A person's achievement is claimed to converge on that of the people they spend time with.
The confound is the whole difficulty. People who associate are alike for reasons that precede the association, so observed similarity is mostly selection. Studies using randomly assigned roommates and squadrons do find genuine peer effects on effort and conduct, at magnitudes far below what the maxim implies.
In practice
- A team whose norms are high produces higher output from the same people, and the effect is smaller than the folklore.
- A person who changed circles and changed habits may have changed circles because they were changing habits.
- Possible explanations
- The explanation offered is transmission of standards and habits. The competing explanation is selection: people who associate are alike for reasons that precede the association.
- In its favour
- Peer-effects research using random assignment, in roommates and military squadrons, does find real influence on behaviour and effort.
- Against it
- The measured effects are small and specific compared with the claim, the confound of selection accounts for most observed similarity, and the evidence offered is testimonial and selected on success.
- Standing
- Testimonial evidence selected on success, with no comparison group. Peer-effects research does find influence on behaviour, with effects that are real, modest and heavily confounded by selection: people choose their associates.
- Subject
- Creativity & Work
- Essay
- The success literature and its sceptics