Stefan Wuchty posits that proximity data over time correlates strongly with social norms, and acts as a valid measure of social connection between individuals. Additionally, the technology used to gather this data (Bluetooth proximity) is superior to previous studies, which rely heavily on self-report measures.
Begin with what is assumed to be common knowledge from the field of sociology -- that a social tie is a relationship between two or more people that regularly communicate with each other, or identify each other as a friend, or in which one person has a tendency to successfully approach the other for advice.
Unfortunately, the methods required to collect this kind of information are time-consuming and resource intensive. Traditionally, researchers rely on interviews, surveys, or other forms of self-report to determine who is social tied to whom. Due to these methodological constraints, measures of social ties are often limited to small groups, capture only a brief snapshot of a phenomenon that changes over time, and lack any objective behavioral data.
Comparing the length of proximity, researchers found significant differences in the self-reported assessment and the actually measured proximity between all pairs of subjects.
To compensate, Wuchty introduces a new breed of multimodal analysis by blending mobile sensors worn by human subjects with computational models. As a result, he created dynamic and time-sensitive maps of human interaction. Wuchty compares these new maps to measures of self-report and questions the limitations and merits of each, including the inevitable legal and privacy issues surrounding the treatment of personal behavioral data. Finally, the author concludes that blending observed data such as spatial and temporal proximity is a parsimonious and valid method for determining social ties in large-scale social networks and qualitatively superior to self-report methods.
Supporting evidence is derived from relatively new electronic methods of counting behavior, which allows scientists to compare observed information tracked over time to self-reported information. The author also points to multiple studies produced in the last 11 years to support his claim. These reports come from a mix of computer science, sociology, human-computer interaction, and engineering research articles. He then laments the lack of a holistic view of communication and behavior, as each study he cites only claims knowledge of a tiny keyhole perspective on a single technology such as email, sms, or phone records.
It is telling that none of his sources are developmental, behavioral, or social psychology journals. Why is it that empirical evidence of how people behave relative to each other should be so conspicuously absent from a discipline that claims to be concerned with the human mind and behavior? How might these new methods inform or compete with existing psychological models of human behavior in groups?
Source:
Stefan Wuchty “What is a social tie?”
Wuchty, S. (2009). What is a social tie?. Proceedings of the National Academy of Sciences, 106(36), 15099-15100.