What makes someone popular? It is a conundrum that has long preoccupied schoolchildren, writers of teenage fiction and producers of Hollywood high-school comedies. But researchers have now come up with an empirical answer: 18 friends.
Academics from the Centre for Research on the Wider Benefits of Learning found that the average secondary pupil has 17 friends. Anyone with a larger number is therefore above averagely popular.
The researchers surveyed more than 1,500 secondary pupils in order to determine the friendship patterns of the average teenager. Nearly half the pupils said that they had 21 or more friends in school, while fewer than one in 10 reported having no more than five friends.
Boys were far more likely than girls to claim that they had large numbers of friends, but when called on to list them by name, girls were usually able to provide a longer list than boys.
The handful of pupils who had very small friendship circles tended either to have special needs or to be unusually academic. They were also mainly girls. But they were not unhappy or isolated: many said that they had very close friends.
"Having no or very few friends may not be negatively related to wellbeing," the researchers said. "There may be gender differences in friendship preferences and the value gained from them."
Having few friends did not, in itself, seem to be a problem. By contrast, being perceived by classmates as someone with very few friends was a definite social handicap.
One Year 10 girl told the researchers: "The people who do more schoolwork than socialising aren't very popular."
Friendship circles were clearly divided by sex: there was virtually no crossover between boys' and girls' networks.
Most pupils seemed to prefer to spend time with people from the same ethnic background. A fifth of pupils said most or all of their school friends shared their ethnicity, and two-thirds said that around half of their friends were from the same background.
However, this varied by ethnic group. A third of Bangladeshi pupils said that most of their friends shared their ethnicity, as did similar numbers of Pakistani and black African pupils. This proportion was significantly greater than the numbers of white British and black Caribbean pupils who primarily spent time with people from the same background.
Boys were far more likely than girls to choose friends who shared their own ethnicity. None the less, school friendship groups were more diverse than out-of-school groups. More than half of the girls from Asian backgrounds said that most or all of their friends outside school shared their ethnic background.
The researchers said: "This suggests the particular importance of school as a site of social mixing."
They recommended that schools work with pupils, to try to encourage a sense of social belonging.
But they added: "It is not necessarily clear that loose ties with many people offer greater value than bondedness with a smaller number. Quality of friendships may be key."