Extracting Communication Networks - Aral and Van Alstyne 2011

[jamesallenevans]

Post questions here:

Aral, Sinan and Marshall Van Alstyne. 2011. “The Diversity-Bandwidth Trade-off.” American Journal of Sociology 117(1): 90-171.

[jsmono]

This research is a bit challenging for me to absorb due to the amount of math involved but the result and discussions are quite inspiring. Compared to other articles using emails as the corpora, this essay's explanation of using emails and its analysis is much clearer and convincing. In the discussion section, the authors talked about the implications of the research and I'm wondering if some of their findings can be applied to areas outside of the industry and workplaces. For example, the idea of "weak, structurally diverse ties provide information at a lower rate, with less frequency, less complexity, and more delay. Weak ties are more likely to deliver obsolete information. " seem valuable for analysis of gossips and rumors.

[gracefulghost31]

Would parsing out the diversity component in the diversity-bandwidth frame yield more insights? In this context, would there be an appropriate instrumental variable to make a causal claim about the relationship between access to information and performance?

[vahuja92]

The way the authors measure novel communication with k-means clustering is really interesting. That being said, the final conclusion the authors come to - that weak ties offer less novel information because they happen less often, feels intuitive and has already been discussed in non-computational social science research. I would be interested in understanding what types of people are able to maintain a high bandwidth of communication with weak ties. Presumably recruiters are good at doing just that - could differences in the characteristics of recruiters change the conclusion that close ties over more volume of novel information simply because of more interaction?