchers must try to account for how many of a drug user’s acquaintances would know about his or her drug use.
McCormick and others statisticians are still working to improve the technique, such as figuring out the ideal names to cover different races and ages.
The numbers aren’t perfect. Still, these relatively weak acquaintances are our best window into the job market, the breadth of our social ties, and many of our most-pressing social questions.
Write to Josh Zumbrun at josh.zumbrun@wsj.com
Catch all the Business News, Market News, Breaking News Events and Latest News Updates on Live Mint. Download The Mint News App to get Daily Market Updates.
Business NewsSpecialsYou probably know 611 people. Here’s how we know.
First Published:17 Nov 2023, 02:03 PM IST
- Top Gainers
- Top Losers
- 52 Week High