Communication and Social Networks (Fall 2023)/Travers summary

The paper reports on an experimental study by Jeffrey Travers and Stanley Milgram investigating the "small world problem" - the idea that society is highly interconnected with short paths linking any two people. The study tested this by tracing acquaintance chains from arbitrarily selected "starting persons" to a designated "target person" in Boston. The starters were from Nebraska and Boston. Each starter received a document to mail to the target. They could only send it to a first-name acquaintance who might advance it closer to the target.

64 of the 296 chains reached the target, demonstrating the feasibility of the method. The mean chain length was 5.2 intermediaries, indicating a high level of social interconnectedness. Chains from the Boston starters were significantly shorter than Nebraska ones, showing the effect of geographic proximity.

Funneling of chains through common intermediaries was observed - 3 people accounted for 48% of completions. Participants also showed strong tendencies to select demographically similar recipients based on age, sex and occupation.

The results provide insight into the structure of acquaintance networks and interconnectedness in a large population. The method appears useful for empirically studying social networks. Extensions could involve varying starter and target characteristics.

Overall, the paper makes both a methodological contribution in demonstrating the viability of tracing acquaintance chains, and provides empirical evidence for the small world hypothesis that society is highly interconnected.