Sharing a desk for one semester made friendship more likely
With randomly assigned seats in rural Hungarian schools, mutual friendship rose from just over 15% to just over 22%.
The key idea
Being near someone every day creates more opportunities for friendship, even when a lottery determines the seat.
The study, in plain language
Researchers randomized seating in 182 classrooms and analyzed 2,996 students. At the end of the semester, each could name up to five best friends. A friendship was mutual when both members of a pair named each other.
When two students had shared a desk, the estimated probability that they chose each other rose from just over 15% to just over 22%. The increase also appeared between boys and girls. For Roma and non-Roma students, the result changed depending on how missing data were handled, so it remained uncertain. It is also unclear whether sharing a desk created new friendships or helped preserve existing ones, or how long they lasted afterward.
What they found
- Students who shared a desk were more likely to name each other among their best friends.
- The increase also appeared when the two students were of different genders.
- The result for Roma and non-Roma students remained uncertain: it changed depending on how missing data were handled.
How they studied it
Seats were randomly assigned for a semester in 182 classrooms in rural Hungarian schools, using a preregistered analysis plan. At the end, students named up to five best friends; a pair counted as friends when both chose each other.
What the study cannot show
- Participating schools were rural and volunteered. The study does not distinguish new friendships from preserved ties or measure how long friendships lasted after the semester.
Something to think about
Think of a friendship of yours that began because someone happened to be nearby. This is a way to reflect on your own history, not a formula for making friends.
The original study
Proximity can induce diverse friendships: A large randomized classroom experiment
Source under CC BY 4.0. English adaptation and explanation by PaperFlow, with AI assistance. The text has been reorganized for a general audience; it is not the complete paper and does not imply endorsement by its authors.
One study alone does not represent a field's consensus. Educational content does not replace professional advice. Report a possible error.
