Talk:Learning and Evaluation/Archive/Program Evaluation and Design/Resources/Pilot Reporting Items

Study of Networking effects

I wish to add a suggestion to another field of study. We hold conferences where a large number of persons converge and share knowledge, and network with others to find potential collaboration partners. I think it is intuitively understood that these are essential and justified. It costs a lot to organize such meets. Venue, travel, accommodation, travel, etc cost a lot. What is the return on this investment?

I am currently (Sept 2013) doing a course in Coursera.org on Networked Life. It is interesting to see, that simple metrics such as network diameter and cluster coefficient can be developed to represent a network of people and how easily information can flow through them. What is relevant to study is the impact these metrics have on productivity and efficiency of participants. I find that when a large number of participants get to meet others they have not known before and understand them better, the diameter of the network reduces. At the same time, too large a number of acquaintances may also be shunned by people because it costs them in terms of distraction to maintain the social relationship. The cluster coefficient kicks in here. People generally like to form a cluster of closer members whose company is helpful, and keep off a larger segment that may cause a distraction. So smaller diameters are good, but not so small that the clusters vanish and too large a component emerges. Jointly these two metrics may form an ideal pair that generates the optimum network effect. Optimum networks help spread desirable content virally within the network to those most interested in it, while avoiding distractions.

How many mails are we able to handle? How many topics are we able to handle? On line social media provides opportunities to form clusters. One can join them, evaluate the discussions going on, and either participate and raise the quality, or quit and reduce the distraction. When these clusters mature and come together in a conference, their effectiveness and efficiency may go up due to the additional intimacy in understanding they may bring. Are we able to organize a conference by invitation, to ensure a high productivity? For inviting people, can we find who will benefit better by attending which conference, through a network analysis? Alternately in larger conferences, we must actively facilitate the formation of small clusters of likeminded people and enable them to have a small group meet. The key question is if we enable these, how much will the productivity of the clusters go up? What diameter and cluster coefficient should individuals target. How can conference organizers facilitate this?

I am not sure if this is the right forum to post such a suggestion. I am from Wikimedia India Chapter, and would like to learn from as well as contribute to measurement related debates. TS-Sowmyan (talk) 03:59, 25 September 2013 (UTC)