Making Friends

Hall asked 355 adults who had recently moved, and then 112 first-year students, how many hours they had spent with new acquaintances. People called each other casual friends after about 40 to 60 hours, friends after 80 to 100 hours, and close friends after more than 200. Hours spent working together counted for less than hours spent relaxing together.

Limits of the study: The first study asked 355 adults who had moved in the last six months to estimate their hours with a new acquaintance. The second followed 112 first-year students for nine weeks. Both were in the United States, and both relied on people's own estimates of time. The studies show that closer friends had spent more hours together. They do not show that spending the hours will make any particular person your friend.

For this exercise: People estimated their own hours, and everyone was in the United States. The study shows that closer friends had spent more time together. It does not promise that the hours will make a particular person your friend.

Open the paper ↗

On their first day, 54 new students were given seats at random. A year later, students who had sat next to each other, or in the same row, rated their friendship as stronger than students who had sat further apart.

Limits of the study: This was one group of 54 first-year psychology students in Germany, who were given seats at random in a single introductory session. It shows that a chance meeting can matter a year later. It does not show how large the effect is for adults outside a university.

For this exercise: This was one group of psychology students at one German university. It shows that being near someone by chance can matter a year later. It does not tell you how strong the effect is for adults.

Open the paper ↗

People who expected to be accepted behaved more warmly when they met someone new, and warmer people were liked more. In a second study, the researchers made rejection seem less likely. People who usually expected the worst then became warmer, and they were liked more too.

Limits of the study: One correlational study and one experiment, both with university students meeting someone new in a lab. The effects were measured within a single meeting. The paper does not show that telling yourself you will be liked produces the same warmth.

For this exercise: These were university students meeting once in a lab. The researchers changed the situation, not what people told themselves. Saying the sentence to yourself is our own step.

Open the paper ↗

In two studies, people reported better days when they talked more with acquaintances, as opposed to close friends. This was true when comparing different people, and also when comparing the same person's different days.

For this exercise: The studies found a link, not a cause. They do not show that adding a chat improves your day, or that an acquaintance will become a friend.

Open the paper ↗

In a series of experiments with more than 5,900 people, the person who sent a message guessed how much it would be appreciated, and the person who received it said how much they appreciated it. Senders consistently guessed too low. The gap was larger when the message was a surprise.

Limits of the study: A series of experiments with more than 5,900 people. An independent team repeated one of the experiments with 742 people and found the same result, but the effect was small. The studies measure how much a message is appreciated. They do not measure whether the friendship starts again afterwards.

For this exercise: An independent team repeated one of the experiments with 742 people. They found the same result, but the effect was small. The studies measured how much people appreciated the message, not whether the friendship started again.

Open the paper ↗

People who wrote letters of thanks overestimated how awkward the other person would feel. They underestimated how surprised and how pleased the person would be.

Limits of the study: The experiments used written thanks between people who already knew each other. They measure how badly the writer predicts the other person's reaction. They do not measure whether the thanks changes the relationship afterwards.

For this exercise: The letters went to people the writers already knew well. We are applying the finding to newer friends and to acquaintances, and that is our own step.

Open the paper ↗

In five studies, people underestimated how much the person they talked to liked them. This is called the liking gap. The number you give yourself in the moment is not a measurement.

Limits of the study: The studies used conversations between strangers and new acquaintances, mostly students. They show that people underestimate how much they were liked. They do not show that correcting for this error changes how the next conversation goes.

For this exercise: The studies measured the gap. They did not test what a follow-up message does. Using the gap as a reason to write first is our own step.

Open the paper ↗

In lab and field studies, people underestimated how often others would say yes to a direct request, by as much as 50%.

Limits of the study: The requests were for help, such as directions, borrowing a phone or a favour. They were not invitations. The result held across those settings, but the study says nothing about asking somebody out.

For this exercise: The study looked at requests for help and favours, not invitations. Applying it to invitations is our own step.

Open the paper ↗

Pairs of strangers answered 36 questions that became more personal as they went on. Afterwards they felt closer than pairs who made small talk.

Limits of the study: The method was designed to create closeness between strangers in a lab in about 45 minutes. The authors say clearly that it was not designed to make people fall in love.

For this exercise: The pairs were students who talked for 45 minutes in a lab. The method was designed to create closeness in one sitting. It does not show that the closeness lasts.

Open the paper ↗

This is our own judgement. We know of no study that tests whether organising a regular plan makes friendships form faster. We take it from Hall's finding that friendship follows hours spent together, because a plan that repeats is the most reliable way we know to collect those hours.